CN115625723B - Social distance monitoring robot based on YOLO algorithm - Google Patents
Social distance monitoring robot based on YOLO algorithm Download PDFInfo
- Publication number
- CN115625723B CN115625723B CN202211653065.7A CN202211653065A CN115625723B CN 115625723 B CN115625723 B CN 115625723B CN 202211653065 A CN202211653065 A CN 202211653065A CN 115625723 B CN115625723 B CN 115625723B
- Authority
- CN
- China
- Prior art keywords
- social distance
- circuit board
- human
- control circuit
- pressure sensing
- 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
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J11/00—Manipulators not otherwise provided for
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J11/00—Manipulators not otherwise provided for
- B25J11/0005—Manipulators having means for high-level communication with users, e.g. speech generator, face recognition means
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J19/00—Accessories fitted to manipulators, e.g. for monitoring, for viewing; Safety devices combined with or specially adapted for use in connection with manipulators
- B25J19/02—Sensing devices
-
- 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
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Emergency Alarm Devices (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses a social distance monitoring robot based on a YOLO algorithm, and relates to the technical field of indoor social distance monitoring. According to the invention, the array type pressure sensing film assembly is adopted to monitor the outline image and the position of the pressure signal of the human feet from the ground, and the matching of the image database of the standing state of the human feet is carried out through the YOLOV5 algorithm, so that the miscellaneous influence of directly monitoring the social distance by adopting a video is avoided, and the efficiency and the accuracy of monitoring the social distance are greatly improved.
Description
Technical Field
The invention relates to the technical field of indoor social distance monitoring, in particular to a social distance monitoring robot based on a YOLO algorithm.
Background
More and more indoor public places use robots to perform auxiliary services and complete some guiding work. The robot is generally called a service robot, has a wide application range, and is mainly used for monitoring, security, cleaning, shopping guide and the like.
At present, with factors such as epidemic situations, more and more public places need to arrange professional staff for controlling the social distance of the staff so as to reduce the gathering of the staff, for example, places such as banks and stations, some places also adopt a fixed camera to shoot fixed places, and adopt an image processing method to monitor the social distance.
However, the manual monitoring mode cannot cover the place to be monitored well, one side is often taken into consideration, the other side is ignored, the problem of communication between people also exists, the communication is over excited, conflicts are caused, a large amount of manpower is wasted, and the effect is not good;
the mode of camera monitoring can realize that there is the condition of distance too near to carry out global monitoring, but, the mode of reminding is relatively weak, exists the condition that can discover the distance too near, but can not accurately remind effectively, and the effect is also not good.
The mobile monitoring mode is also adopted, the camera carried by the robot is used for shooting pictures, image processing is carried out, and the social distance is obtained, so that the robot can accurately remind a place with an excessively small social distance. In the mode, because the robot is in the continuous motion process, the shot pictures are limited and unstable, the image processing result is inaccurate, the situation of misinformation often occurs, the object to be reminded is disliked and even complained, and the effect is also poor.
Disclosure of Invention
In order to solve the technical problems of accurately and quickly monitoring the social distance of indoor personnel and accurately reminding correction, the invention provides a social distance monitoring robot based on a YOLO algorithm. The following technical scheme is adopted:
a social distance monitoring robot based on a YOLO algorithm comprises an array type pressure sensing film assembly, a wireless transmission module, a service robot and a social distance reminding module, wherein the array type pressure sensing film assembly is laid below a floor of a required social distance monitoring area and is in wireless communication connection with a main control circuit board of the service robot through the wireless transmission module, and the service robot judges social distance according to data monitored by the array type pressure sensing film assembly;
the method for judging the social distance comprises the following steps: monitoring contour information, weight value information and contour position information of a pressure generating source above a floor at intervals of 30-60 seconds by the array type pressure sensing film assembly, transmitting the contour information, the weight value information and the contour position information to a main control circuit board through a wireless transmission module, and carrying out imaging processing on the contour of the pressure generating source by a processor of the main control circuit board to obtain a contour image;
all outlines of the outline images are compared with preset standard biped outlines of human standing postures by adopting a YOLOV5 algorithm, the similarity IOU value is larger than a set similarity threshold value, a pressure generation source is judged to be a human, the central point of the biped outlines is marked as a social distance monitoring point, when the distance between any two social distance monitoring points is smaller than a set distance warning threshold value, the social distance is determined to be too small, the master control circuit board controls the service robot to move to one side of the occurrence position of the too small social distance, and the social distance reminding module is started to remind that the social distance is kept.
Through the technical scheme, the array type pressure sensing film assembly laid in advance is used for monitoring the pressure signal from the ground, the contour information, the weight information and the position information of the pressure source can be accurately identified, monitoring is carried out at intervals of 30-60 seconds instead of continuous monitoring, and once the situation that the social distance is too close is found, the service robot needs time for on-site treatment, and the good monitoring effect cannot be achieved at overlong time intervals.
After a monitoring period is completed, the master control circuit board obtains profile information, weight value information and position information of a pressure generation source above the array type pressure sensing film assembly monitoring floor through the wireless transmission module, a processor of the master control circuit board compares the monitored profile information with a human biped profile library preset in a database in multiple states one by one, a YOLO algorithm is mainly adopted for comparing similarity, when the similarity IOU value is larger than a set similarity threshold value, the profile pressure generation source can be determined to be human, a central point of the biped profile is marked, the distance between the central points is the social distance obtained by monitoring, when the distance between any two social distance monitoring points is smaller than a set distance warning threshold value, the social distance is determined to be too small, the service robot can move to one side where the social distance is generated, the social distance reminding module is started to remind that the social distance is kept to be too small, and efficient and accurate social distance monitoring and accurate reminding of an indoor area can be achieved.
Optionally, the array pressure sensing membrane assembly is a piezoresistive distributed flexible membrane.
According to the technical scheme, the piezoresistive distributed flexible film is manufactured by transferring materials such as sensing force sensitive materials and silver paste to a flexible film base material through a precision printing process, and drying and curing the materials. The method can adopt the arrangement of a plurality of rows of array force-sensitive materials according to the size of an application scene, the total sensing point number is related to the density of the arranged sensing force-sensitive materials, when the pressure is applied, the resistance is reduced along with the increase of the pressure, the piezoresistive characteristic is expressed that the resistance and the pressure are in a power function relationship, and the reciprocal of the resistance and the pressure are in an approximate linear relationship.
The piezoresistive distributed flexible film has the advantages that the film thickness is not more than 0.6mm, the effective detection area is large, the plantar pressure distribution below 50 yards can be measured, the response speed is high, the response time is less than 1ms, the recovery time is less than 15ms, the high-speed dynamic scanning requirement is met, the detection pressure range is large, a single point can detect 1Mpa maximally, customization can be realized, and the external dimension, the pressure point dimension, the density and the pressure measurement range can be customized according to the user requirement.
Optionally, the social distance warning module includes electronic cloud platform and projection arrangement, electronic cloud platform fixed mounting is at service robot's top, projection arrangement installs on the rotation portion of electronic cloud platform, and the orientation of electronic cloud platform rotation portion of main control circuit board control, projection arrangement projects the pattern of setting for shape and colour on ground, the shape of pattern is straight line, curve or rectangular, the colour of pattern is green, yellow or red.
Optionally, the projection device is a laser projection lamp, and the pattern further includes text reminding information.
Through the technical scheme, the social distance reminding module does not adopt a traditional independent voice reminding mode, the situation that a person to be reminded is not clear to remind himself or herself, when the social distance of a related person needs to be reminded to be too close, the main control circuit board controls the rotating portion of the electric pan-tilt to face the central point connecting line area of the outline of the two feet determined to be too close in social distance, the projection device is turned on, the projection device can project red straight lines or other warning patterns in the central point connecting line area, and red text information of 'the distance between your people is \8230, the distance between your people and the person is too close, please keep the distance' can be projected, so that the person who is too close in social distance is reminded to keep the social distance, the reminding mode is more direct, more striking and the reminding effect is better.
Optionally, the social distance reminding module further includes a voice reminding sound box.
Optionally, the voice reminding sound is a voice sound of the service robot.
Through above-mentioned technical scheme, the mode that the pronunciation were reminded can be added again to the mode of reminding, further reminds the keeping of social distance, and plays the warning effect to other people on every side.
Optionally, the pressure sensing film assembly further comprises an interval power supply module, wherein the interval power supply module comprises a power adapter and a time control switch, and the power adapter is connected to the mains supply and is electrically connected with the array pressure sensing film assembly through the time control switch.
Through the technical scheme, the mode of the time control switch is adopted to realize the interval power supply of the array type pressure sensing film assembly, and the array type pressure sensing film assembly is completely disconnected when not being monitored, so that the energy is saved.
Optionally, the wireless transmission module is a WIFI wireless module.
Through above-mentioned technical scheme, wireless transmission module can be the wireless WIFI module that indoor scene erect, and data transmission is fast, and is more stable.
Optionally, the method for determining the social distance specifically includes:
and 5, the master control circuit board controls the service robot to carry out path planning, the service robot moves to one side of a connecting line of central points of two adjacent human biped standing state images with too small social distance, a projection device is started to project patterns to the ground, meanwhile, a voice reminding sound box starts voice reminding until a person to be reminded increases the social distance S, and when S is larger than 1.2X, the master control circuit board controls the service robot to close the projection device and the voice reminding sound box and leave.
In step 3, the similarity IOU value is 0.9-0.95; in step 4, the distance threshold X is 0.5-1.5 meters.
Through the technical scheme, after the array type pressure sensing film assembly is laid and the floor is covered, the pressure value monitored by the array type pressure sensing film assembly can be subjected to zero treatment, the weight value m measured later is set as a monitoring threshold, when the weight value m is more than 20kg, the pressure profile generated by human can be determined, other miscellaneous influences such as a pet meeting chair and the like on human social distance monitoring are eliminated, and the time consumption of an algorithm is reduced;
a plurality of human plantar pressure contour images are prestored in a human biped standing state image database, and the human plantar pressure contour images comprise three foot types including a high arch foot, a flat foot and a normal foot, and the three foot types wear the human plantar pressure contour images under different sole states. The YOLOV5 algorithm is adopted to realize comparison and identification of the monitored plantar pressure contour images, and the method also belongs to training of a YOLOv5 model for target identification in the continuous identification process, so that the identification precision and speed can be greatly improved in the using process.
In summary, the invention includes at least one of the following beneficial technical effects:
the invention provides a social distance monitoring robot based on a YOLO algorithm, which adopts an array type pressure sensing film assembly to monitor contour images and positions of human biped pressure signals from the ground, and then matches a human biped standing state image database through a YOLOV5 algorithm, thereby avoiding the miscellaneous influence of directly monitoring the social distance by adopting a video and greatly improving the efficiency and the accuracy of monitoring the social distance;
by adopting the dual reminding mode of matching the projection device with voice reminding, people with too short social distance can immediately and accurately realize that the design distance is too close, and the reminding and correcting effect is better.
Drawings
FIG. 1 is a schematic diagram of the electrical device connection principle of the structure of the present invention;
FIG. 2 is a schematic view of a pressure sensing diaphragm assembly monitoring a pressure profile image in accordance with an embodiment of the present invention;
FIG. 3 is a diagram illustrating a target detection and recognition result of the pressure contour image by the YOLO algorithm according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of an assembled state of the projection device and the service robot according to the present invention.
Description of the reference numerals: 1. an array pressure sensing membrane assembly; 2. a wireless transmission module; 3. a service robot; 31. a main control circuit board; 4. an electric pan-tilt; 5. a projection device; 6. a voice reminding sound box; 71. a power adapter; 72. and (4) a time control switch.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings.
The embodiment of the invention discloses a social distance monitoring robot based on a YOLO algorithm.
Referring to fig. 1-4, a social distance monitoring robot based on a YOLO algorithm includes an array pressure sensing thin film assembly 1, a wireless transmission module 2, a service robot 3 and a social distance reminding module, wherein the array pressure sensing thin film assembly 1 is laid under a floor of a required social distance monitoring area and is in wireless communication connection with a main control circuit board 31 of the service robot 3 through the wireless transmission module 2, and the service robot 3 determines a social distance according to data monitored by the array pressure sensing thin film assembly 1;
the method for judging the social distance comprises the following steps: the array type pressure sensing film assembly 1 monitors contour information, weight value information and contour position information of a pressure generating source above a floor at intervals of 30-60 seconds and transmits the contour information, the weight value information and the contour position information to the main control circuit board 31 through the wireless transmission module 2, and a processor of the main control circuit board 31 carries out imaging processing on the contour of the pressure generating source to obtain a contour image;
all outlines of the outline images are compared with preset standard biped outlines of human standing postures by adopting a YOLOV5 algorithm, the similarity IOU value is larger than a set similarity threshold value, a pressure generation source is judged to be a human, the central point of the biped outlines is marked as a social distance monitoring point, when the distance between any two social distance monitoring points is smaller than a set distance warning threshold value, the social distance is determined to be too small, the main control circuit board 31 controls the service robot 3 to move to one side of a place where the social distance is too small, and a social distance reminding module is started to remind that the social distance is kept.
The pressure signal from the ground is monitored by the aid of the pre-laid array type pressure sensing film assembly 1, contour information, weight information and position information of a pressure source can be accurately identified, monitoring is carried out at intervals of 30-60 seconds instead of continuous monitoring, and once the situation that the social distance is too close is found, the service robot 3 needs time to go to the site for treatment, and good monitoring effect cannot be achieved at overlong time intervals.
After a monitoring period is completed, the main control circuit board 31 obtains profile information, weight value information and position information of a pressure generation source above a monitoring floor of the array type pressure sensing film assembly 1 through the wireless transmission module 2, a processor of the main control circuit board 31 compares the monitored profile information with a human biped foot type library in multiple preset states in a database one by one, similarity comparison is mainly carried out by adopting a YOLO algorithm, when the IOU value of the similarity is larger than a set similarity threshold value, the profile pressure generation source can be determined to be human, at the moment, a central point of the biped profile is marked, the distance between the central points is the social distance obtained by monitoring, when the distance between any two social distance monitoring points is smaller than a set distance warning threshold value, the social distance is determined to be too small, the service robot 3 can move to the side where the social distance occurs, a social distance reminding module with too small social distance is started to remind that the social distance is kept, and efficient and accurate monitoring and reminding of the social distance in an indoor area can be realized.
The array pressure sensing diaphragm assembly 1 is a piezoresistive distributed flexible diaphragm.
The piezoresistive distributed flexible film is manufactured by transferring materials such as sensing force sensitive materials, silver paste and the like onto a flexible film base material through a precision printing process, and drying and curing. The method can adopt the arrangement of a plurality of rows of array force-sensitive materials according to the size of an application scene, the total sensing point number is related to the density of the arranged sensing force-sensitive materials, when the pressure is applied, the resistance is reduced along with the increase of the pressure, the piezoresistive characteristic is expressed that the resistance and the pressure are in a power function relationship, and the reciprocal of the resistance and the pressure are in an approximate linear relationship.
The piezoresistive distributed flexible film has the advantages that the film thickness is not more than 0.6mm, the effective detection area is large, the plantar pressure distribution below 50 yards can be measured, the response speed is high, the response time is less than 1ms, the recovery time is less than 15ms, the high-speed dynamic scanning requirement is met, the detection pressure range is large, a single point can detect 1Mpa maximally, customization can be realized, and the external dimension, the pressure point dimension, the density and the pressure measurement range can be customized according to the user requirement.
The social distance reminding module comprises an electric holder 4 and a projection device 5, the electric holder 4 is fixedly installed at the top of the service robot 3, the projection device 5 is installed on a rotating portion of the electric holder 4, the main control circuit board 31 controls the orientation of the rotating portion of the electric holder 4, the projection device 5 projects a pattern with a set shape and color on the ground, the pattern is a straight line, a curve or a strip, and the pattern is green, yellow or red in color.
The projection device 5 is a laser projection lamp, and the pattern also comprises text reminding information.
The social distance reminding module does not adopt a traditional independent voice reminding mode any more, the situation that a person to be reminded is not clear to remind himself or herself and a related person needs to be reminded that the social distance is too short is avoided, the main control circuit board 31 controls the rotating part of the electric holder 4 to face a central point connecting line area of a two-foot outline determined to be too short in social distance, the projection device 5 is turned on afterwards, the projection device 5 can project a red straight line or other warning patterns in the central point connecting line area, and red text information of ' the distance between your is 8230 ' \ 8230, meter, too short in distance and please keep the distance ' can be projected, so that the person who is too short in social distance is reminded to keep the social distance, the reminding mode is more direct, more striking and better in reminding effect.
The social distance reminding module further comprises a voice reminding sound box 6.
The voice prompt sound 6 is a voice sound of the service robot 3.
The mode of reminding can add the mode that the pronunciation was reminded again, further reminds keeping of social distance, and plays the warning effect to other people around.
The pressure sensing film assembly further comprises an interval power supply module, the interval power supply module comprises a power adapter 71 and a time control switch 72, the power adapter 71 is connected to mains supply, and is electrically connected with the array type pressure sensing film assembly 1 through the time control switch 72.
The mode of the time control switch 72 is adopted to realize the interval power supply for the array type pressure sensing film assembly 1, and the power supply is completely disconnected when the monitoring is not carried out, so that the energy is saved.
The wireless transmission module 2 is a wireless WIFI module.
The method for judging the social distance specifically comprises the following steps:
step 1: the time control switch 72 controls the array type pressure sensing membrane assembly 1 to be electrified for 10 seconds at the time interval of 30-60 seconds;
and 2, step: in a power-on state, the array type pressure sensing film assembly 1 wirelessly transmits an outline signal and a weight value of a pressure generation source above a monitored floor to the main control circuit board 31 through the wireless transmission module 2, and a processor of the main control circuit board 31 converts an outline of a pressure signal with the weight value m greater than 20kg monitored in the power-on 10 th second into an image A;
and 3, step 3: adopting a YOLOV5 algorithm to carry out target detection and identification on different foot types of the pressure signal profile image, judging a paired foot type area with the identification similarity IOU value larger than a set similarity threshold value as a human biped standing state, carrying out maximum profile frame selection on the paired human biped standing state area, marking the framed area as Fn, wherein n represents the number of the identified human biped standing state images, marking the central point Cn of the Fn, and obtaining the position coordinate Sn of the Cn in the images;
and 4, step 4: calculating the space S, S = | of any two adjacent human biped standing state images
| setting a distance threshold value X, and when S is less than X, indicating that the human social distance corresponding to the image position of the human biped standing state is too small;
and step 5, the main control circuit board 31 controls the service robot 3 to carry out path planning, the service robot moves to one side of a connecting line of central points of two adjacent human biped standing state images with too small social distance, the projection device 5 is started to project patterns to the ground, meanwhile, the voice reminding sound 6 starts voice reminding until the social distance S is increased by a person to be reminded, and when S is larger than 1.2X, the main control circuit board 31 controls the service robot 3 to stop the projection device 5 and the voice reminding sound 6 and leave.
In step 3, the similarity IOU value is 0.9-0.95; in step 4, the distance threshold X is 0.5-1.5 meters.
After the array type pressure sensing film assembly 1 is laid and a floor is covered, the pressure value monitored by the array type pressure sensing film assembly 1 can be subjected to zero resetting, a monitoring threshold is set for the weight value m measured later, when m is larger than 20kg, the pressure profile generated by human can be determined, other miscellaneous influences such as a small pet meeting chair on human social distance monitoring are eliminated, and the time consumption of an algorithm is reduced;
a plurality of human plantar pressure contour images including three foot types of a high arch foot, a flat foot and a normal foot are prestored in a human biped standing state image database, and the three foot types wear the human plantar pressure contour images in different sole states. The YOLOV5 algorithm is adopted to realize comparison and identification of the monitored plantar pressure contour images, and the method also belongs to training of a YOLOv5 model for target identification in the continuous identification process, so that the identification precision and speed can be greatly improved in the using process.
The invention relates to a social distance monitoring robot specific implementation principle based on a YOLO algorithm, which comprises the following steps:
under the social distance monitoring application scene of a certain bank, the time control switch 72 controls the array type pressure sensing film assembly 1 to be electrified for 10 seconds at the time interval of 60 seconds;
in a certain monitoring cycle, in a power-on state, the array type pressure sensing film assembly 1 wirelessly transmits an outline signal and a weight value of a pressure generation source above a monitored floor to the main control circuit board 31 through the wireless transmission module 2, and a processor of the main control circuit board 31 converts an outline of a pressure signal with a weight value m greater than 20kg monitored in the power-on 10 th second into an image A, which is shown in fig. 2;
adopting a YOLOV5 algorithm to carry out target detection and identification on different foot types of the pressure signal profile image, judging a paired foot type area with the identification similarity IOU value larger than a set similarity threshold value as a human biped standing state, carrying out maximum profile frame selection on the paired human biped standing state area, marking the framed area as Fn, wherein n represents the number of the identified human biped standing state images, marking the central point Cn of the Fn, and obtaining the position coordinate Sn of the Cn in the images, which is shown in figure 3;
calculating the space S, S = | of any two adjacent human biped standing state images
| setting a distance threshold value X =0.8 m, and calculating to obtain that the distance S =0.73 m and is smaller than 0.8 m of F2 and F3, which indicates that the social distance of the human corresponding to the image position of the human biped standing state is too small;
master control circuit board 31 control service robot 3 carries out route planning, move to the one side that takes place two adjacent people 'S both feet standing state image central point lines of social distance undersize, open projection arrangement 5, to ground projection pattern, pronunciation warning stereo set 6 opens pronunciation simultaneously and reminds, until being reminded personnel with social distance S increase, when F2 and F3' S interval S > 0.96 meters, master control circuit board 31 control service robot 3 closes projection arrangement 5 and pronunciation warning stereo set 6, and leave, whole warning and correction process are accomplished in 15 seconds, high efficiency, it is more accurate.
The above are all the preferred embodiments of the present invention, and the protection scope of the present invention is not limited thereby, so: all equivalent changes made according to the structure, shape and principle of the invention are covered by the protection scope of the invention.
Claims (10)
1. A social distance monitoring robot based on a YOLO algorithm is characterized in that: the intelligent social distance monitoring system comprises an array type pressure sensing film assembly (1), a wireless transmission module (2), a service robot (3) and a social distance reminding module, wherein the array type pressure sensing film assembly (1) is laid below the floor of a required social distance monitoring area and is in wireless communication connection with a main control circuit board (31) of the service robot (3) through the wireless transmission module (2), and the service robot (3) judges the social distance according to data monitored by the array type pressure sensing film assembly (1);
the method for judging the social distance comprises the following steps: the method comprises the following steps that the array type pressure sensing film assembly (1) monitors outline information, weight value information and outline position information of a pressure generating source above a floor at intervals of 30-60 seconds, the outline information, the weight value information and the outline position information are transmitted to a main control circuit board (31) through a wireless transmission module (2), and a processor of the main control circuit board (31) carries out imaging processing on the outline of the pressure generating source to obtain an outline image;
all outlines of the outline images are compared with preset standard biped outlines of human standing postures by adopting a YOLOV5 algorithm, the similarity IOU value is larger than a set similarity threshold value, a pressure generation source is judged to be human, the central point of the biped outlines is marked as a social distance monitoring point, when the distance between any two social distance monitoring points is smaller than a set distance warning threshold value, the social distance is determined to be too small, a main control circuit board (31) controls a service robot (3) to move to one side of a place where the social distance is too small, and a social distance reminding module is started to remind that the social distance is kept.
2. The YOLO algorithm based social distance monitoring robot of claim 1, wherein: the array type pressure sensing film component (1) is a piezoresistive distributed flexible film.
3. The YOLO algorithm-based social distance monitoring robot of claim 1, wherein: social distance reminds module includes electronic cloud platform (4) and projection arrangement (5), electronic cloud platform (4) fixed mounting is at the top of service robot (3), projection arrangement (5) are installed on the rotation portion of electronic cloud platform (4), and the orientation of electronic cloud platform (4) rotation portion is controlled in main control circuit board (31), projection arrangement (5) throw out the pattern of setting for shape and colour on ground, the shape of pattern is straight line, curve or rectangular, the colour of pattern is green, yellow or red.
4. The yo algorithm-based social distance monitoring robot of claim 3, wherein: the projection device (5) is a laser projection lamp, and the patterns further comprise text reminding information.
5. The YOLO algorithm based social distance monitoring robot of claim 3, wherein: the social distance reminding module further comprises a voice reminding sound box (6).
6. The yo algorithm-based social distance monitoring robot of claim 5, wherein: the voice reminding sound box (6) is a voice sound box of the service robot (3).
7. The YOLO algorithm-based social distance monitoring robot of claim 1, wherein: the pressure sensing film assembly is characterized by further comprising an interval power supply module, wherein the interval power supply module comprises a power adapter (71) and a time control switch (72), the power adapter (71) is connected to a mains supply and is electrically connected with the array type pressure sensing film assembly (1) through the time control switch (72).
8. The YOLO algorithm-based social distance monitoring robot of claim 1, wherein: the wireless transmission module (2) is a wireless WIFI module.
9. The YOLO algorithm-based social distance monitoring robot according to any one of claims 1-8, wherein the method for determining the social distance specifically comprises:
step 1, a time control switch (72) controls an array type pressure sensing membrane assembly (1) to be electrified for 10 seconds at a time interval of 30-60 seconds;
step 2, in a power-on state, the array type pressure sensing film assembly (1) wirelessly transmits a contour signal and a weight value of a pressure generation source above a monitored floor to a main control circuit board (31) through a wireless transmission module (2), and a processor of the main control circuit board (31) converts a pressure signal contour of which the weight value m is more than 20kg and which is monitored in the power-on 10 th second into an image A;
step 3, adopting a YOLOV5 algorithm to perform target detection and identification on different foot types of the pressure signal profile image, judging a paired foot type area with the identification similarity IOU value larger than a set similarity threshold value as a human biped standing state, performing maximum profile frame selection on the paired human biped standing state area, marking the selected area as Fn, wherein n represents the number of the identified human biped standing state images, marking the central point Cn of the Fn, and obtaining the position coordinate Sn of the Cn in the images;
step 4, calculating the space S, S = |, between any two adjacent human bipedal standing state images
| setting a distance threshold value X, and when S is smaller than X, indicating that the human social distance corresponding to the image position of the human biped standing state is too small;
and step 5, the main control circuit board (31) controls the service robot (3) to carry out path planning, the service robot moves to one side of a connecting line of central points of two adjacent human biped standing state images with too small social distance, the projection device (5) is started to project patterns to the ground, meanwhile, the voice reminding sound (6) starts voice reminding until the social distance S is increased by a person to be reminded, and when S is larger than 1.2X, the main control circuit board (31) controls the service robot (3) to close the projection device (5) and the voice reminding sound (6) and leave.
10. The YOLO algorithm-based social distance monitoring robot of claim 9, wherein: in step 3, the similarity IOU value is 0.9-0.95; in step 4, the distance threshold X is 0.5-1.5 meters.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211653065.7A CN115625723B (en) | 2022-12-22 | 2022-12-22 | Social distance monitoring robot based on YOLO algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211653065.7A CN115625723B (en) | 2022-12-22 | 2022-12-22 | Social distance monitoring robot based on YOLO algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115625723A CN115625723A (en) | 2023-01-20 |
| CN115625723B true CN115625723B (en) | 2023-03-10 |
Family
ID=84910358
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211653065.7A Active CN115625723B (en) | 2022-12-22 | 2022-12-22 | Social distance monitoring robot based on YOLO algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115625723B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2017033165A (en) * | 2015-07-30 | 2017-02-09 | 株式会社日立製作所 | Residence condition measuring apparatus, residence condition measuring system, and residence condition measuring method |
| CN106846909A (en) * | 2017-02-24 | 2017-06-13 | 青岛智慧城市产业发展有限公司 | Intelligent road anticollision accessory system |
| CN107066917A (en) * | 2015-12-03 | 2017-08-18 | 国际联合航空集团 | Queue monitor based on marking profile |
| CN108431880A (en) * | 2015-10-30 | 2018-08-21 | 光学感应器控股有限公司 | Monitor the magnitude of traffic flow |
| CN108898726A (en) * | 2018-07-22 | 2018-11-27 | 薛永芳 | A kind of uniform queuing guide device for being lined up troop |
| CN110363695A (en) * | 2019-07-01 | 2019-10-22 | 深圳勇艺达机器人有限公司 | A kind of crowd's queue control method and device based on robot |
| JP2022143869A (en) * | 2021-03-18 | 2022-10-03 | 株式会社Ｊｖｃケンウッド | Determination device and determination method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN211196192U (en) * | 2019-11-07 | 2020-08-07 | 中国计量大学 | Fresh air control system intelligently adjusted along with saturation of carrying passengers in subway carriage |
-
2022
- 2022-12-22 CN CN202211653065.7A patent/CN115625723B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2017033165A (en) * | 2015-07-30 | 2017-02-09 | 株式会社日立製作所 | Residence condition measuring apparatus, residence condition measuring system, and residence condition measuring method |
| CN108431880A (en) * | 2015-10-30 | 2018-08-21 | 光学感应器控股有限公司 | Monitor the magnitude of traffic flow |
| CN107066917A (en) * | 2015-12-03 | 2017-08-18 | 国际联合航空集团 | Queue monitor based on marking profile |
| CN106846909A (en) * | 2017-02-24 | 2017-06-13 | 青岛智慧城市产业发展有限公司 | Intelligent road anticollision accessory system |
| CN108898726A (en) * | 2018-07-22 | 2018-11-27 | 薛永芳 | A kind of uniform queuing guide device for being lined up troop |
| CN110363695A (en) * | 2019-07-01 | 2019-10-22 | 深圳勇艺达机器人有限公司 | A kind of crowd's queue control method and device based on robot |
| JP2022143869A (en) * | 2021-03-18 | 2022-10-03 | 株式会社Ｊｖｃケンウッド | Determination device and determination method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115625723A (en) | 2023-01-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110142785A (en) | A kind of crusing robot visual servo method based on target detection | |
| WO2003092291A1 (en) | Object detection device, object detection server, and object detection method | |
| CN105631961B (en) | It is a kind of based on far infrared to the racing track timekeeping system and its application method of pipe | |
| CN111685633A (en) | Tumble detection method | |
| CN207602041U (en) | Vehicle, indoor and outdoor alignment system are sought in wisdom parking | |
| CN109389641A (en) | Indoor map integrated data generation method and indoor method for relocating | |
| CN105760832A (en) | Escaped prisoner recognition method based on Kinect sensor | |
| KR20190094954A (en) | Apparatus and method for tracking a movement of eletronic device | |
| CN108693532A (en) | Wearable barrier-avoiding method and device based on enhanced binocular camera Yu 3D millimetre-wave radars | |
| CN213131856U (en) | Long jump detection device based on OpenCV image-text recognition technology | |
| CN109059929A (en) | Air navigation aid, device, wearable device and storage medium | |
| TWI774904B (en) | Pet monitoring method and pet monitoring system | |
| US11494935B2 (en) | Scaled human video tracking | |
| CN107300398A (en) | A kind of communicated based on WIFI supports the electric inspection process device of positioning and data transfer simultaneously | |
| CN110975270A (en) | Standing long jump detection method based on marks and computer vision | |
| CN112613381A (en) | Image mapping method and device, storage medium and electronic device | |
| CN115625723B (en) | Social distance monitoring robot based on YOLO algorithm | |
| CN110170993A (en) | A kind of network-enabled intelligent robot | |
| CN109512435A (en) | It is a kind of can real-time detection walk appearance correcting instrument | |
| CN109145697A (en) | A kind of method of voice calling intelligent home Kang Hu robot | |
| CN111862170A (en) | Optical motion capture system and method | |
| CN112582075A (en) | Epidemic situation prevention and control intelligent system based on face recognition algorithm and detection method thereof | |
| CN209916004U (en) | Optical and inertial hybrid motion capture system | |
| CN106595647B (en) | Behavior state monitoring method, device and system | |
| WO2023009935A3 (en) | Survey device, system and method |
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 |