CN113044694B - System and method for counting number of persons in building elevator based on deep neural network - Google Patents
System and method for counting number of persons in building elevator based on deep neural network Download PDFInfo
- Publication number
- CN113044694B CN113044694B CN202110243399.6A CN202110243399A CN113044694B CN 113044694 B CN113044694 B CN 113044694B CN 202110243399 A CN202110243399 A CN 202110243399A CN 113044694 B CN113044694 B CN 113044694B
- Authority
- CN
- China
- Prior art keywords
- module
- elevator
- unit
- data
- people
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 15
- 238000001514 detection method Methods 0.000 claims abstract description 32
- 238000010276 construction Methods 0.000 claims abstract description 24
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000013145 classification model Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 5
- 238000004148 unit process Methods 0.000 claims description 4
- 230000035772 mutation Effects 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 2
- 238000005303 weighing Methods 0.000 description 4
- 238000007726 management method Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B5/00—Applications of checking, fault-correcting, or safety devices in elevators
- B66B5/02—Applications of checking, fault-correcting, or safety devices in elevators responsive to abnormal operating conditions
- B66B5/14—Applications of checking, fault-correcting, or safety devices in elevators responsive to abnormal operating conditions in case of excessive loads
- B66B5/145—Applications of checking, fault-correcting, or safety devices in elevators responsive to abnormal operating conditions in case of excessive loads electrical
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B1/00—Control systems of elevators in general
- B66B1/34—Details, e.g. call counting devices, data transmission from car to control system, devices giving information to the control system
- B66B1/3476—Load weighing or car passenger counting devices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/14—Error detection or correction of the data by redundancy in operation
- G06F11/1402—Saving, restoring, recovering or retrying
- G06F11/1446—Point-in-time backing up or restoration of persistent data
- G06F11/1448—Management of the data involved in backup or backup restore
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/53—Recognition of crowd images, e.g. recognition of crowd congestion
-
- 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
-
- 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
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N5/00—Details of television systems
- H04N5/76—Television signal recording
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30242—Counting objects in image
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Automation & Control Theory (AREA)
- Human Computer Interaction (AREA)
- Biomedical Technology (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Quality & Reliability (AREA)
- Signal Processing (AREA)
- Mechanical Engineering (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Indicating And Signalling Devices For Elevators (AREA)
- Elevator Control (AREA)
Abstract
The invention discloses a system and a method for counting the number of persons in a building site elevator based on a deep neural network, which belong to the technical field of computer vision and target detection. The invention has the advantages of low cost, strong applicability, long-time continuous work, and contribution to reducing the control and management difficulty of the number of workers in the construction site elevator and improving the management efficiency.
Description
Technical Field
The invention relates to the technical field of computer vision and target detection, in particular to a system and a method for counting the number of persons in a construction elevator based on a deep neural network.
Background
The construction site elevator and the lifting operation platform are multifunctional lifting mechanical equipment, and can be divided into fixed type, movable type, crank arm type, scissor type and the like, and are commonly used for rescue and construction. The number control is an important measure for safety management of the construction site elevator, and a manual counting method and an elevator weighing method are generally adopted in the existing protection to control the number of persons in the construction site elevator not to exceed the upper limit of the weight load of the construction site elevator.
The following problems exist in the prior art: the people number control method needs staff to control the number of people, so that the labor cost is increased, and the accuracy of the people number control and the long-time work are difficult to ensure; the upper limit of the weight of the construction elevator is far higher than the theoretical total weight of people under the condition of full load, so that the weighing method is only suitable for civil elevators and is not suitable for the construction elevators.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide a system and a method for counting the number of persons in a construction elevator based on a deep neural network, which have the advantages of low cost, strong applicability, long-time continuous operation, contribution to reducing the difficulty of controlling and managing the number of persons in the construction elevator and improvement of management efficiency.
The system and the method for counting the number of persons in the building site elevator based on the deep neural network comprise a camera module, edge computing equipment and a prompt module, wherein the camera module collects image data in the building site elevator and transmits the image data to the edge computing equipment, and when the edge computing equipment detects that the number of persons in the image data exceeds the threshold, the prompt module sends out a threshold exceeding prompt.
The invention is further provided with: the edge computing module comprises a power supply module, an AI core computing module, a network module and a TTL serial port, wherein the TTL serial port is connected with a prompt module, the network module is connected with a cloud server, the prompt module sends out overguard prompt, the network module uploads overguard records to the cloud server, and the network module only stores the overguard records in a local storage unit under the condition that the network module cannot be connected to cloud service.
The invention is further provided with: a two-class type module is arranged between the camera module and the AI core operation module and is used for detecting whether the elevator door is closed or not, the two-class type detection module is connected with a physical switch of the elevator door of the construction site, and the AI core operation module starts to detect the number of people in the elevator and count the number when the two-class type module detects that the elevator door is closed and the physical switch is closed;
when the two classification model modules cannot be connected to the physical switch of the elevator door, the AI core operation module starts to detect and count the personnel in the elevator when the two classification model modules detect that the elevator door is closed.
The invention is further provided with: the AI core operation module comprises a data acquisition unit, a data processing unit and a model application unit, wherein the data acquisition unit intercepts data pictures in video shot by the camera unit, the data processing unit carries out preprocessing and enhancement on the input pictures through a computer vision algorithm, the data processing unit comprises color space conversion, image size conversion and image projection conversion operation, the model application unit detects the overload condition of human image data, and when the number of people exceeds the number of people, the data acquisition unit transmits the overload condition to the prompt module and opens the camera module to record the video.
The invention is further provided with: the model application unit comprises a detection model unit, a judging unit and a target counting unit, wherein the detection model unit identifies each person entering the site elevator, and the judging unit compares the position of the target with the theoretical movement limit when the detection model unit fails to identify.
The invention is further provided with: the TTL serial port comprises a 485-to-TTL serial port and a 232-to-TTL serial port, and the network module is one of various wired network modules and wireless network modules.
The invention is further provided with: and the detection model unit is connected with a KCF correlation filtering unit.
The invention is further provided with: the camera module is any one of a fisheye camera and a wide-angle camera, and the prompting module is any one of a tweeter and a loudspeaker.
The invention is further provided with: the AI core operation module adopts a model trained by a special number set and a derivative model thereof, wherein the model comprises a FASTER-RCNN, SSD and Yolo.
A system and a method for counting the number of persons in a building elevator based on a deep neural network are characterized by comprising the following steps:
s1, a camera module collects image data in site elevators at different times, and an AI core operation module starts to detect targets when an elevator door is detected to be closed;
s2, the data acquisition unit intercepts the data picture and transmits the data picture to the data processing unit;
s3, the data processing unit processes the image intercepted by the data acquisition unit and transmits the processed image to the detection model unit;
s4, the detection model unit identifies and tracks the personnel so as to ensure that tracking identification is not lost when the personnel entering the elevator is temporarily shielded, the detection model unit transmits the number of the personnel to the counting unit, and the prompting module is connected and a prompt is sent out when the number of the personnel exceeds the number of the personnel;
s5, when the fact that the elevator door of the construction site is closed for the second time and the number of people in the construction site is not reduced is detected, the prompting module alarms again, and at the moment, the volume is increased;
and S6, when the overtime alarm occurs, starting a video recording function, acquiring the picture data in a period of time before and after the time point by the equipment end, and transmitting the picture data and the video data to a cloud server for backup through a network module.
In summary, the invention has the following beneficial effects:
1. the camera module, the edge computing equipment and the prompting module are arranged to count and judge the number of people in the elevator at the construction site, and prompt the people when overtaking occurs, and can accurately count for a long time;
2. through the cloud server connected with the network module, when a secondary overman condition occurs in the site elevator, uploading the image to the cloud server for subsequent responsibility following;
3. and the target counting unit and the counting unit compare the target positions after the detection model unit fails to judge through a model application unit consisting of the detection model unit, the judging unit and the target counting unit, so that the counting accuracy is improved.
Drawings
FIG. 1 is a schematic diagram of the connection of various modules in the present invention;
FIG. 2 is a flow chart of a method for processing gray scale between video frames according to the present invention;
fig. 3 is a flow chart of a method of job site elevator demographics in the present invention.
Detailed Description
Further advantages and effects of the present invention will become apparent to those skilled in the art from the disclosure of the present invention when taken in conjunction with the accompanying drawings. The terms such as "upper", "lower", "left", "right", "middle" and the like are used in the present specification for convenience of description, but are not intended to limit the scope of the present invention, and the relative changes or modifications thereof are also regarded as the scope of the present invention without substantial modification of the technical content.
Examples:
as shown in fig. 1 to 3, the system and the method for counting the number of persons in a building site elevator based on a deep neural network designed in the invention comprise a camera module, an edge computing device and a prompt module, wherein the camera module is centrally arranged on a top plate of the building site elevator, the camera module collects image data in the building site elevator and transmits the image data to the edge computing device, the edge computing module comprises a power supply module, an AI core computing module, a network module and a TTL serial port, the TTL serial port is electrically connected with the prompt module, the network module is connected with a cloud server, the prompt module sends out a prompt for overtaking when the AI core computing module detects that the number of persons in the image data exceeds the threshold, the network module uploads the image data recorded by the overtaking to the cloud server, and the overtaking record is only stored in a local storage unit under the condition that the network module cannot be connected to cloud service. The network module is set to be a 4G network module, the camera module is set to be any one of a fisheye camera and a wide-angle camera, and the prompting module is any one of a tweeter and a loudspeaker.
As shown in fig. 2, a two-class model module is arranged between the camera module and the AI core operation module, the two-class model module is used for detecting whether the elevator door is closed, the two-class detection module is connected with a physical switch of the elevator door of the construction site, and the AI core operation module starts to detect the number of people in the elevator and counts the number when the two-class model module detects that the elevator door is closed and the physical switch is closed; when the two classification model modules cannot be connected to the physical switch of the elevator door, the AI core operation module starts to detect and count the personnel in the elevator when the two classification model modules detect that the elevator door is closed. The AI core operation module adopts a model trained by a special number set and a derivative model thereof, comprising a FASTER-RCNN, SSD and Yolo, wherein a block and a convolution layer are used in a CNN structure of a convolution neural network model.
As shown in fig. 2 and 3, the AI core operation module includes a data acquisition unit, a data processing unit and a model application unit, the data acquisition unit intercepts a data picture in a video shot by the camera unit, the data processing unit processes the data picture, the model application unit detects that the image data is out of stock, and when out of stock, the data is transmitted to the prompt module and the camera module is opened for video recording.
Head feature recognition based on head feature detection is generally classified into face recognition and head feature detection, wherein the head feature is easy to extract, and main features of the head are easy to recognize in a construction site, so head detection is adopted in the embodiment. The data acquisition unit intercepts medium data acquired by the video of the camera module, the data processing unit carries out inter-frame gray processing on the video transmitted by the camera module, binarizes the differential image, filters the binarized black-and-white image, refines the image, extracts the whole skeleton outline and transmits the skeleton outline to the model application unit, and finally the model application unit detects the edge of the head by using Hough transformation, and the flow is shown in figure 2.
As shown in fig. 3, the model application unit includes a detection model unit, a decision unit and a target counting unit, wherein the detection model unit identifies each person entering the site elevator and transmits the personnel to the target counting unit, counts the number of people in the site elevator, and tracks the movement track and direction of each person entering the site elevator.
As shown in fig. 3, the detection model unit is connected with a filtering unit, and in this embodiment, the filtering unit is set as a KCF correlation filtering unit, and a target detector is trained in the process of tracking the head of a person, and the target detector is used to detect whether the next predicted position has a target, and then the training data is updated according to the new detection result, so as to obtain a new and more accurate target detector, thereby solving the problem that the target object is blocked. When the identification of the tracked personnel fails due to factors such as shielding, rapid change of light, unstable plug flow of a camera and the like, the algorithm can be combined with the position information of the personnel identified before the identification failure and the position information of the personnel identified after the identification failure to be compared, if the theoretical movement limit is not exceeded and the counted number of people does not have unreasonable mutation, the repositioning is considered to be successful, the number of people is counted correctly, and the accuracy of the number of people counting is effectively guaranteed not to be influenced.
The invention also provides a method for counting the number of persons in the elevator on the basis of the deep neural network, which comprises the following steps:
s1, a camera module collects image data in site elevators at different times, and an AI core operation module starts to detect targets when an elevator door is detected to be closed;
s2, the data acquisition unit intercepts the data picture and transmits the data picture to the data processing unit;
s3, the data processing unit processes the image intercepted by the data acquisition unit and transmits the processed image to the detection model unit;
s4, the detection model unit identifies and tracks the personnel so as to ensure that tracking identification is not lost when the personnel entering the elevator is temporarily shielded, the detection model unit transmits the number of the personnel to the counting unit, and the prompting module is connected and a prompt is sent out when the number of the personnel exceeds the number of the personnel;
s5, when the fact that the elevator door of the construction site is closed for the second time and the number of people in the construction site is not reduced is detected, the prompting module alarms again, and at the moment, the volume is increased;
and S6, when the overtime alarm occurs, starting a video recording function, acquiring the picture data in a period of time before and after the time point by the equipment end, and transmitting the picture data and the video data to a cloud server for backup through a network module.
The camera module continuously collects images in the elevator at the construction site, the data collecting unit intercepts data collected by the camera unit, the data processing unit selects pictures with persons in downloaded pictures, the pictures are processed by adopting an inter-frame gray level method, the difference map is binarized, and image labeling and label selection are carried out on the difference map. After the elevator door is closed, the data processing unit transmits the processed image to the model application unit, the detection model unit identifies the head marked in the image and tracks the movement track and direction of the head marked in the image, the target counting unit counts the number of people, and when the number of people in the elevator exceeds the upper load limit, the prompting module is connected to send out an alarm.
When the elevator door is closed again, the number of people in the site elevator is detected again, when the number of people in the site elevator is detected not to be reduced, the prompt module is switched on again to send out a loud alarm sound, the camera module starts video recording, at the moment, the time point of overtaking in the site elevator is acquired from the web end, then the equipment end acquires picture data in a period of time before and after the time point, and the picture data and the video are uploaded to the cloud server end by the network module together with the video and are sent to site responsible people.
When the detection model unit fails to identify, the judgment unit combines the position information of the person identified once before the identification fails, the position information of the person identified once after the identification fails is compared, and if the theoretical movement limit is not exceeded and the number of people counted by the target counting unit does not have unreasonable mutation, the number of people is judged to be correct in statistics. The elevator weighing system replaces the manual counting and elevator weighing method in the prior art, can continuously count and monitor the number of people in the elevator for a long time, reduces the difficulty of controlling and managing the number of people in the construction site elevator, and improves the management efficiency.
The present embodiment is only for explanation of the present invention and is not to be construed as limiting the present invention, and modifications to the present embodiment, which may not creatively contribute to the present invention as required by those skilled in the art after reading the present specification, are all protected by patent laws within the scope of claims of the present invention.
Claims (6)
1. A building elevator people counting system based on a deep neural network is characterized in that:
the system comprises a camera module, edge computing equipment and a prompt module, wherein the camera module collects image data in a building site elevator and transmits the image data to the edge computing equipment;
the edge computing equipment comprises a power supply module, an AI core operation module, a network module and a TTL (transistor-transistor logic) serial port, wherein the TTL serial port is connected with a prompt module, the network module is connected with a cloud server, the prompt module sends out overranging prompt, the network module uploads overranging records to the cloud server, and the network module only stores the overranging records in a local storage unit under the condition that the network module cannot be connected to cloud service;
a two-class model module is arranged between the camera module and the AI core operation module and is used for detecting whether the elevator door is closed or not, the two-class model module is connected with a physical switch of the elevator door of the construction site, and when the two-class model module detects that the elevator door is closed and the physical switch is closed, the AI core operation module starts to detect the number of people in the elevator and counts the number;
when the two classification model modules cannot be connected to the physical switch of the elevator door, and when the two classification model modules detect that the elevator door is closed, the AI core operation module starts to detect personnel in the elevator and count the number;
the AI core operation module comprises a data acquisition unit, a data processing unit and a model application unit, wherein the data acquisition unit intercepts data pictures in video shot by the camera module, the data processing unit carries out preprocessing and enhancement on the input pictures through a computer vision algorithm, the data processing unit comprises color space conversion, image size conversion and image projection conversion operation, the model application unit detects the overload condition of human image data, and when the number of people exceeds the number of people, the model application unit transmits the overload condition to the prompt module and opens the camera module to record the video;
the model application unit comprises a detection model unit, a judging unit and a target counting unit, wherein the detection model unit identifies each person entering the site elevator, when the detection model unit fails to identify, the judging unit compares the position of the target with the theoretical movement limit, the position information of the person identified at the last time of the identification failure and the position information of the person identified at the last time of the identification failure are compared, if the theoretical movement limit is not exceeded and the counted number of people does not have unreasonable mutation, the repositioning is considered to be successful, and the number of people is counted correctly;
the camera module collects image data within the worksite elevator at different times.
2. The deep neural network-based work elevator people counting system according to claim 1, wherein:
the TTL serial port comprises a 485-to-TTL serial port and a 232-to-TTL serial port, and the network module is one of various wired network modules and wireless network modules.
3. The deep neural network-based work elevator people counting system according to claim 1, wherein:
and the detection model unit is connected with a KCF correlation filtering unit.
4. The deep neural network-based work elevator people counting system according to claim 1, wherein:
the camera module is any one of a fisheye camera and a wide-angle camera, and the prompting module is any one of a tweeter and a loudspeaker.
5. The deep neural network-based industrial personal elevator people counting system according to claim 4, wherein:
the AI core operation module adopts a model trained by a special number set and a derivative model thereof, wherein the model comprises a FASTER-RCNN, SSD and Yolo.
6. A method for counting the number of persons in a building elevator based on a deep neural network according to any one of claims 1 to 5,
the method comprises the following steps:
s1, a camera module collects image data in site elevators at different times, and an AI core operation module starts to detect targets when an elevator door is detected to be closed;
s2, the data acquisition unit intercepts the data picture and transmits the data picture to the data processing unit;
s3, the data processing unit processes the image intercepted by the data acquisition unit and transmits the processed image to the detection model unit;
s4, the detection model unit identifies and tracks the personnel so as to ensure that tracking identification is not lost when the personnel entering the elevator is temporarily shielded, the detection model unit transmits the number of the personnel to the counting unit, and the prompting module is connected and a prompt is sent out when the number of the personnel exceeds the number of the personnel;
s5, when the fact that the elevator door of the construction site is closed for the second time and the number of people in the construction site is not reduced is detected, the prompting module alarms again, and at the moment, the volume is increased;
and S6, when the overtime alarm occurs, starting a video recording function, and transmitting the picture data and the video data to a cloud server for backup through a network module after and before the overtime time point of the building site ladder acquired from the Web end.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110243399.6A CN113044694B (en) | 2021-03-05 | 2021-03-05 | System and method for counting number of persons in building elevator based on deep neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110243399.6A CN113044694B (en) | 2021-03-05 | 2021-03-05 | System and method for counting number of persons in building elevator based on deep neural network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113044694A CN113044694A (en) | 2021-06-29 |
CN113044694B true CN113044694B (en) | 2023-10-24 |
Family
ID=76510227
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110243399.6A Active CN113044694B (en) | 2021-03-05 | 2021-03-05 | System and method for counting number of persons in building elevator based on deep neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113044694B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114604700B (en) * | 2021-11-22 | 2023-08-29 | 杭州西奥电梯有限公司 | Method for counting and detecting number of people in elevator car |
CN114314243A (en) * | 2022-01-10 | 2022-04-12 | 徐工汉云技术股份有限公司 | Elevator overload alarm system and method based on video identification technology |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109544532A (en) * | 2018-11-20 | 2019-03-29 | 四川长虹电器股份有限公司 | Construction elevator demographic method based on image recognition |
CN110002314A (en) * | 2018-08-09 | 2019-07-12 | 浙江新再灵科技股份有限公司 | A kind of stranded number detection system of the elevator passenger based on deep learning |
CN110040592A (en) * | 2019-04-15 | 2019-07-23 | 福建省星云大数据应用服务有限公司 | Lift car carrying number detection method and system based on the analysis of two-way monitor video |
CN111199220A (en) * | 2020-01-21 | 2020-05-26 | 北方民族大学 | Lightweight deep neural network method for people detection and people counting in elevator |
CN111353377A (en) * | 2019-12-24 | 2020-06-30 | 浙江工业大学 | Elevator passenger number detection method based on deep learning |
-
2021
- 2021-03-05 CN CN202110243399.6A patent/CN113044694B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110002314A (en) * | 2018-08-09 | 2019-07-12 | 浙江新再灵科技股份有限公司 | A kind of stranded number detection system of the elevator passenger based on deep learning |
CN109544532A (en) * | 2018-11-20 | 2019-03-29 | 四川长虹电器股份有限公司 | Construction elevator demographic method based on image recognition |
CN110040592A (en) * | 2019-04-15 | 2019-07-23 | 福建省星云大数据应用服务有限公司 | Lift car carrying number detection method and system based on the analysis of two-way monitor video |
CN111353377A (en) * | 2019-12-24 | 2020-06-30 | 浙江工业大学 | Elevator passenger number detection method based on deep learning |
CN111199220A (en) * | 2020-01-21 | 2020-05-26 | 北方民族大学 | Lightweight deep neural network method for people detection and people counting in elevator |
Also Published As
Publication number | Publication date |
---|---|
CN113044694A (en) | 2021-06-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110263686A (en) | A kind of construction site safety of image cap detection method based on deep learning | |
EP4105101A1 (en) | Monitoring system, monitoring method, and monitoring device for railway train | |
CN113044694B (en) | System and method for counting number of persons in building elevator based on deep neural network | |
CN101465033B (en) | Automatic tracking recognition system and method | |
CN112233300A (en) | Community passing epidemic prevention monitoring joint defense system and method based on artificial intelligence | |
CN209543514U (en) | Monitoring and alarm system based on recognition of face | |
CN113411542A (en) | Intelligent working condition monitoring equipment | |
CN112560745B (en) | Method for discriminating personnel on electric power operation site and related device | |
CN109614906A (en) | A kind of security system and security alarm method based on deep learning | |
CN111540152A (en) | Intelligent security early warning system | |
CN116563797B (en) | Monitoring management system for intelligent campus | |
CN111079694A (en) | Counter assistant job function monitoring device and method | |
CN214896720U (en) | Multifunctional access control device for community epidemic prevention and control | |
KR20080044812A (en) | The automatic guard system to prevent the crime and accident using computer video image analysis technology | |
CN111510675A (en) | Intelligent security system based on face recognition and big data analysis | |
CN111339811A (en) | Image processing method, device, equipment and storage medium | |
CN112102574B (en) | Intelligent alarm system for security room | |
KR101989376B1 (en) | Integrated track circuit total monitoring system | |
CN110809137A (en) | Campus safety trampling prevention monitoring system and method | |
CN111523386A (en) | Machine vision-based high-speed railway platform door monitoring and protecting method and system | |
CN105095891A (en) | Human face capturing method, device and system | |
CN112532927A (en) | Intelligent safety management and control system for construction site | |
CN109614916B (en) | Real-time monitoring and early warning computer system for station passenger state based on face recognition | |
CN116912037A (en) | Safety monitoring system and method for building construction based on image processing | |
CN114783040B (en) | Detection method and detection device for candid shooting in building |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |