EP2553628A1 - Automated monitoring and control of contamination in a production area - Google Patents
Automated monitoring and control of contamination in a production areaInfo
- Publication number
- EP2553628A1 EP2553628A1 EP11717800A EP11717800A EP2553628A1 EP 2553628 A1 EP2553628 A1 EP 2553628A1 EP 11717800 A EP11717800 A EP 11717800A EP 11717800 A EP11717800 A EP 11717800A EP 2553628 A1 EP2553628 A1 EP 2553628A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- individual
- production area
- image
- contamination control
- image data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
- G08B21/24—Reminder alarms, e.g. anti-loss alarms
- G08B21/245—Reminder of hygiene compliance policies, e.g. of washing hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
- G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
- G16H40/20—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Definitions
- the invention is directed to automated monitoring and control of contamination in a production area, particularly automated monitoring and control of contamination of the production area by an individual working in the production area.
- Contamination of a production area by workers can occur if, for example, a worker handles food product without wearing sterile gloves, or without wearing a hair net.
- Contamination of a production area can easily occur in a wide variety of industries, where hygiene is needed to prevent contamination by biological contaminants and other contaminants.
- industries include the food industry, the pharmaceutical industry, hospitals, doctors' offices, outpatient clinics, and other health-services and health-product related industries, in which microbial and other contamination can have adverse consequences on the consumer of the services and/or products.
- CCE contamination control equipment
- a first aspect of the invention is directed to an automated process for monitoring and controlling contamination in a production area.
- the process comprises capturing image data from the production area, processing the image data, and activating a contamination control device if the article of contamination control equipment is not present and properly positioned on the individual while the individual is working in the production area.
- the image data is processed to determine: (i) whether an individual is present within the production area in which the individual is to be wearing an article of contamination control equipment; and (ii) whether the article of contamination control equipment is present and properly positioned on the individual while the individual is working in the production area.
- the contamination control equipment comprises at least one member selected from the group consisting of a glove, a face mask, a suit, a gown, and a hair net.
- the image data is captured by scanning at least a portion of the production area with a camera.
- activating of the contamination control device comprises activating at least one member selected from group consisting of: (i) a means for contamination control, (ii) an alarm to notify the individual that the at least one article of contamination control equipment is not present or is not properly positioned, (iii) the generation of a report that the article of contamination control equipment was not present while the individual was present in the production area, or was not properly positioned while the individual was present in the production area.
- the means for contamination control comprises at least one member selected from the group consisting of: (i) cutting off power to at least one machine in the production area, and (ii) interjecting a physical restraint or barrier between the individual and the machine in the production area.
- activating the contamination control device comprises setting off the alarm, and the alarm comprises at least one member selected from the group consisting of an audible alarm, a visual alarm, and a vibratory alarm.
- the transmission of the report can comprise at least one member selected from the group consisting of transmission of an electronic report and transmission of a hard copy report.
- the image data can be captured over a time period, with the processing of the image data being carried out to find an image of at least a portion of an individual in motion, using a stabilization algorithm to determine whether the image data satisfies a threshold image value for a threshold time period, with the threshold image value being a pre-determined minimum image value correlating with an absence of the contamination control equipment properly positioned on the individual, and the threshold time period being a pre-determined minimum time period that the threshold image value is satisfied, with the
- the activating of the contamination control device comprises activating at least one member selected from group consisting of: (i) a means for contamination control; (ii) an alarm to notify the individual that the at least one article of contamination control equipment is not present or is not properly positioned; and (iii) the generation of a report that the article of contamination control equipment was not present while the individual was present in the production area, or was not properly positioned while the individual was present in the production area.
- the report includes an image of the individual in the work zone while the threshold image value is satisfied for the threshold time period, and a notation of a time at which the image was captured.
- a second aspect is directed to an automated system for monitoring and controlling contamination in a production area.
- the system comprises a computer, an imaging sensor in communication with the computer, the imaging sensor being configured and arranged to capture image data of at least a portion of the production area, and a computer-readable program code disposed on the computer.
- the computer-readable program code comprises: (i) a first executable portion for processing image data and creating an image of the production area, (ii) a second executable portion for processing image data to find an image of an individual or a portion of an individual in the production area, (iii) a third executable portion for processing image data and determining whether an article of contamination control equipment is present in association with the image of the individual or the image of the portion of the individual, (iv) a fourth executable portion for processing image data and determining if the article of contamination control equipment is properly positioned on the individual while the individual is in the production area, (v) a sixth executable portion for activating a contamination control device if the article of contamination control equipment is not present and properly positioned on the individual while the individual is present in the production area.
- the second aspect can utilize any feature the various embodiments of the first aspect can utilize, and vice versa.
- At least one member selected from the production area, the individual, and the article of CCE has secondary or supplemental source of data, such as an identifiable symbol, text marking, coloration, RFID tag, etc.
- the automated system further comprises a seventh executable portion comprising a stabilization algorithm to determine whether the image data satisfies a threshold image value for a threshold time period, with the threshold image value being a pre-determined minimum image value correlating an absence of the contamination control equipment properly positioned on the individual, and the threshold time period being a pre-determined minimum time period that the threshold image value is satisfied.
- the imaging sensor is a first imaging sensor and the system further comprises a second imaging sensor in communication with the computer, with the computer-readable program code disposed on the computer being provided with executable first, second, third, and fourth executable portions for creating and processing image data of at least a portion of the production area from the second imaging sensor, with the creating and processing of the image data from the second imaging sensor being carried out in a manner corresponding with the executable portions for capturing and processing image data from the first imaging sensor.
- the imaging sensor is a scanning imaging sensor configured and arranged to scan a production area.
- the automated system further comprises a data entry device that is in communication with the computer,
- a secondary image data capturing/processing system can be used to obtain and process data from a selected area of the field of view monitored by a primary image data capturing/processing system.
- the primary image data capturing/processing system which is utilized to identify personnel, CCE, and activate one or more CC devices, can also be used to direct the secondary image data capturing/processing system.
- the secondary image data capturing/processing system can include hyperspectral imaging systems, thermal imaging systems, radio frequency detection devices, microwave detection devices, colorimetric detection devices, gas chromatography, as well as electromechanical focusing equipment.
- the data processing of the primary image data capturing/processing system can be designed to activate the secondary image data capturing/processing system upon the detection of a condition that the secondary image data
- capturing/processing system has the capability to further assess in a desired manner.
- the data processing of the primary image data capturing/processing system can be designed to activate the secondary image data capturing/processing system upon the detection of a condition that the secondary image data capturing/processing system has the capability to further assess in a desired manner.
- a primary image data capturing/processing system can be used to monitor a work area at a sandwich shop, find an individual working behind a counter, and then subsequently define the arms and hands of a person that is making a sandwich.
- the primary image data capturing/processing system may determine whether the individual is wearing gloves and may then subsequently activate a secondary image data capturing/processing system utilizing a hyperspectral imaging camera (e.g., a HySpexTM hyperspectral camera such as HySpexTM model VNIR-640s hyperspectral camera available from Norsk Elektro Optikk AS), to observe just the defined hand area and determine if the hands are contaminated with bacteria such as ecoli; further enabling the activation of an alarm system if the bacteria is found.
- a hyperspectral imaging camera e.g., a HySpexTM hyperspectral camera such as HySpexTM model VNIR-640s hyperspectral camera available from Norsk Elektro Optikk AS
- This parallel process with selective focusing using multiple cameras
- the automated system further comprises a printer that is in communication with the computer and is capable of printing a report of a determination of whether contamination control equipment is properly positioned on the individual in the production area.
- FIG. 1 is a schematic diagram illustrating an automated machine vision process and system for monitoring and controlling contamination in a production area through the monitoring and control of the wearing of one or more articles of CCE by one or more individuals in a production area.
- FIG. 2 is a representative schematic of loop process for determining whether one or more persons in a production area are properly wearing CCE.
- FIG. 3 s a representative schematic of a process for tracking images of individuals, or particular portions of individuals, in a production environment.
- FIG. 4 is an illustration of the tracking of a plurality of faces in a given image from the production area.
- FIG. 5 is a representative schematic of the overall process for determining whether a tracked face is wearing an article of PPE.
- FIG. 6 is a schematic diagram illustrating an automated machine vision process and system for monitoring and controlling contamination in a production area through the monitoring and control of the wearing of one or more gloves by one or more individuals in the production area.
- FIG, 7 is a schematic illustrating a manner of assessing an arm blob to determine whether CCE is present.
- FIG. 8 is a schematic illustrating the process of judging whether CCE is on or off of an arm.
- FIG. 9 is a schematic illustrating an automated machine vision process and system for monitoring and controlling contamination in the preparation of a sandwich.
- FIG. 10 is a schematic illustrating a broom having warning lights that can be activated upon monitoring whether sanitary gloves are being contaminated while the broom is in use.
- FIG. 1 1 illustrates a rack holding sausage, with warning lights associated with each hanging sausage.
- FIG. 12 illustrates a cart holding various items intended to remain uncontaminated, the cart having warning lights thereon.
- the phrase "automated process” is used with reference to processes utilizing computer vision and/or machine vision in obtaining and processing image data.
- the image data is captured using one or more imaging sensors in communication with a computer. While the process can be carried out using only image data, additional data can be input from machine-readable or human-readable sensors and identifiers, radio frequency identification transponder (RFID) or other transmitting sensors, time stamps or biometric identification, object recognition, texture definition, database management and other software, data interface equipment consisting of serial, parallel, or network communication, binary data such as switches, gates, push buttons, current sensors, as well as additional forms of data input.
- RFID radio frequency identification transponder
- One or more computers can process image data and optionally other data from other sensors, identifiers, etc., using algorithms designed to determine whether the computer is to activate a control device, particularly a contamination control device (hereinafter "CC device").
- CC device contamination control device
- imaging sensor refers to a component of a vision system that captures image data, e.g., a camera or other image capturing device.
- one or more imaging sensors are configured and arranged to capture image data of a one or more objects within the production area.
- Imaging sensors include analog video cameras, digital video cameras, color and monochrome cameras, closed-circuit television (CCTV) cameras, charge-coupled device (CCD) sensors, complementary metal oxide semiconductor
- CMOS complementary metal-oxide-semiconductor
- PC cameras PC cameras
- pan-tilt-zoom cameras (PTZ) web cameras
- infra-red imaging devices any other devices that can capture image data.
- the selection of the particular camera type for a particular facility may be based on factors including environmental lighting conditions, the frame rate and data acquisition rate, and the ability to process data from the lens of the camera within the electronic circuitry of the camera control board, the size of the camera and associated electronics, the ease with which the camera can be mounted as well as powered, the lens attributes which are required based on the physical layout of the facility and the relative position of the camera to the objects, and the cost of the camera.
- Exemplary cameras that may be used in the practice of the invention are available from Sony such as Sony Handycam Camcorder model number DCR-SR80.
- Image data is captured and processed to determine the presence of one or more individuals, as well as in the presence of one or more articles of CCE.
- Image data can be processed in a manner to determine whether an article of CCE is being properly worn by an individual.
- the computer can be programmed to send a signal that automatically activates a CC device.
- CCPA controlled contamination production area
- the computer system i.e., one or more computers, can be programmed to process the image data to identify individuals as well as other objects in motion, and separate the moving objects from the non-moving background images.
- the computer system can be programmed to distinguish images of individuals from images of other moving objects.
- the computer system can be programmed to process image data for individuals required to be wearing CCE, and determine whether an individual is properly wearing a required article of CCE in a production area.
- Computer-readable program codes include program modules, algorithms, rules, and combinations thereof.
- the computer system may include computer- readable program codes that process the image data of one or more objects being monitored, in order to perform one or more of the following functions: identifying an object being monitored; tracking an object as it moves within the production area; locating an object in the production area; and associating information with an object.
- the computer system may process image data utilizing program modules, algorithms, rules, and combinations thereof.
- Computer vision may utilize one or more of the following: camera, computer, object recognition and tracking using blob analysis, texture definition, data base management and other software, data interface equipment consisting of serial, parallel, or network communication, specific activity based, founding data originating from the person or CCE (containing information on the individual or the CCE), and integration of other discrete characterization data such as RFID tags, binary data such as switches, gates, push buttons, or current sensors.
- the computer vision system may utilize an algorithm model or vision- based software to correctly identify a person from the environment. This may involve the use of multiple cameras and the geometric correlation of the perspective of a plurality of cameras having overlapping views or views from different perspectives. Algorithms such as the background subtraction method, Canny imaging, Harris corner imaging, Shen-Castan edge detection, grey level segmentation, skeletonization, etc., can be used to process image data in a manner that identifies the visual features of a person, e.g., eyes, ears, nose, head, arms, hands, and other body parts. See also J.R. Parker, "Algorithms for Image Processing and Computer Vision, John Wiley & Sons, (1997), and D.A. Forsyth and J. Ponce, "Computer Vision a Modern Approach", Prentiss Hall (January 2003), both of which is hereby incorporated in their entireties, by reference thereto.
- algorithm model or vision- based software to correctly identify a person from the environment. This may involve
- the safety equipment is further identified and associated to the person and the
- the interface summary and detection data may be printed to a report, burned to an electronic chip, or compact disc or other storage device or stored in a computer database and referenced by a unique identifier including name, CCE type or location.
- Image data can be processed using video content analysis (VCA) techniques.
- VCA video content analysis
- VCA techniques see, for example, Nathanael Rota and Monique Thonnat, "Video Sequence Interpretation for Visual Surveillance,” in Proc. of the 3d IEEE Int'l Workshop on Visual Surveillance, 59-67, Dublin, Ireland (Jul. 1 , 2000), and Jonathan Owens and Andrew Hunter, "Application in the Self-Organizing Map to Trajectory Classification," in Proc. Of the 3d IEEE Int'l Workshop on Visual Surveillance, 77-83, Dublin, Ireland (Jul. 1, 2000), both of which are hereby incorporated by reference.
- the VCA techniques are employed to recognize various features in the images obtained by the image capture devices.
- the computer system may use one or more Item Recognition Modules (IRM) to process image data for the recognition of a particular individual or other object in motion, and/or an article of CCE.
- IRM Item Recognition Modules
- LRM Location Recognition Module
- MRM Movement Recognition Modules
- the computer may use IRM in combination with LRM and/or MRM in identifying and tracking movements of particular individual or other object in motion, or article of CCE for the purpose of assessing velocity of movement and/or conformational movement characteristics, as well as in assessing whether contamination control requirements are being violated.
- the IRM, LRM, and MRM can be configured to operate independently or in conjunction with one another,
- the image data can be analyzed using human classification techniques that can be employed for the purpose of confirming whether an object is a human, as well as for analyzing the facial features. Face detection may be performed in accordance with the teachings described in, for example, any one or more of the following, each of which is incorporated, in its entirety, by reference thereto: International Patent WO 9932959, entitled “Method and System for Gesture Based Option Selection", and Damian Lyons and Daniel Pelletier, "A line-Scan Computer vision Algorithm for Identifying Human Body Features," Gesture '99, 85-96 France (1999); M.H. Yang and N. Ahuj , "Detecting Human Faces in Color Images", Proc. Int 'l Conf.
- production area refers to any area in which an automated system is used in a process of monitoring and controlling sanitation as individuals and/or machines work in an environment to make any form of measurable progress. While a typical production area would be a factory in which articles of manufacture are being produced, the phrase "production area” includes restaurants, gas stations, construction sites, offices, hospitals, etc., i.e., anywhere a product is being produced and/or a service is being rendered. The criteria for controlling contamination of a production area depend upon the particular nature of the production area, i.e., what articles are being produced and/or services offered, and the contamination control requirements associated with those products and/or services. With regard to minimizing the amount of contamination in any specified area, the area could also be referred to as a "sanitation area".
- work zone refers to a discrete area that can correspond with an entire production area, one or more discrete regions of a production area, or even an entire production area plus an additional area. Different regions within a production area can have different contamination control
- a first work zone could include only a defined area immediately surrounding a particular machine in a factory.
- the contamination control requirements for the machine operator and others within a specified distance of the machine may be greater than the contamination control requirements just a few meters away from the machine.
- a factory can have many different work zones within a single production area, such as 2-100 work zones, 2-50 work zones, or 2-10 work zones.
- a factory can have uniform CCE requirements throughout the production area, which can be one single work zone.
- CCE contamination control equipment
- articles of CCE include face masks, gloves, gowns, suits, aprons, hair nets, etc.
- the phrase "contamination control device” includes any device that, when activated, is designed to prevent, reduce the likelihood of, or reduce the degree of, the release of contamination from the individual into the production area.
- the CC device can be designed to immediately prevent the release of contamination and/or reduce the likelihood of the release of contamination, and/or reduce the degree of contamination released by the individual.
- the activation of the CC device could discontinue power to a machine, or interject a physical barrier or restraint between an individual and product that could be contaminated.
- the CC device could provide a more delayed effect on prevention or reduction of contamination.
- the CC device could be in the form of an alarm to alert one or more individuals of the heightened risk of contamination associated with the absence of a required article of CCE on an individual within the production area. The individuals could be left to decide how to address the condition in response to the alarm.
- the CC device could generate and transmit a report to a production manager, agent, safety officer, etc., for the purpose of modifying behavior so that the absence of the required article of CCE would be less likely to occur in the future.
- movement includes movements of objects in which the location of the center of gravity of the individual or object changes, as well as movements in which the center of gravity does not change, but the conformation of the individual or object changes. Changes in the location of the center of gravity of an individual or object in an ascertainable time period correlate with the velocity of the individual or object. "Conformational movements” are movements in which there is a substantial change in the location of the individual or object, but only a small (or no) change in the location of the center of gravity of the individual or object.
- the automated process for monitoring and controlling contamination in a production area utilizes algorithm-based computer vision to: (i) identify an individual or a portion of an individual; (ii) identify whether a required article of CCE is present in association with the individual or the portion of the individual, and/or determine whether the individual or portion of the individual has the required article of CCE properly positioned thereon; (iii) send a signal to automatically activate a CC device in the event that the required article of CCE is not present in association with the individual or the portion of the individual, and/or that the required article of CCE is not properly positioned on the individual or portion of the individual.
- FIG. 1 is a schematic diagram illustrating an automated machine vision process and system 10 for monitoring and controlling contamination in a production area through the monitoring and control of the wearing of one or more articles of CCE by one or more individuals in a production area.
- Computer vision system 18 for monitoring and controlling contamination in production area 12 captures and processes data related to one or more individuals wearing CCE,
- Production area 12 has multiple work zones 14 therein.
- image data capturing devices 16 e.g., cameras
- the one or more image data capturing devices 16 could be within production area 12 but not within any of work zones 14, or some or all image data capturing devices 16 could be within one or more of work zones 14.
- Image data capturing devices 16 provide image data input to one or more computer vision system 18 with data tracking and identifying personnel or body parts thereof including location in production area 12, as well as whether an individual is within one of work zones 14.
- other CCE-related data can be provided to computer vision system(s) 18 via other data input means such as symbolic alpha, or numeric information embodied in or on a machine or machine- readable or human-readable identifier such as a tag or label (e.g., bar coded tag or label), a hole pattern, a radio frequency identification transponder (RFID) or other transmitting sensors, machine readable sensors, time stamps or biometric
- RFID radio frequency identification transponder
- the resulting automated process system 10 provides data that is compared to predetermined fault criteria programmed into the one or more fault-detection analysis computer 19.
- the fault criteria are met if an individual is present in the production area 12 and/or one or more of work zones 14 without wearing one or more articles of CCE required for the respective production area 12 or work zone 14, or without having the one or more required articles of CCE properly positioned while the individual is in the respective production area 12 or work zone 14.
- Contamination control device 22 takes one or more actions selected from the group consisting of (i) activating a contamination control means, (ii) activating an alarm, and (iii) activating the generation and transmission of a report of a violation of contamination control protocol.
- the machine vision system can be designed to view the scene and detect the face of an individual and perform segmentation based on proportionality to find the eyes.
- the machine vision system can be designed to find features associated with the face mask (including color mismatch, etc) and can be designed to remove non- moving objects, and zoom and/or read information on associated objects or persons and activate electromechanical circuit(s).
- the machine vision system can be designed to view the scene and perform background subtraction and detect the face of an individual, and perform segmentation based on proportionality to find the arms of the individual, and perform segmentation based on proportionality to find the hands of the individual.
- the machine vision system can be designed to find features associated with gloves, including color mismatch, etc.
- the machine vision system can be designed to find features associated with one or more gloves (including color mismatch, etc) and can be designed to remove non- moving objects and zoom and/or read information on associated objects or individuals, and activate electromechanical circuit(s).
- the machine vision system can be designed to view the scene and perform background subtraction and detect the face of an individual, and perform
- the machine vision system can be designed to find confirmation features associated with the face mask (including color mismatch, etc) and can be designed to remove non-moving objects and zoom and/or read information on associated objects or individuals, and activate electromechanical circuit(s).
- the machine vision system can be designed to view the scene and perform background subtraction and detect the face of an individual, and perform segmentation based on proportionality to find the head of the individual.
- the machine vision system can be designed to find confirmation features associated with the presence or absence of a hair net, and can be designed to remove non-moving objects and zoom and/or read information on associated objects or individuals, and activate electromechanical circuit(s).
- the machine vision system can be designed to view the scene and perform background subtraction and detect the body of an individual, and perform
- the machine vision system can be designed to analyze proportionality ratios to confirm the presence or absence of the gown (including color mismatch, etc) and can be designed to remove non-moving objects and zoom and/or read information on associated objects or individuals, and activate electromechanical circuit(s).
- FIG. 2 illustrates a representative schematic of loop process for determining whether one or more persons in a production area are properly wearing CCE to be placed on the face.
- the process of FIG. 2 includes: (i) primary data processing module 40 for finding a moving face within a production area, (ii) secondary data processing module 42 for determining the presence or absence of CCE such as a face mask on the associated face, as well as whether the CCE is properly positioned on the face, and (iii) tertiary data processing module 44 which utilizes a stabilization algorithm that tracks the face within the production area to ensure consistent data reporting.
- Stabilization algorithm 44 completes a data processing feedback loop to prevent "false positives" from occurring.
- the stabilization algorithm of tertiary data processing module 44 requires a combination of (a) assessment of a pre-determined quality of image (i.e., a minimum image value) associated with the face in the absence of properly positioned CCE, and that this quality of image be present for at least a pre-determined minimum time period, before the system reports a CCE non-compliance event.
- a pre-determined quality of image i.e., a minimum image value
- the images can be processed so that an image having a very high image quality correlating with non-compliance can be saved as a record of the non-compliance event.
- it can have the date, hour, and location provided therewith, together with other data such as the duration of the period of non-compliance, etc.
- the stabilization algorithm can be carried out as follows. First, obtain an image value (i.e., detection result) of the subject matter being monitored which relates to whether, for example, CCE is present and properly positioned, for each frame in which a person or other subject matter is being monitored. Image value is determined for each frame, regardless of whether particular image features are detected or not. The image value can be related to "CCE present and properly positioned" or "CCE not present or not properly positioned", depending on the project or algorithm. For example, for glove detection, both glove detection and glove non-detection have features. The image features can include (i) skin blob is long and narrow, and (ii) hand is detected, both of which are related to "glove not detected”.
- the features of "glove detected” include (i) non-skin pixels around the hand area, and (ii) edges around wrist.
- Adjusted image value adjusted image value from previous frame + 1 ,
- Adjusted image value adjusted image value from previous frame - 1 ,
- an adjusted image value is above a pre-selected maximum value, it is trimmed to the maximum value. If an adjusted image value is below a pre-selected minimum value, it is trimmed to the minimum value.
- each adjusted image value trimmed to the maximum value is compared against a preset maximum threshold image value and deemed to correspond with CCE not present or not properly positioned; and each adjusted image value trimmed to the minimum value is compared against a preset minimum threshold image value and deemed to correspond with CCE present and properly positioned. All image values that are not adjusted or that are adjusted and do not exceed the preselected maximum threshold image value or fall below the pre-selected minimum threshold image value are deemed neutral and are not used to determine whether CCE is or is not present and properly positioned.
- a low pass filter can be used as a stabilization algorithm.
- a low pass filter stabilization algorithm can be carried out by obtaining a set of image values (i.e., detection results) over a period of time (e.g., to, ti, etc), in which successive frames each provide a particular image value (V) of the subject matter being monitored which relates to whether, for example, CCE is present and properly positioned, or not present or not properly positioned.
- V image value
- V t0 + (V t i - V t0 ) a, in which V (0 represents the image value of the frame acquired at time to, and V t i represents the image value of the frame at time t
- each adjusted image value which is greater than a selected upper threshold value is deemed to correspond with CCE not present or not properly positioned; and each adjusted image value which is less than a selected lower threshold image value is deemed to correspond with CCE present and properly positioned. All adjusted image values that do not exceed the selected upper threshold value or do not fall below the selected lower threshold value are deemed neutral and are not used to determine whether CCE is present and properly positioned, or is not present or not properly positioned.
- the first step in the process of monitoring and controlling contamination in a production area associated with the use of CCE is to find the image of a face in motion in a production area. This can be carried out by using Haar-Iike feature detection. Alternatively, the number of skin pixels within a face region can be counted in assessing that a particular image is that of a face. In a third method, an image is determined to be something other than a face if dividing the number of skin pixels by the number of pixels in the face region produces a result less than a threshold value, otherwise it is a face.
- / is object image
- B is background image.
- the image can be judged as non-moving if Dif is less than a pre -determined threshold.
- the background image can be assessed using low pass filtering over time, in which:
- r is a predetermined time constant
- B is a low pass filtered background image
- / is an image
- FIG. 3 illustrates a second step in the process, i.e., the step of tracking individual faces in the production area.
- computation is made of the location of each face of the current image (46) and the locations of the features of the known faces in the previous image (48), i.e., distances are computed between each of the faces of the current image and the faces known from the image immediately preceding in time. Determinations are made as to which faces are closest to one another (50) between the faces in current image (46) and the faces in the immediately prior image (48).
- the speed of imaging is likely high enough (e.g., 200 milliseconds between images) that the likelihood is greatest that closest faces in the respective current and prior images in fact represent the same face.
- Locations and feature properties are then updated for the new image (52), and the new locations properties are stored (54).
- the old image of the production area including the old faces (48), can then be removed from the stack (58) (i.e., group) of closest faces in the current image
- a "reminder" is provided to ensure removal of the non-essential prior images of the faces.
- Feature distance D can be determined as:
- o y j , ⁇ 2 , ⁇ ⁇ 3 are pre-determined variances obtained from samples of the same object in continuous (i.e., successive) frames.
- Properties can then be updated by characterization of the image life, i.e., by measurement of how long the image has been successfully tracked, by
- ⁇ is a predetermined time constant.
- FIG. 4 is an illustration of the tracking of a plurality of faces in a given image from the production area.
- Image 60 is taken at TV In image 60, Face A, Face B, Face C, and Face D appear at particular locations.
- Image 62 is taken at time T 2 , a fraction of a second after TV Image 62 shows tracked Face A, tracked Face B, tracked Face C, and tracked Face D at particular locations of image 62. While tracked Face A and tracked Face B are in approximately the same locations at T 2 as at T tracked Faces B and C appear in different positions at T 2 , showing their relative movement between T[ and T 2 .
- D include their "life” (i.e., how long they have been present in the image, including how long they have been present at or near their current location), the image value of the low pass Filter PPE on/off value, their location (i.e., position), size, and color histogram.
- the update of the properties can be assessed by the increment life value, the decrement life, and the low pass filter on/off value, as described above.
- FIG. 5 is a representative schematic of the overall process for determining whether a tracked face is wearing an article of CCE. This is the portion of the process and system that are designed to provide a data feedback loop to prevent "false positives" from occurring.
- the feedback loop of the stabilization algorithm is set up to determine, with a high degree of accuracy, whether the face actually is wearing a required article of CCE in a manner conforming to contamination protocol within the production area. Without the use of the stabilization algorithm, a multitude of false positives have been found to occur when using image capturing and processing of faces in motion in a production area.
- each tracked face is assessed using a low pass filter (64), assessing whether the image value corresponds with the face properly wearing the required article of CCE, or not properly wearing the required article of CCE, A pre- determined image value threshold is used in processing the image of the tracked face. If the image of the tracked face is such that the assessed image value is less than the threshold image value, the image is assessed as either being unstable or that the required article of CCE is being properly worn by the face (66). In such an instance, no safety control device is activated (66).
- the processing is continued by assessing whether the time period over which the image value threshold is met is a time period that meets or exceeds a pre-determined threshold time period (68). If the image value threshold has not been met for the duration of the threshold time period, the result is that time no CC device is activated (66). However, if the threshold image value is satisfied for the threshold time period, a signal is sent that the face-associated CCE is "off and that tracking is stable (70), with the result that a CC device is activated (70).
- markers on the CCE can be provided to assess the presence or absence of properly positioned CCE on the face.
- the markers can have particular color and intensity patterns located at pre-determined positions, relative to the face, making it easier to determine whether the required CCE is properly worn on the face.
- the measure of the marker existence can be xj. For example, if marker is a blue marker, X] can equal the difference between the target number of pixels and the number of blue pixels.
- high intensity points can be assessed, as the number of high intensity points represents the reflection of face-associated equipment.
- X2 can equal the number of pixels having an intensity greater than a pre-determined threshold intensity value.
- a horizontal edge under the eyes can also be assessed, as the existence of an edge, and the strength of the edge located pre-determined position under the eyes and relative to the face, corresponds with the presence of properly worn CCE (e.g., facemask) on the face. This can be assessed as follows:
- x 3 I/1-/2I where and h being on the same horizontal axis but on different vertical axes.
- N of predetermined pixel sample represents skin: (s ,S2,sj, ,s N ). Pixel distance (di, f3 ⁇ 4,
- Thresholding can be carried out using a pre-determined value th. If the distance is smaller than th, the pixel is skin, otherwise, the pixel is not skin.
- face-associated CCE is judged as "OFF".
- the second method for determining whether the face associated CCE is "ON” or "OFF” utilizes Bayesian classifier:
- Face-associated CCE is judged as "ON” if:
- the algorithm consists of several modules, specifically: (a) a primary module that finds a moving object from a background within a work environment; (b) a secondary algorithm that finds an arm blob from the primary object; (c) a judgment algorithm that determines whether the gloves are located on the arm blob; and (d) a optional stabilization algorithm using tracking and time life may to ensure accurate reporting.
- FIG. 6 is a schematic of an automated process for detecting gloves on hands (or other hand associated CCE) using a computer algorithm further coupled to hardware.
- the algorithm is carried out by obtaining an image and detecting skin color (82) in the image, followed by detecting motion (84) of the image, followed by a labeling method (86), followed by finding an arm blob (88), followed by judging whether a glove is "on” or "off the arm (90).
- Skin color detection can be assessed in accordance with the methodology disclosed above for the tracking of faces.
- Motion can be detected using a motion subtraction method. Motion exists if:
- Motion detector devices can also be used.
- the labeling method can be carried out by obtaining a blob from a binary image.
- the arm blob can be found by finding the closest blob to a pre-determined object.
- FIG. 7 illustrates arm blob 100 with a dotted line illustrating ellipse 102 that approximates the size of arm blob 100.
- Ellipse 102 has length "f ", width 'V, and an aspect ratio of C: w or C/w
- FIG. 8 is a schematic illustrating the process of judging whether the glove is on or off the arm.
- Features are extracted from each blob (92), following which the data is processed to determine whether the smallest feature distance from a model is less than a threshold value th (94), If the smallest feature distance is not less than the threshold value, the glove is determined to be "on” (96). If the smallest feature distance is determined to be less than the threshold value, the glove is determined to be "off (98).
- the extraction of features from each blob is carried out by determining the long radius of the fitted ellipse, determining the short radius of fitted ellipse, determining the distance from a model contour by (a) finding the closest point in the object contour
- M is matrix often used as inverse covariance.
- the glove is determined to be "OFF”. Otherwise, the glove is determined to be "ON”.
- Sequence breaks are found by supposing to, t ⁇ , , . . . are instances when a motion is detected. If (trak + i- ⁇ > threshold, then there is a sequence break between /ford + i and / admir. Otherwise, trak + ⁇ and / admir are in the same sequence. The results are grouped by sequence. Focusing on each sequence, count the number of glove OFF images
- NOFF - > threshold then output warning with image.
- Find a warning image in the sequence i.e., an image used in activating a CC device, by identifying the most representative hand-like blob image, and by considering the images at the midpoint of the continuous OFF images.
- a cutting board is located at a sandwich making station.
- a sandwich-maker is located at the cutting board and is monitored by a video camera such as a Trendnet ® TV IP1 10 internet camera server network camera.
- the camera sends a visual data wirelessly via a router (e.g., NETGEAR ® - RangeMax 802.1 lg Wireless Router, model WPN824, available from Best Buy, P.O. Box 9312,
- the computer processes the data in a near real time manner to determine if the sandwich- maker is complying with proper contamination-prevention protocol such as wearing gloves and a cap.
- the output signal from the computer controls light emitting diodes embedded within the cutting board.
- the cutting board made with food-grade polyethylene, may have light emitting diodes embedded in a corner, overlaid with a translucent printing identifying a hat, gloves, or other contamination-related identifiers. Additional signal-receiving circuitry may be embedded in the cutting board so that a signal from the transmitter at the control computer can be received, further illuminating the board in the proper location to help warn the sandwich maker of any safety requirement being violated.
- FIG. 9 is a schematic diagram illustrating an automated machine vision process and system for monitoring and controlling contamination in the sandwich- making example above.
- Camera 104 captures image data 106 that is processed by data processing computer 108. If data processing computer 108 detects a protocol violation, it sends that transmitter 110 then sends to cutting board 112.
- warning circuitry and light emitting diodes including colored lights to further help in training.
- a green, yellow, and red light may indicate the number of violations identified in a period of time.
- Warning lights may be positioned independently in locations easily viewed by workers. In some cases, the lights may be accompanied with acoustics including warning messages. A sign may have back-lit lettering such as "Did you wash your hands?" or "Sanitize your shoes! These signs may be activated by the computer vision system as described above when a fault is detected.
- signs may be located at a blender to identify if a batch of food has been contaminated by a lost glove and a relay may be activated to shut the blender down.
- a relay may be activated to shut the blender down.
- the occurrence of such an event, left undetected, can contaminate a large number of food products.
- a machine vision safety system can be used to identify the error when it occurs, as well as identifying the contaminated products prior to distribution.
- the sandwich-maker can be identified when he touches a sandwich or edible product, and the product may be tracked through to its wrapping or packaging state.
- the package Prior to delivery to the customer, the package may be sent through a printing, labeling, or laser marking station (e.g., LP-V 10 Series Laser Markers F AYb manufactured by Panasonic) and the wrapping or package marked or labeled with words or symbols to convey "contaminated food" or "do not use".
- the sandwich or food may be placed on a conveyor for transit and an electromechanical or pneumatic system may divert the contaminated product to a disposal station or a cleaning station.
- uncontaminated food may be placed in a "safe zone" or labeled "handled safely”.
- Hyperspectral imaging devices may be utilized as primary or secondary data capturing devices, in combination with appropriate data processing of the data therefrom.
- FIG. 10 illustrates a broom 114 having warning lights 1 16 that can be activated upon monitoring whether gloves are being worn while the broom is in use.
- warning lights 116 positioned on the handle of broom 1 14 notify the user of this unsafe practice, as well as to dispose of gloves contaminated through the unauthorized contact.
- the power to a cash register may be terminated via a relay (e.g., High-Amp & Medium-Amp Relays, available from McMaster-Carr Supply Company, P.O. Box 54960, Los Angeles, CA 90054-0960), or, in the case of a setting in which a machine (e.g., an automatic power slicer for slicing deli meat) is to be used in combination with sanitary gloves, the power to the machine may be shut off if the individual is not wearing gloves while loading and unloading food from the machine.
- a relay e.g., High-Amp & Medium-Amp Relays, available from McMaster-Carr Supply Company, P.O. Box 54960, Los Angeles, CA 90054-0960
- the power to the machine may be shut off if the individual is not wearing gloves while loading and unloading food from the machine.
- FIG. 1 1 illustrates a rack 118 holding sausage 120.
- each hook light emitting diode 122 is affixed, with each hook being connected to a sensor (not illustrated, but, e.g., Compact Digital Hanging Scale Legal-for-Trade 61b/2.7 Kg capacity, item number 3952T41 , from McMaster-Carr Supply Company, P.O. Box 54960, Los Angeles, CA 90054-0960) that detects when a sausage is added or loaded to a cart.
- a sensor not illustrated, but, e.g., Compact Digital Hanging Scale Legal-for-Trade 61b/2.7 Kg capacity, item number 3952T41 , from McMaster-Carr Supply Company, P.O. Box 54960, Los Angeles, CA 90054-0960
- a sensor not illustrated, but, e.g., Compact Digital Hanging Scale Legal-for-Trade 61b/2.7 Kg capacity, item number 3952T41 , from McMaster-Carr Supply Company, P.O. Box
- a warning light may be installed within a mechanical fixture used to hold food which may be activated to identify that the food hanging, contained, attached, or associated to the fixture has been handled by an operator that was not practicing safe food handling protocol such as wearing gloves.
- the contamination monitoring and control system can be designed to be capable of detecting whether a worker has coughed on food or sanitary items while the food is on a cart, as illustrated in FIG. 12, in which cart 124, holding various items intended to remain
- Warning lights 126 are to be activated in the event of contamination of the items on cart 124, such as by unsanitary handling without gloves, sneezing, coughing, etc.
- a system could, for example, help to prevent to prevent the spread of germs to patients in a health care facility such as a hospital, outpatient center, doctor's office, or retirement home.
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Epidemiology (AREA)
- Public Health (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
In an automated process for monitoring and controlling contamination in a production area, captured image data is processed to determine whether an individual is to wearing an article of contamination control equipment (e.g., gloves, face mask, etc) and whether the equipment is properly positioned on the individual. A determination that the contamination control equipment is not present or not properly positioned automatically activates a contamination control device such as contamination control means (e.g., shutting off power to machine) or an alarm or generates and sends a report of the contamination control protocol violation. An automated system for monitoring and controlling contamination includes a computer, an imaging sensor in communication with the computer, and a computer-readable program code disposed on the computer.
Description
AUTOMATED MONITORING AND CONTROL OF CONTAMINATION IN
A PRODUCTION AREA
This application claims the benefit of, and incorporates by reference the entirety of Provisional Application No. 61/341 ,655 filed April 1 , 2010.
Background
[0001] The invention is directed to automated monitoring and control of contamination in a production area, particularly automated monitoring and control of contamination of the production area by an individual working in the production area.
[0002] Contamination of a production area by workers can occur if, for example, a worker handles food product without wearing sterile gloves, or without wearing a hair net. Contamination of a production area, particularly contamination of product being produced in a production area, can easily occur in a wide variety of industries, where hygiene is needed to prevent contamination by biological contaminants and other contaminants. Such industries include the food industry, the pharmaceutical industry, hospitals, doctors' offices, outpatient clinics, and other health-services and health-product related industries, in which microbial and other contamination can have adverse consequences on the consumer of the services and/or products.
[0003] Yet other industries require contamination-free, high purity production environments because contamination can negatively impact the function and/or aesthetics of the resulting product. One such industry is the manufacture of semiconductor devices, microelectronics, microchips, and integrated circuits.
Another industry in which contamination negatively impacts product quality is the coating (e.g., painting) of various products, including automotive, furniture, construction, and other products, in which contaminants adversely affect the appearance, feel, or durability of the coating applied to the product. The presence of skin oils, airborne dust, and other contaminants from workers can result in product defects during one or more phases of the manufacturing process.
[0004] Although there are numerous vendors supplying the market with contamination control equipment (hereinafter, "CCE") such as gloves, face masks, hair nets, aprons, gowns, etc., and although employers require employees to wear
CCE, the cost of contamination in the workplace remains high. Manual monitoring of employees, vendors, and visitors through close circuit camera or direct supervision is both expensive and subjective. The overall reporting of violations can be inaccurate and unverifiable.
[0005] There is a need for a system that accurately monitors individuals as they work in a production area, to ensure that they wear required articles of CCE and to ensure that the required contamination-control protocol is followed to avoid contamination of products and/or injury to the ultimate consumer of the product . Summary
[0006] A first aspect of the invention is directed to an automated process for monitoring and controlling contamination in a production area. The process comprises capturing image data from the production area, processing the image data, and activating a contamination control device if the article of contamination control equipment is not present and properly positioned on the individual while the individual is working in the production area. The image data is processed to determine: (i) whether an individual is present within the production area in which the individual is to be wearing an article of contamination control equipment; and (ii) whether the article of contamination control equipment is present and properly positioned on the individual while the individual is working in the production area.
[0007] In an embodiment, the contamination control equipment comprises at least one member selected from the group consisting of a glove, a face mask, a suit, a gown, and a hair net.
[0008] In an embodiment, the image data is captured by scanning at least a portion of the production area with a camera.
[0009] In an embodiment, activating of the contamination control device comprises activating at least one member selected from group consisting of: (i) a means for contamination control, (ii) an alarm to notify the individual that the at least one article of contamination control equipment is not present or is not properly positioned, (iii) the generation of a report that the article of contamination control equipment was not present while the individual was present in the production area, or was not properly positioned while the individual was present in the production area.
[0010] In an embodiment, the means for contamination control comprises at least one member selected from the group consisting of: (i) cutting off power to at least one machine in the production area, and (ii) interjecting a physical restraint or barrier between the individual and the machine in the production area.
[0011] In an embodiment, activating the contamination control device comprises setting off the alarm, and the alarm comprises at least one member selected from the group consisting of an audible alarm, a visual alarm, and a vibratory alarm. The transmission of the report can comprise at least one member selected from the group consisting of transmission of an electronic report and transmission of a hard copy report.
[0012) In an embodiment, the image data can be captured over a time period, with the processing of the image data being carried out to find an image of at least a portion of an individual in motion, using a stabilization algorithm to determine whether the image data satisfies a threshold image value for a threshold time period, with the threshold image value being a pre-determined minimum image value correlating with an absence of the contamination control equipment properly positioned on the individual, and the threshold time period being a pre-determined minimum time period that the threshold image value is satisfied, with the
contamination control device being activated if the threshold image value is satisfied for the threshold time period. In an embodiment, the activating of the contamination control device comprises activating at least one member selected from group consisting of: (i) a means for contamination control; (ii) an alarm to notify the individual that the at least one article of contamination control equipment is not present or is not properly positioned; and (iii) the generation of a report that the article of contamination control equipment was not present while the individual was present in the production area, or was not properly positioned while the individual was present in the production area. In an embodiment, the report includes an image of the individual in the work zone while the threshold image value is satisfied for the threshold time period, and a notation of a time at which the image was captured.
[0013] A second aspect is directed to an automated system for monitoring and controlling contamination in a production area. The system comprises a computer, an imaging sensor in communication with the computer, the imaging sensor being
configured and arranged to capture image data of at least a portion of the production area, and a computer-readable program code disposed on the computer. The computer-readable program code comprises: (i) a first executable portion for processing image data and creating an image of the production area, (ii) a second executable portion for processing image data to find an image of an individual or a portion of an individual in the production area, (iii) a third executable portion for processing image data and determining whether an article of contamination control equipment is present in association with the image of the individual or the image of the portion of the individual, (iv) a fourth executable portion for processing image data and determining if the article of contamination control equipment is properly positioned on the individual while the individual is in the production area, (v) a sixth executable portion for activating a contamination control device if the article of contamination control equipment is not present and properly positioned on the individual while the individual is present in the production area. The second aspect can utilize any feature the various embodiments of the first aspect can utilize, and vice versa.
[0014] In an embodiment, at least one member selected from the production area, the individual, and the article of CCE has secondary or supplemental source of data, such as an identifiable symbol, text marking, coloration, RFID tag, etc.
[0015] In an embodiment, the automated system further comprises a seventh executable portion comprising a stabilization algorithm to determine whether the image data satisfies a threshold image value for a threshold time period, with the threshold image value being a pre-determined minimum image value correlating an absence of the contamination control equipment properly positioned on the individual, and the threshold time period being a pre-determined minimum time period that the threshold image value is satisfied.
[0016] In an embodiment, the imaging sensor is a first imaging sensor and the system further comprises a second imaging sensor in communication with the computer, with the computer-readable program code disposed on the computer being provided with executable first, second, third, and fourth executable portions for creating and processing image data of at least a portion of the production area from the second imaging sensor, with the creating and processing of the image data from
the second imaging sensor being carried out in a manner corresponding with the executable portions for capturing and processing image data from the first imaging sensor.
[0017] In an embodiment, the imaging sensor is a scanning imaging sensor configured and arranged to scan a production area.
[0018] In an embodiment, the automated system further comprises a data entry device that is in communication with the computer,
[0019] In an embodiment, a secondary image data capturing/processing system can be used to obtain and process data from a selected area of the field of view monitored by a primary image data capturing/processing system. The primary image data capturing/processing system, which is utilized to identify personnel, CCE, and activate one or more CC devices, can also be used to direct the secondary image data capturing/processing system. The secondary image data capturing/processing system can include hyperspectral imaging systems, thermal imaging systems, radio frequency detection devices, microwave detection devices, colorimetric detection devices, gas chromatography, as well as electromechanical focusing equipment.
[0020] The data processing of the primary image data capturing/processing system can be designed to activate the secondary image data capturing/processing system upon the detection of a condition that the secondary image data
capturing/processing system has the capability to further assess in a desired manner.
The data processing of the primary image data capturing/processing system can be designed to activate the secondary image data capturing/processing system upon the detection of a condition that the secondary image data capturing/processing system has the capability to further assess in a desired manner.
[00211 For example, a primary image data capturing/processing system can be used to monitor a work area at a sandwich shop, find an individual working behind a counter, and then subsequently define the arms and hands of a person that is making a sandwich. The primary image data capturing/processing system may determine whether the individual is wearing gloves and may then subsequently activate a secondary image data capturing/processing system utilizing a hyperspectral imaging camera (e.g., a HySpex™ hyperspectral camera such as HySpex™ model VNIR-640s hyperspectral camera available from Norsk Elektro Optikk AS), to observe just the
defined hand area and determine if the hands are contaminated with bacteria such as ecoli; further enabling the activation of an alarm system if the bacteria is found. This parallel process with selective focusing using multiple cameras increases the speed and efficiency with which data can be analyzed.
[0022] In an embodiment, the automated system further comprises a printer that is in communication with the computer and is capable of printing a report of a determination of whether contamination control equipment is properly positioned on the individual in the production area. Brief Description of the Drawings
[0023] FIG. 1 is a schematic diagram illustrating an automated machine vision process and system for monitoring and controlling contamination in a production area through the monitoring and control of the wearing of one or more articles of CCE by one or more individuals in a production area.
[0024] FIG. 2 is a representative schematic of loop process for determining whether one or more persons in a production area are properly wearing CCE.
[0025] FIG. 3 s a representative schematic of a process for tracking images of individuals, or particular portions of individuals, in a production environment.
[0026] FIG. 4 is an illustration of the tracking of a plurality of faces in a given image from the production area.
[0027] FIG. 5 is a representative schematic of the overall process for determining whether a tracked face is wearing an article of PPE.
[0028] FIG. 6 is a schematic diagram illustrating an automated machine vision process and system for monitoring and controlling contamination in a production area through the monitoring and control of the wearing of one or more gloves by one or more individuals in the production area.
[0029] FIG, 7 is a schematic illustrating a manner of assessing an arm blob to determine whether CCE is present.
[0030] FIG. 8 is a schematic illustrating the process of judging whether CCE is on or off of an arm.
[0031] FIG. 9 is a schematic illustrating an automated machine vision process and system for monitoring and controlling contamination in the preparation of a sandwich.
[0032] FIG. 10 is a schematic illustrating a broom having warning lights that can be activated upon monitoring whether sanitary gloves are being contaminated while the broom is in use.
[0033] FIG. 1 1 illustrates a rack holding sausage, with warning lights associated with each hanging sausage.
[0034] FIG. 12 illustrates a cart holding various items intended to remain uncontaminated, the cart having warning lights thereon.
Detailed Description
[0035] As used herein, the phrase "automated process" is used with reference to processes utilizing computer vision and/or machine vision in obtaining and processing image data. The image data is captured using one or more imaging sensors in communication with a computer. While the process can be carried out using only image data, additional data can be input from machine-readable or human-readable sensors and identifiers, radio frequency identification transponder (RFID) or other transmitting sensors, time stamps or biometric identification, object recognition, texture definition, database management and other software, data interface equipment consisting of serial, parallel, or network communication, binary data such as switches, gates, push buttons, current sensors, as well as additional forms of data input. One or more computers can process image data and optionally other data from other sensors, identifiers, etc., using algorithms designed to determine whether the computer is to activate a control device, particularly a contamination control device (hereinafter "CC device").
[0036] As used herein, the phrase "imaging sensor" refers to a component of a vision system that captures image data, e.g., a camera or other image capturing device. In computer vision and machine vision systems, one or more imaging sensors are configured and arranged to capture image data of a one or more objects within the production area. Imaging sensors include analog video cameras, digital video cameras, color and monochrome cameras, closed-circuit television (CCTV) cameras, charge-coupled device (CCD) sensors, complementary metal oxide semiconductor
(CMOS) sensors, analog and digital cameras, PC cameras, pan-tilt-zoom cameras (PTZ), web cameras, infra-red imaging devices, and any other devices that can
capture image data. The selection of the particular camera type for a particular facility may be based on factors including environmental lighting conditions, the frame rate and data acquisition rate, and the ability to process data from the lens of the camera within the electronic circuitry of the camera control board, the size of the camera and associated electronics, the ease with which the camera can be mounted as well as powered, the lens attributes which are required based on the physical layout of the facility and the relative position of the camera to the objects, and the cost of the camera. Exemplary cameras that may be used in the practice of the invention are available from Sony such as Sony Handycam Camcorder model number DCR-SR80.
[0037] Image data is captured and processed to determine the presence of one or more individuals, as well as in the presence of one or more articles of CCE. Image data can be processed in a manner to determine whether an article of CCE is being properly worn by an individual. Upon determination that a required article of CCE is not in place, the computer can be programmed to send a signal that automatically activates a CC device.
[0038] Since motion takes place over a period of time, processing of image data to determine individuals in motion requires capturing image data over a period of time. Differences in location and/or conformation of objects as a function of time allows the computer to process image data in a manner that distinguishes moving objects from non-moving background. Image data can also be processed using one or more threshold values to determine whether there is a violation of one or more
predetermined standards for a controlled contamination production area (hereinafter "CCPA"), with activation of a CC device in the event that the predetermined standard is being violated.
[0039] The computer system, i.e., one or more computers, can be programmed to process the image data to identify individuals as well as other objects in motion, and separate the moving objects from the non-moving background images. The computer system can be programmed to distinguish images of individuals from images of other moving objects. The computer system can be programmed to process image data for individuals required to be wearing CCE, and determine whether an individual is properly wearing a required article of CCE in a production area.
[0040] Computer-readable program codes include program modules, algorithms, rules, and combinations thereof. The computer system may include computer- readable program codes that process the image data of one or more objects being monitored, in order to perform one or more of the following functions: identifying an object being monitored; tracking an object as it moves within the production area; locating an object in the production area; and associating information with an object. The computer system may process image data utilizing program modules, algorithms, rules, and combinations thereof.
[0041] Computer vision may utilize one or more of the following: camera, computer, object recognition and tracking using blob analysis, texture definition, data base management and other software, data interface equipment consisting of serial, parallel, or network communication, specific activity based, founding data originating from the person or CCE (containing information on the individual or the CCE), and integration of other discrete characterization data such as RFID tags, binary data such as switches, gates, push buttons, or current sensors.
[0042] The computer vision system may utilize an algorithm model or vision- based software to correctly identify a person from the environment. This may involve the use of multiple cameras and the geometric correlation of the perspective of a plurality of cameras having overlapping views or views from different perspectives. Algorithms such as the background subtraction method, Canny imaging, Harris corner imaging, Shen-Castan edge detection, grey level segmentation, skeletonization, etc., can be used to process image data in a manner that identifies the visual features of a person, e.g., eyes, ears, nose, head, arms, hands, and other body parts. See also J.R. Parker, "Algorithms for Image Processing and Computer Vision, John Wiley & Sons, (1997), and D.A. Forsyth and J. Ponce, "Computer Vision a Modern Approach", Prentiss Hall (January 2003), both of which is hereby incorporated in their entireties, by reference thereto.
[0043] Using the same types of vision algorithms applied for tracking people, the safety equipment is further identified and associated to the person and the
environment in which the CCE is required. Monitoring of both the initially-tracked individual and his immediate association with one or more articles of CCE can be done simultaneously. The coupling of data from auxiliary equipment from markers
such as RFID tags, physical interface monitors, and electronic controls (such as inline current sensing units) to the CCE and the person provides additional monitoring capability.
[0044] The software's recognition of actions may trigger parent-child
relationships to other pieces of equipment and the analysis of a continuous stream of data from the cameras may initiate additional correlations of the individual as he moves through a monitored area. The interface summary and detection data may be printed to a report, burned to an electronic chip, or compact disc or other storage device or stored in a computer database and referenced by a unique identifier including name, CCE type or location.
[0045] Image data can be processed using video content analysis (VCA) techniques. For a detailed discussion of suitable VCA techniques, see, for example, Nathanael Rota and Monique Thonnat, "Video Sequence Interpretation for Visual Surveillance," in Proc. of the 3d IEEE Int'l Workshop on Visual Surveillance, 59-67, Dublin, Ireland (Jul. 1 , 2000), and Jonathan Owens and Andrew Hunter, "Application in the Self-Organizing Map to Trajectory Classification," in Proc. Of the 3d IEEE Int'l Workshop on Visual Surveillance, 77-83, Dublin, Ireland (Jul. 1, 2000), both of which are hereby incorporated by reference. Generally, the VCA techniques are employed to recognize various features in the images obtained by the image capture devices.
[0046] The computer system may use one or more Item Recognition Modules (IRM) to process image data for the recognition of a particular individual or other object in motion, and/or an article of CCE. In addition, the computer system may use one or more Location Recognition Module (LRM) to determine the location of a particular individual or other object in motion, or an article of CCE. In addition, the computer system may use one or more Movement Recognition Modules (MRM) to process movement data for the recognition of a particular individual or other object in motion, or article of CCE. The computer may use IRM in combination with LRM and/or MRM in identifying and tracking movements of particular individual or other object in motion, or article of CCE for the purpose of assessing velocity of movement and/or conformational movement characteristics, as well as in assessing whether
contamination control requirements are being violated. The IRM, LRM, and MRM can be configured to operate independently or in conjunction with one another,
[0047] The image data can be analyzed using human classification techniques that can be employed for the purpose of confirming whether an object is a human, as well as for analyzing the facial features. Face detection may be performed in accordance with the teachings described in, for example, any one or more of the following, each of which is incorporated, in its entirety, by reference thereto: International Patent WO 9932959, entitled "Method and System for Gesture Based Option Selection", and Damian Lyons and Daniel Pelletier, "A line-Scan Computer vision Algorithm for Identifying Human Body Features," Gesture '99, 85-96 France (1999); M.H. Yang and N. Ahuj , "Detecting Human Faces in Color Images", Proc. Int 'l Conf. IEEE Image Processing, pp. 127-139, Oct. 1998; I. Haritaoglu, D. Harwood, L. Davis, "Hydra: Multiple People Detection and Tracking Using Silhouettes," Computer Vision and Pattern Recognition, Second Workshop of Vidoe Surveillance (CVPR, 1999); A. Cohnenarez and T.S. Huang, "Maximum Likelihood Face Detection",
International Conference On Face and Gesture Recognition, pp 164-169, ilington, Vt. (Oct. 14-16, 1 96); Owens, J. and Hunter, A., "Application of the Self- Organising Map to Trajectory Classification", Proc. 3rd IEEE International Workshop on Visual Surveillance, IEEE Comput. Soc, Los Alamitos, CA, USA, pages 77-83 (2000); N. Rota and M. Thonnat, "Video Sequence Interpretation For Video
Surveillance, Proceedings of the Third IEEE International Workshop on Visual SurveillanceftOOO); Srinivas Gutta, Jeffrey Huang, Ibrahim F. Imam, Harry
Wechsler, "Face and Hand Gesture Recognition Using Hybrid Classifiers",
Proceedings of the International Conference on Automatic Face and Gesture
Recognition, ICAFGR 96, 164- 169, illington ( 1996); and A. Criminisi, A.
Zisserman, L. Van Gool, Bramble S., and D. Compton, "A New Approach To Obtain Height Measurements from Video", Proc. ofSPIE, Boston, Massachussets, USA, volume 3576, pp. 227-238 (1 -6 November 1998).
[0048] As used herein, the phrase "production area" refers to any area in which an automated system is used in a process of monitoring and controlling sanitation as individuals and/or machines work in an environment to make any form of measurable progress. While a typical production area would be a factory in which articles of
manufacture are being produced, the phrase "production area" includes restaurants, gas stations, construction sites, offices, hospitals, etc., i.e., anywhere a product is being produced and/or a service is being rendered. The criteria for controlling contamination of a production area depend upon the particular nature of the production area, i.e., what articles are being produced and/or services offered, and the contamination control requirements associated with those products and/or services. With regard to minimizing the amount of contamination in any specified area, the area could also be referred to as a "sanitation area".
[0049] As used herein, the phrase "work zone" refers to a discrete area that can correspond with an entire production area, one or more discrete regions of a production area, or even an entire production area plus an additional area. Different regions within a production area can have different contamination control
requirements. For example, a first work zone could include only a defined area immediately surrounding a particular machine in a factory. The contamination control requirements for the machine operator and others within a specified distance of the machine may be greater than the contamination control requirements just a few meters away from the machine. A factory can have many different work zones within a single production area, such as 2-100 work zones, 2-50 work zones, or 2-10 work zones. Alternatively, a factory can have uniform CCE requirements throughout the production area, which can be one single work zone.
[0050] As used herein, the phrase "contamination control equipment" (i.e., "CCE") refers to any article to be worn by an individual for the purpose of controlling the emission of contamination from the individual into the production environment. As such, articles of CCE include face masks, gloves, gowns, suits, aprons, hair nets, etc.
[0051] As used herein, the phrase "contamination control device" (i.e., "CC device") includes any device that, when activated, is designed to prevent, reduce the likelihood of, or reduce the degree of, the release of contamination from the individual into the production area. The CC device can be designed to immediately prevent the release of contamination and/or reduce the likelihood of the release of contamination, and/or reduce the degree of contamination released by the individual.
[0052] For example, the activation of the CC device could discontinue power to a machine, or interject a physical barrier or restraint between an individual and product that could be contaminated. Alternatively, the CC device could provide a more delayed effect on prevention or reduction of contamination. For example, the CC device could be in the form of an alarm to alert one or more individuals of the heightened risk of contamination associated with the absence of a required article of CCE on an individual within the production area. The individuals could be left to decide how to address the condition in response to the alarm. Alternatively, the CC device could generate and transmit a report to a production manager, agent, safety officer, etc., for the purpose of modifying behavior so that the absence of the required article of CCE would be less likely to occur in the future.
[0053] As used herein, the term "movement" includes movements of objects in which the location of the center of gravity of the individual or object changes, as well as movements in which the center of gravity does not change, but the conformation of the individual or object changes. Changes in the location of the center of gravity of an individual or object in an ascertainable time period correlate with the velocity of the individual or object. "Conformational movements" are movements in which there is a substantial change in the location of the individual or object, but only a small (or no) change in the location of the center of gravity of the individual or object.
[0054] The automated process for monitoring and controlling contamination in a production area utilizes algorithm-based computer vision to: (i) identify an individual or a portion of an individual; (ii) identify whether a required article of CCE is present in association with the individual or the portion of the individual, and/or determine whether the individual or portion of the individual has the required article of CCE properly positioned thereon; (iii) send a signal to automatically activate a CC device in the event that the required article of CCE is not present in association with the individual or the portion of the individual, and/or that the required article of CCE is not properly positioned on the individual or portion of the individual.
[0055] One or more embodiments of the present invention now will be described with reference to the accompanying drawings, in which some, but not all
embodiments of the invention are shown. The invention may be embodied in many different forms and should not be construed as limited to the embodiments set forth
herein. Rather, these embodiments are provided so that this disclosure will satisfy applicable legal requirements. Like numbers refer to like elements throughout.
[0056] FIG. 1 is a schematic diagram illustrating an automated machine vision process and system 10 for monitoring and controlling contamination in a production area through the monitoring and control of the wearing of one or more articles of CCE by one or more individuals in a production area. Computer vision system 18 for monitoring and controlling contamination in production area 12 captures and processes data related to one or more individuals wearing CCE, Production area 12 has multiple work zones 14 therein. Although image data capturing devices 16 (e.g., cameras) are shown outside of production area 12, they could be within production area 12. The one or more image data capturing devices 16 could be within production area 12 but not within any of work zones 14, or some or all image data capturing devices 16 could be within one or more of work zones 14. Image data capturing devices 16 provide image data input to one or more computer vision system 18 with data tracking and identifying personnel or body parts thereof including location in production area 12, as well as whether an individual is within one of work zones 14. In addition to data provided by image data capturing devices 16, other CCE-related data can be provided to computer vision system(s) 18 via other data input means such as symbolic alpha, or numeric information embodied in or on a machine or machine- readable or human-readable identifier such as a tag or label (e.g., bar coded tag or label), a hole pattern, a radio frequency identification transponder (RFID) or other transmitting sensors, machine readable sensors, time stamps or biometric
identification, CCE markers or designs or coloration, etc., as illustrated by other input data 20 from production area 12.
[0057] The resulting automated process system 10 provides data that is compared to predetermined fault criteria programmed into the one or more fault-detection analysis computer 19. The fault criteria are met if an individual is present in the production area 12 and/or one or more of work zones 14 without wearing one or more articles of CCE required for the respective production area 12 or work zone 14, or without having the one or more required articles of CCE properly positioned while the individual is in the respective production area 12 or work zone 14. If computer vision system 18 in combination with fault-detection computer 19 determine that one or
more individuals are not wearing the required article(s) of CCE in respective production area 12 or work zone 14, and/or if the automated process determines that required article(s) of CCE are not properly positioned on the one or more individuals in the production area 12 or work zone 14, data input from computer vision system 18 to fault-detection computer 19 results in the assessment of the existence of a fault, causing fault-detection computer 19 to trigger contamination control device 22. Contamination control device 22 takes one or more actions selected from the group consisting of (i) activating a contamination control means, (ii) activating an alarm, and (iii) activating the generation and transmission of a report of a violation of contamination control protocol.
[0058] If the automated process is directed to the presence and proper use of a face mask, the machine vision system can be designed to view the scene and detect the face of an individual and perform segmentation based on proportionality to find the eyes. The machine vision system can be designed to find features associated with the face mask (including color mismatch, etc) and can be designed to remove non- moving objects, and zoom and/or read information on associated objects or persons and activate electromechanical circuit(s).
[0059] If the automated process is directed to the presence and proper use of one or more gloves (or any other form of safety equipment for one or both hands), the machine vision system can be designed to view the scene and perform background subtraction and detect the face of an individual, and perform segmentation based on proportionality to find the arms of the individual, and perform segmentation based on proportionality to find the hands of the individual. The machine vision system can be designed to find features associated with gloves, including color mismatch, etc. The machine vision system can be designed to find features associated with one or more gloves (including color mismatch, etc) and can be designed to remove non- moving objects and zoom and/or read information on associated objects or individuals, and activate electromechanical circuit(s).
[0060] If the automated process is directed to the presence and proper use of a face mask, the machine vision system can be designed to view the scene and perform background subtraction and detect the face of an individual, and perform
segmentation based on proportionality to find the mouth and nose of the individual.
The machine vision system can be designed to find confirmation features associated with the face mask (including color mismatch, etc) and can be designed to remove non-moving objects and zoom and/or read information on associated objects or individuals, and activate electromechanical circuit(s).
[0061] If the automated process is directed to the presence and proper positioning of a hair net on an individual, the machine vision system can be designed to view the scene and perform background subtraction and detect the face of an individual, and perform segmentation based on proportionality to find the head of the individual. The machine vision system can be designed to find confirmation features associated with the presence or absence of a hair net, and can be designed to remove non-moving objects and zoom and/or read information on associated objects or individuals, and activate electromechanical circuit(s).
[0062] If the automated process is directed to the presence and proper use of gown (or a skirt or apron or any other form of safety equipment for the body of an individual), the machine vision system can be designed to view the scene and perform background subtraction and detect the body of an individual, and perform
segmentation based on proportionality to find the hips, shoulders, and feet of the individual. The machine vision system can be designed to analyze proportionality ratios to confirm the presence or absence of the gown (including color mismatch, etc) and can be designed to remove non-moving objects and zoom and/or read information on associated objects or individuals, and activate electromechanical circuit(s).
[0063] FIG. 2 illustrates a representative schematic of loop process for determining whether one or more persons in a production area are properly wearing CCE to be placed on the face. The process of FIG. 2 includes: (i) primary data processing module 40 for finding a moving face within a production area, (ii) secondary data processing module 42 for determining the presence or absence of CCE such as a face mask on the associated face, as well as whether the CCE is properly positioned on the face, and (iii) tertiary data processing module 44 which utilizes a stabilization algorithm that tracks the face within the production area to ensure consistent data reporting.
[0064] Stabilization algorithm 44 completes a data processing feedback loop to prevent "false positives" from occurring. In the absence of stabilization algorithm 44, it can be difficult to set up the image capturing device and associated primary data processing module 40 and second processing module 42 so that together they consistently maintain an accurate determination of the presence or absence of properly positioned CCE on an individual in motion in the production area. The motion of the face, the motion of other objects in the production area, and various other factors have been determined to make it difficult to consistently make accurate determinations of the presence and placement of CCE on a moving face in the production area. As a result, inaccurate conclusions of non-compliance (i.e., "false positives") have been found to occur at a high rate, particularly when image data is being captured at a rate of, for example, 50 images per second. Single occurrences of images which show the presence of a face but which are inaccurately assessed by the data processing to be in the absence of CCE, can soar to thousands per hour. The stabilization algorithm of tertiary data processing module 44 requires a combination of (a) assessment of a pre-determined quality of image (i.e., a minimum image value) associated with the face in the absence of properly positioned CCE, and that this quality of image be present for at least a pre-determined minimum time period, before the system reports a CCE non-compliance event. In this manner, the process can be carried out using a stabilization algorithm that reduces the activation of a CC device due to a false positive to, for example, less than 0.1 percent of all determinations of non-compliance determinations. In addition, the images can be processed so that an image having a very high image quality correlating with non-compliance can be saved as a record of the non-compliance event. Optionally, it can have the date, hour, and location provided therewith, together with other data such as the duration of the period of non-compliance, etc.
[0065] The stabilization algorithm can be carried out as follows. First, obtain an image value (i.e., detection result) of the subject matter being monitored which relates to whether, for example, CCE is present and properly positioned, for each frame in which a person or other subject matter is being monitored. Image value is determined for each frame, regardless of whether particular image features are detected or not. The image value can be related to "CCE present and properly positioned" or "CCE
not present or not properly positioned", depending on the project or algorithm. For example, for glove detection, both glove detection and glove non-detection have features. The image features can include (i) skin blob is long and narrow, and (ii) hand is detected, both of which are related to "glove not detected". If such features are not detected, the absence of these features is related to "gloves on." Algorithms can have features from both states, i.e., "glove on" (= glove detected) and "glove off (= glove not detected). The features of "glove detected" include (i) non-skin pixels around the hand area, and (ii) edges around wrist. Algorithms having features from both states have image values encompassing three values: "glove on" = 1 ; "neutral" = 0; "glove off ' = - 1.
[0066] Second, adjust the image value, which may be different for each image acquired, as follows:
Adjusted image value = adjusted image value from previous frame + 1 ,
for each image value detected which corresponds with
CCE not present or not properly positioned;
and
Adjusted image value = adjusted image value from previous frame - 1 ,
for each image value detected which corresponds with
CCE present and properly positioned.
If an adjusted image value is above a pre-selected maximum value, it is trimmed to the maximum value. If an adjusted image value is below a pre-selected minimum value, it is trimmed to the minimum value.
[0067] Third, each adjusted image value trimmed to the maximum value is compared against a preset maximum threshold image value and deemed to correspond with CCE not present or not properly positioned; and each adjusted image value trimmed to the minimum value is compared against a preset minimum threshold image value and deemed to correspond with CCE present and properly positioned. All image values that are not adjusted or that are adjusted and do not exceed the preselected maximum threshold image value or fall below the pre-selected minimum threshold image value are deemed neutral and are not used to determine whether CCE is or is not present and properly positioned.
[0068] Alternatively, a low pass filter can be used as a stabilization algorithm. In one embodiment, a low pass filter stabilization algorithm can be carried out by obtaining a set of image values (i.e., detection results) over a period of time (e.g., to, ti, etc), in which successive frames each provide a particular image value (V) of the subject matter being monitored which relates to whether, for example, CCE is present and properly positioned, or not present or not properly positioned.
[0069] Second, a low pass filtration is conducted on the image values over time, using the following formula:
Adjusted image value at t| = Vt0 + (Vti - Vt0) a, in which V(0 represents the image value of the frame acquired at time to, and Vti represents the image value of the frame at time t|, and a represents a sensitivity multiplier.
[0070] Third, each adjusted image value which is greater than a selected upper threshold value is deemed to correspond with CCE not present or not properly positioned; and each adjusted image value which is less than a selected lower threshold image value is deemed to correspond with CCE present and properly positioned. All adjusted image values that do not exceed the selected upper threshold value or do not fall below the selected lower threshold value are deemed neutral and are not used to determine whether CCE is present and properly positioned, or is not present or not properly positioned.
[0071] The first step in the process of monitoring and controlling contamination in a production area associated with the use of CCE is to find the image of a face in motion in a production area. This can be carried out by using Haar-Iike feature detection. Alternatively, the number of skin pixels within a face region can be counted in assessing that a particular image is that of a face. In a third method, an image is determined to be something other than a face if dividing the number of skin pixels by the number of pixels in the face region produces a result less than a threshold value, otherwise it is a face.
[0072] Finding facial images of one or more individuals in a production area can be reasonably limited to finding images of faces in motion in the production area. This can be performed by computing the difference between the image of the face and the background image, in which:
Dif =∑(within region)|/-f?|,
where / is object image, and B is background image. The image can be judged as non-moving if Dif is less than a pre -determined threshold. The background image can be assessed using low pass filtering over time, in which:
Β = τΒ + { \ - τ)Ι,
where r is a predetermined time constant, B is a low pass filtered background image, and / is an image.
[0073] FIG. 3 illustrates a second step in the process, i.e., the step of tracking individual faces in the production area. As shown in FIG. 3 computation is made of the location of each face of the current image (46) and the locations of the features of the known faces in the previous image (48), i.e., distances are computed between each of the faces of the current image and the faces known from the image immediately preceding in time. Determinations are made as to which faces are closest to one another (50) between the faces in current image (46) and the faces in the immediately prior image (48). The speed of imaging is likely high enough (e.g., 200 milliseconds between images) that the likelihood is greatest that closest faces in the respective current and prior images in fact represent the same face. Locations and feature properties are then updated for the new image (52), and the new locations properties are stored (54). The old image of the production area including the old faces (48), can then be removed from the stack (58) (i.e., group) of closest faces in the current image
(52), with faces of the new image then being stored together with the storage their new properties (54). A "reminder" is provided to ensure removal of the non-essential prior images of the faces.
[0074] The computation of feature distances can be carried out by evaluation of differences in facial position (y\), differences in face size (y2), and differences in color histogram differences ( 3). Feature distance D can be determined as:
where oyj , σ 2 , σγ3 are pre-determined variances obtained from samples of the same object in continuous (i.e., successive) frames.
[0075] Properties can then be updated by characterization of the image life, i.e., by measurement of how long the image has been successfully tracked, by
measurement of a low pass filtered determination of the CCE "on/off value" of the
face, and by characterization of the features of the face, including position, size, and color histogram. Properties can be updated by the Increment Life value if the tracked face is associated with the face found in the current frame, as well as by Decrement Life if no face is associated to this tracked face. An example of determination of the low pass filter "on/off value" of the PPE on the face is as follows:
LPF ^ xLPF + (l-x)status
here τ is a predetermined time constant.
[0076] FIG. 4 is an illustration of the tracking of a plurality of faces in a given image from the production area. Image 60 is taken at TV In image 60, Face A, Face B, Face C, and Face D appear at particular locations. Image 62 is taken at time T2, a fraction of a second after TV Image 62 shows tracked Face A, tracked Face B, tracked Face C, and tracked Face D at particular locations of image 62. While tracked Face A and tracked Face B are in approximately the same locations at T2 as at T tracked Faces B and C appear in different positions at T2, showing their relative movement between T[ and T2. As described above, the properties of each of Faces A-
D include their "life" (i.e., how long they have been present in the image, including how long they have been present at or near their current location), the image value of the low pass Filter PPE on/off value, their location (i.e., position), size, and color histogram. The update of the properties can be assessed by the increment life value, the decrement life, and the low pass filter on/off value, as described above.
[0077] FIG. 5 is a representative schematic of the overall process for determining whether a tracked face is wearing an article of CCE. This is the portion of the process and system that are designed to provide a data feedback loop to prevent "false positives" from occurring. In short, the feedback loop of the stabilization algorithm is set up to determine, with a high degree of accuracy, whether the face actually is wearing a required article of CCE in a manner conforming to contamination protocol within the production area. Without the use of the stabilization algorithm, a multitude of false positives have been found to occur when using image capturing and processing of faces in motion in a production area.
[0078] In FIG. 5, each tracked face is assessed using a low pass filter (64), assessing whether the image value corresponds with the face properly wearing the required article of CCE, or not properly wearing the required article of CCE, A pre-
determined image value threshold is used in processing the image of the tracked face. If the image of the tracked face is such that the assessed image value is less than the threshold image value, the image is assessed as either being unstable or that the required article of CCE is being properly worn by the face (66). In such an instance, no safety control device is activated (66).
[0079] However, if the image value threshold is met during the low pass filter processing of the image of the tracked face (64), the processing is continued by assessing whether the time period over which the image value threshold is met is a time period that meets or exceeds a pre-determined threshold time period (68). If the image value threshold has not been met for the duration of the threshold time period, the result is that time no CC device is activated (66). However, if the threshold image value is satisfied for the threshold time period, a signal is sent that the face-associated CCE is "off and that tracking is stable (70), with the result that a CC device is activated (70).
[0080] Various features of the tracked face can be assessed in order to determine the image value of the face. Markers on the CCE can be provided to assess the presence or absence of properly positioned CCE on the face. The markers can have particular color and intensity patterns located at pre-determined positions, relative to the face, making it easier to determine whether the required CCE is properly worn on the face. The measure of the marker existence can be xj. For example, if marker is a blue marker, X] can equal the difference between the target number of pixels and the number of blue pixels.
[0081] Similarly, high intensity points can be assessed, as the number of high intensity points represents the reflection of face-associated equipment. For example, X2 can equal the number of pixels having an intensity greater than a pre-determined threshold intensity value.
[0082] A horizontal edge under the eyes can also be assessed, as the existence of an edge, and the strength of the edge located pre-determined position under the eyes and relative to the face, corresponds with the presence of properly worn CCE (e.g., facemask) on the face. This can be assessed as follows:
x3 = I/1-/2I
where
and h being on the same horizontal axis but on different vertical axes.
[0083] Skin color can also be assessed as an indicator of whether CCE is properly positioned on the face, by determination of the ratio of pixels within skin color range in pre-determined range, relative to the face, e.g., where 4 = number of skin color pixels.
[0084] Skin color detection can be assessed as follows. First, for each pixel Pi=[R G B] and p2=[R G B], pixel distance d is defined as
where∑ is a matrix, in which inverse of covariance matrix is often used. N of predetermined pixel sample represents skin: (s ,S2,sj, ,sN). Pixel distance (di, f¾,
<¾,. .. ,df/) is computed from each pre-determined pixel (s/,S ,sj, ,siV). The minimum distance within N set of distances is found using: dmin = m n {di,d2,d3,
,dN} . Thresholding can be carried out using a pre-determined value th. If the distance is smaller than th, the pixel is skin, otherwise, the pixel is not skin.
[0085] Another method of skin color detection, which is faster, utilizes color vector analysis wherein p=[R G B], with pre-determined vectors ai,a2,a3,. - " p is skin pixel if
(ailp<; ii) n (a2 p< j¾2)
n...
[0086] in determining whether the face associated CCE is "ON" or "OFF", either of the following methods can be used. Using simple thresholding, assume features X , X2, X3, X4 and predetermined threshold
judge face-associated CCE as "ON" if :
(xi>/Ai) n (x2>/ ?2) o (x >tfa) n
Otherwise, face-associated CCE is judged as "OFF".
[0087] The second method for determining whether the face associated CCE is "ON" or "OFF" utilizes Bayesian classifier:
x = [x\ X2 X3 4]T
Face-associated CCE is judged as "ON" if:
/?ON(X)>/?OFF(X)
where /JO (X) an /¾FF(X) are probability functions predetermined by samples.
Normal distribution is assumed.
[0088] A detailed method for monitoring and controlling contamination in a production area will now be described for a system for monitoring and controlling the use of gloves on an individual. The algorithm consists of several modules, specifically: (a) a primary module that finds a moving object from a background within a work environment; (b) a secondary algorithm that finds an arm blob from the primary object; (c) a judgment algorithm that determines whether the gloves are located on the arm blob; and (d) a optional stabilization algorithm using tracking and time life may to ensure accurate reporting.
[0089] FIG. 6 is a schematic of an automated process for detecting gloves on hands (or other hand associated CCE) using a computer algorithm further coupled to hardware. The algorithm is carried out by obtaining an image and detecting skin color (82) in the image, followed by detecting motion (84) of the image, followed by a labeling method (86), followed by finding an arm blob (88), followed by judging whether a glove is "on" or "off the arm (90).
[0090] Skin color detection can be assessed in accordance with the methodology disclosed above for the tracking of faces.
[0091] Motion can be detected using a motion subtraction method. Motion exists if:
■"(region of interest){|/«(xj'Hit- ,y)| } > threshold
Most web cameras have this function. Motion detector devices can also be used.
[0092] The labeling method can be carried out by obtaining a blob from a binary image. The arm blob can be found by finding the closest blob to a pre-determined object.
[0093] Features include the size and the aspect ratio (i.e., ratio of horizontal to vertical size), with the blob being approximated as an ellipse, with the aspect ratio being the ratio of the long radius to the short radius. FIG. 7 illustrates arm blob 100 with a dotted line illustrating ellipse 102 that approximates the size of arm blob 100. Ellipse 102 has length "f ", width 'V, and an aspect ratio of C: w or C/w
[0094] FIG. 8 is a schematic illustrating the process of judging whether the glove is on or off the arm. Features are extracted from each blob (92), following which the data is processed to determine whether the smallest feature distance from a model is less than a threshold value th (94), If the smallest feature distance is not less than the
threshold value, the glove is determined to be "on" (96). If the smallest feature distance is determined to be less than the threshold value, the glove is determined to be "off (98).
[0095] The extraction of features from each blob is carried out by determining the long radius of the fitted ellipse, determining the short radius of fitted ellipse, determining the distance from a model contour by (a) finding the closest point in the object contour
from model contour, and (b) summing the distances.
[0096| The smallest feature distance (greatest similarity value) is determined by assessing the blob feature as:
X = ( \ , Xi , X2, X3, . . . )T,
assessing the model feature as:
and assessing the feature distance as:
d = {x-y) vM{x-y).
Since there can be more than one model, find minimum of d. M is matrix often used as inverse covariance.
[0097] Judging whether the smallest feature distance is less than the threshold value can be carried out as follows:
if (x i > thi) length of blob > threshold and
(x2 < th2) width of blob < threshold
(X3 > f 3) skin pixels > threshold,
then the glove is determined to be "OFF". Otherwise, the glove is determined to be "ON".
[0098] Tracking a hand blob and maintaining stable properties of the hand- associated CCE so that these properties can be used to make consistent determinations of whether the glove is "on" or "off, are carried out as follows:
[0099] Sequence breaks are found by supposing to, t\ , , . . . are instances when a motion is detected. If (t„+i- ^ > threshold, then there is a sequence break between /„+i and /„. Otherwise, t„+\ and /„ are in the same sequence. The results are grouped by sequence. Focusing on each sequence, count the number of glove OFF images
If NOFF -> threshold, then output warning with image. Find a warning image
in the sequence, i.e., an image used in activating a CC device, by identifying the most representative hand-like blob image, and by considering the images at the midpoint of the continuous OFF images.
Examples
As an example, a cutting board is located at a sandwich making station. A sandwich-maker is located at the cutting board and is monitored by a video camera such as a Trendnet® TV IP1 10 internet camera server network camera. The camera sends a visual data wirelessly via a router (e.g., NETGEAR® - RangeMax 802.1 lg Wireless Router, model WPN824, available from Best Buy, P.O. Box 9312,
Minneapolis , MN 55440) to a computer (e.g., eMachines - Netbook with Intel
Atom™ Processor, Model: EM250-1915, also available from Best Buy). The computer processes the data in a near real time manner to determine if the sandwich- maker is complying with proper contamination-prevention protocol such as wearing gloves and a cap. The output signal from the computer controls light emitting diodes embedded within the cutting board. The cutting board, made with food-grade polyethylene, may have light emitting diodes embedded in a corner, overlaid with a translucent printing identifying a hat, gloves, or other contamination-related identifiers. Additional signal-receiving circuitry may be embedded in the cutting board so that a signal from the transmitter at the control computer can be received, further illuminating the board in the proper location to help warn the sandwich maker of any safety requirement being violated.
FIG. 9 is a schematic diagram illustrating an automated machine vision process and system for monitoring and controlling contamination in the sandwich- making example above. Camera 104 captures image data 106 that is processed by data processing computer 108. If data processing computer 108 detects a protocol violation, it sends that transmitter 110 then sends to cutting board 112.
[00100] In a similar manner, other items such as hats, gloves, shoes, utensils, tools, furniture, conveyors can be fitted with warning circuitry and light emitting diodes including colored lights to further help in training. For example, a green, yellow, and red light may indicate the number of violations identified in a period of time.
Warning lights may be positioned independently in locations easily viewed by workers. In some cases, the lights may be accompanied with acoustics including
warning messages. A sign may have back-lit lettering such as "Did you wash your hands?" or "Sanitize your shoes!" These signs may be activated by the computer vision system as described above when a fault is detected.
[00101 ] Similarly, signs may be located at a blender to identify if a batch of food has been contaminated by a lost glove and a relay may be activated to shut the blender down. The occurrence of such an event, left undetected, can contaminate a large number of food products. A machine vision safety system can be used to identify the error when it occurs, as well as identifying the contaminated products prior to distribution.
[00102] For example, if a sandwich maker is not wearing gloves, or if the gloves are not properly positioned, the sandwich-maker can be identified when he touches a sandwich or edible product, and the product may be tracked through to its wrapping or packaging state. Prior to delivery to the customer, the package may be sent through a printing, labeling, or laser marking station (e.g., LP-V 10 Series Laser Markers F AYb manufactured by Panasonic) and the wrapping or package marked or labeled with words or symbols to convey "contaminated food" or "do not use". In some cases, the sandwich or food may be placed on a conveyor for transit and an electromechanical or pneumatic system may divert the contaminated product to a disposal station or a cleaning station. In contrast, uncontaminated food may be placed in a "safe zone" or labeled "handled safely". Hyperspectral imaging devices may be utilized as primary or secondary data capturing devices, in combination with appropriate data processing of the data therefrom.
[00103] The handling of non-food preparation items such as money, brooms, or door knobs can also be identified by creating monitored zones in which a worker should remove his gloves. FIG. 10 illustrates a broom 114 having warning lights 1 16 that can be activated upon monitoring whether gloves are being worn while the broom is in use. By tracking an individual as he enters a zone where he should not be wearing gloves, such as using broom 114, warning lights 116 positioned on the handle of broom 1 14 notify the user of this unsafe practice, as well as to dispose of gloves contaminated through the unauthorized contact. In addition, the power to a cash register may be terminated via a relay (e.g., High-Amp & Medium-Amp Relays, available from McMaster-Carr Supply Company, P.O. Box 54960, Los Angeles, CA
90054-0960), or, in the case of a setting in which a machine (e.g., an automatic power slicer for slicing deli meat) is to be used in combination with sanitary gloves, the power to the machine may be shut off if the individual is not wearing gloves while loading and unloading food from the machine.
[00104] FIG. 1 1 illustrates a rack 118 holding sausage 120. Above each hook light emitting diode 122 is affixed, with each hook being connected to a sensor (not illustrated, but, e.g., Compact Digital Hanging Scale Legal-for-Trade 61b/2.7 Kg capacity, item number 3952T41 , from McMaster-Carr Supply Company, P.O. Box 54960, Los Angeles, CA 90054-0960) that detects when a sausage is added or loaded to a cart. Consider a sausage maker who has forgotten to wear a hair net. When the sausage-maker, identified by the algorithms as not following the safety protocol, approaches the cart, a warning status alarm is transmitted to the cart and to the hooks that have changed weight when the worker touches the sausage. The red light above each of these hooks warns that the sausage is unfit for sale.
[00105] In some cases a warning light may be installed within a mechanical fixture used to hold food which may be activated to identify that the food hanging, contained, attached, or associated to the fixture has been handled by an operator that was not practicing safe food handling protocol such as wearing gloves. The contamination monitoring and control system can be designed to be capable of detecting whether a worker has coughed on food or sanitary items while the food is on a cart, as illustrated in FIG. 12, in which cart 124, holding various items intended to remain
uncontaminated, has warning lights 126 thereon. Warning lights 126 are to be activated in the event of contamination of the items on cart 124, such as by unsanitary handling without gloves, sneezing, coughing, etc. Such a system could, for example, help to prevent to prevent the spread of germs to patients in a health care facility such as a hospital, outpatient center, doctor's office, or retirement home.
Claims
WHAT IS CLAIMED IS:
Claim 1 : An automated process for monitoring and controlling contamination in a production area, comprising:
(A) capturing image data from the production area;
(B) processing the image data to determine:
(i) whether an individual is present within the production area in which the individual is to be wearing an article of contamination control equipment; and
(ii) whether the article of contamination control equipment is present and properly positioned on the individual while the individual is working in the production area; and
(C) activating a contamination control device if the article of
contamination control equipment is not present and properly positioned on the individual while the individual is working in the production area.
Claim 2: The automated process according to claim 1 , wherein the image data is captured over a time period, and the processing of the image data is carried out to find an image of at least a portion of an individual in motion, using a stabilization algorithm to determine whether the image data satisfies a threshold image value for a threshold time period, with the threshold image value being a pre-determined minimum image value correlating with an absence of the contamination control equipment properly positioned on the individual, and the threshold time period being a pre-determined minimum time period that the threshold image value is satisfied, with the contamination control device being activated if the threshold image value is satisfied for the threshold time period.
Claim 3 : The automated process according to claim 2, wherein the activating of the contamination control device comprises activating at least one member selected from group consisting of:
(i) a means for contamination control;
(ii) an alarm to notify the individual that the at least one article of
contamination control equipment is not present or is not properly positioned;
(iii) the generation of a report that the article of contamination control equipment was not present while the individual was present in the production area, or was not properly positioned while the individual was present in the production area.
Claim 4: The automated process according to claim 3, wherein the report includes an image of the individual in the work zone while the threshold image value is satisfied for the threshold time period, and a notation of a time at which the image was captured.
Claim 5 : The automated process according to claim 1 , wherein the contamination control equipment comprises at least one member selected from the group consisting of a glove, a face mask, a suit, a gown, and a hair net.
Claim 6: The automated process according to claim 1 , wherein the image data is captured by scanning at least a portion of the production area with a camera.
Claim 7: The automated process according to Claim 1 , wherein at least one member selected from the work zone, the individual, and the article of personal protective equipment has an RFID tag thereon.
Claim 8: The automated process according to claim 1 , wherein the activating of the contamination control device comprises activating at least one member selected from group consisting of:
(i) a means for contamination control;
(ii) an alarm to notify the individual that the at least one article of
contamination control equipment is not present or is not properly positioned;
(iii) the generation of a report that the article of contamination control equipment was not present while the individual was present in the production area, or was not properly positioned while the individual was present in the production area.
Claim 9: The automated process according to claim 8, wherein the means for contamination control comprises at least one member selected from the group consisting of: (i) cutting off power to at least one machine in the production area, and (ii) interjecting a physical restraint or barrier between the individual and the machine in the production area.
Claim 10: The automated process according to claim 8, wherein activating the contamination control device comprises setting off the alarm, and the alarm comprises at least one member selected from the group consisting of an audible alarm, a visual alarm, and a vibratoiy alarm.
Claim 1 1 : The automated process according to Claim 8, further comprising the transmission of the report, with the transmission of the report comprising at least one member selected from the group consisting of transmission of an electronic report and transmission of a hard copy report.
Claim 12: An automated system for monitoring and controlling contamination in a production area, the system comprising:
(A) a computer;
(B) an imaging sensor in communication with the computer, the imaging
sensor being configured and arranged to capture image data of at least a portion of the production area;
(C) computer-readable program code disposed on the computer, the computer- readable program code comprising:
(i) a first executable portion for processing image data and creating an image of the production area,
(ii) a second executable portion for processing image data to find an image of an individual or a portion of an individual in the production area,
(iii) a third executable portion for processing image data and determining whether an article of contamination control equipment is present in
association with the image of the individual or the image of the portion of the individual,
(iv) a fourth executable portion for processing image data and determining if the article of contamination control equipment is properly positioned on the individual while the individual is in the production area.
(v) a sixth executable portion for activating a contamination control device if the article of contamination control equipment is not present and properly positioned on the individual while the individual is present in the production area.
Claim 13 : The automated system according to claim 12, further comprising a seventh executable portion comprising a stabilization algorithm to determine whether the image data satisfies a threshold image value for a threshold time period, with the threshold image value being a pre-determined minimum image value correlating an absence of the contamination control equipment properly positioned on the individual, and the threshold time period being a pre-determined minimum time period that the threshold image value is satisfied.
Claim 14: The automated system according to claim 12, wherein the imaging sensor is a first imaging sensor and the system further comprises a second imaging sensor in communication with the computer, with the computer-readable program code disposed on the computer being provided with executable first, second, third, and fourth executable portions for creating and processing image data of at least a portion of the production area from the second imaging sensor, with the creating and processing of the image data from the second imaging sensor being carried out in a manner corresponding with the executable portions for capturing and processing image data from the first imaging sensor.
Claim 1 5 : The automated system according to claim 12, wherein the imaging sensor is a scanning imaging sensor configured and arranged to scan a production area.
Claim 16: The automated system according to claim 12, further comprising a data entry device that is in communication with the computer. Claim 17: The automated system according to claim 12, further comprising a printer that is in communication with the computer and is capable of printing a report of a determination of whether contamination control equipment is properly positioned on the individual in the production area.
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US34165510P | 2010-04-01 | 2010-04-01 | |
US12/928,362 US9189949B2 (en) | 2010-12-09 | 2010-12-09 | Automated monitoring and control of contamination in a production area |
PCT/US2011/030870 WO2011123743A1 (en) | 2010-04-01 | 2011-04-01 | Automated monitoring and control of contamination in a production area |
Publications (1)
Publication Number | Publication Date |
---|---|
EP2553628A1 true EP2553628A1 (en) | 2013-02-06 |
Family
ID=44063561
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP11717800A Withdrawn EP2553628A1 (en) | 2010-04-01 | 2011-04-01 | Automated monitoring and control of contamination in a production area |
Country Status (3)
Country | Link |
---|---|
EP (1) | EP2553628A1 (en) |
CA (1) | CA2795144A1 (en) |
WO (1) | WO2011123743A1 (en) |
Families Citing this family (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3080922A4 (en) * | 2013-12-11 | 2017-09-06 | Antisep - Tech Ltd. | Method and system for monitoring activity of an individual |
CA3015179A1 (en) | 2016-03-08 | 2016-12-08 | Antisep - Tech Ltd. | Method and system for monitoring activity of an individual |
US11288945B2 (en) | 2018-09-05 | 2022-03-29 | Honeywell International Inc. | Methods and systems for improving infection control in a facility |
US10978199B2 (en) | 2019-01-11 | 2021-04-13 | Honeywell International Inc. | Methods and systems for improving infection control in a building |
US11620594B2 (en) | 2020-06-12 | 2023-04-04 | Honeywell International Inc. | Space utilization patterns for building optimization |
US11783652B2 (en) | 2020-06-15 | 2023-10-10 | Honeywell International Inc. | Occupant health monitoring for buildings |
US11914336B2 (en) | 2020-06-15 | 2024-02-27 | Honeywell International Inc. | Platform agnostic systems and methods for building management systems |
US11783658B2 (en) | 2020-06-15 | 2023-10-10 | Honeywell International Inc. | Methods and systems for maintaining a healthy building |
US11823295B2 (en) | 2020-06-19 | 2023-11-21 | Honeywell International, Inc. | Systems and methods for reducing risk of pathogen exposure within a space |
US11184739B1 (en) | 2020-06-19 | 2021-11-23 | Honeywel International Inc. | Using smart occupancy detection and control in buildings to reduce disease transmission |
US11619414B2 (en) | 2020-07-07 | 2023-04-04 | Honeywell International Inc. | System to profile, measure, enable and monitor building air quality |
US11402113B2 (en) | 2020-08-04 | 2022-08-02 | Honeywell International Inc. | Methods and systems for evaluating energy conservation and guest satisfaction in hotels |
US11894145B2 (en) | 2020-09-30 | 2024-02-06 | Honeywell International Inc. | Dashboard for tracking healthy building performance |
US11372383B1 (en) | 2021-02-26 | 2022-06-28 | Honeywell International Inc. | Healthy building dashboard facilitated by hierarchical model of building control assets |
US11662115B2 (en) | 2021-02-26 | 2023-05-30 | Honeywell International Inc. | Hierarchy model builder for building a hierarchical model of control assets |
US11474489B1 (en) | 2021-03-29 | 2022-10-18 | Honeywell International Inc. | Methods and systems for improving building performance |
US12038187B2 (en) | 2021-09-28 | 2024-07-16 | Honeywell International Inc. | Multi-sensor platform for a building |
CN115546902B (en) * | 2022-11-30 | 2023-03-10 | 江苏未来网络集团有限公司 | Safety production management method and system based on industrial internet full-connection management |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6176782B1 (en) | 1997-12-22 | 2001-01-23 | Philips Electronics North America Corp. | Motion-based command generation technology |
AU2006337971B2 (en) * | 2006-02-10 | 2011-07-07 | Hyintel Limited | A system and method for monitoring hygiene standards compliance |
WO2007129289A1 (en) * | 2006-05-04 | 2007-11-15 | Provost Fellows And Scholars Of The College Of The Holy And Undivided Trinity Of Queen Elizabeth Near Dublin | A hand washing monitoring system |
KR20110051265A (en) * | 2008-09-03 | 2011-05-17 | 하이지넥스 인코포레이티드 | Methods and systems for monitoring hygiene habits |
-
2011
- 2011-04-01 WO PCT/US2011/030870 patent/WO2011123743A1/en active Application Filing
- 2011-04-01 EP EP11717800A patent/EP2553628A1/en not_active Withdrawn
- 2011-04-01 CA CA2795144A patent/CA2795144A1/en not_active Abandoned
Non-Patent Citations (2)
Title |
---|
None * |
See also references of WO2011123743A1 * |
Also Published As
Publication number | Publication date |
---|---|
WO2011123743A1 (en) | 2011-10-06 |
CA2795144A1 (en) | 2011-10-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9189949B2 (en) | Automated monitoring and control of contamination in a production area | |
EP2553628A1 (en) | Automated monitoring and control of contamination in a production area | |
US9143843B2 (en) | Automated monitoring and control of safety in a production area | |
US9406212B2 (en) | Automated monitoring and control of contamination activity in a production area | |
EP2553313B1 (en) | Automated monitoring and control of safety in a production area | |
US10198711B2 (en) | Methods and systems for monitoring or tracking products in a retail shopping facility | |
US10810414B2 (en) | Movement monitoring system | |
US9011607B2 (en) | Automated monitoring and control of cleaning in a production area | |
US20180165733A1 (en) | Notification system and methods for use in retail environments | |
ES2379196T3 (en) | Method and system to associate source information of a unit of origin, with a product transformed from it | |
US20190012794A1 (en) | Movement monitoring system | |
US20200226363A1 (en) | Detection and identification systems for humans or objects | |
US11238597B2 (en) | Suspicious or abnormal subject detecting device | |
US20230051146A1 (en) | Monitoring and Tracking Interactions With Inventory in a Retail Environment | |
US20230177458A1 (en) | Methods and systems for monitoring on-shelf inventory and detecting out of stock events | |
US10636321B2 (en) | Methods and systems for improving hand hygiene | |
WO2022092032A1 (en) | Activity estimating device, activity estimating method, program, and activity estimating system | |
US11302161B1 (en) | Monitoring and tracking checkout activity in a retail environment | |
Mamun et al. | Patient condition monitoring modular hospital robot | |
WO2023152893A1 (en) | Management device, management system, management method, and program | |
US20230290173A1 (en) | Circuitry and method | |
EP4371088A1 (en) | Touch-free seal check systems and methods for respiratory protection devices | |
CN115661854A (en) | Personnel detection device and detection method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
17P | Request for examination filed |
Effective date: 20121002 |
|
AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
DAX | Request for extension of the european patent (deleted) | ||
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: EXAMINATION IS IN PROGRESS |
|
17Q | First examination report despatched |
Effective date: 20170921 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN |
|
18D | Application deemed to be withdrawn |
Effective date: 20180202 |