CN112347906B - Method for detecting abnormal aggregation behavior in bus - Google Patents

Method for detecting abnormal aggregation behavior in bus Download PDF

Info

Publication number
CN112347906B
CN112347906B CN202011218749.5A CN202011218749A CN112347906B CN 112347906 B CN112347906 B CN 112347906B CN 202011218749 A CN202011218749 A CN 202011218749A CN 112347906 B CN112347906 B CN 112347906B
Authority
CN
China
Prior art keywords
bus
crowd
abnormal
carriage
aggregation
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
Application number
CN202011218749.5A
Other languages
Chinese (zh)
Other versions
CN112347906A (en
Inventor
王玉全
费玉婧楠
王力
何忠贺
刘鹏
徐龙
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
North China University of Technology
Original Assignee
North China University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by North China University of Technology filed Critical North China University of Technology
Priority to CN202011218749.5A priority Critical patent/CN112347906B/en
Publication of CN112347906A publication Critical patent/CN112347906A/en
Application granted granted Critical
Publication of CN112347906B publication Critical patent/CN112347906B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/26—Government or public services
    • 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
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B31/00—Predictive alarm systems characterised by extrapolation or other computation using updated historic data
    • 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
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

The invention relates to a method for detecting abnormal aggregation behavior in a bus, which comprises the following steps: s1: an infrared camera arranged in the bus for shooting video; s2: the method comprises the steps that an infrared camera in a bus in S1 is utilized to obtain video segments in the bus, video frame image preprocessing is carried out on the obtained video segments, and video frame images are converted into a data set; s3: and (3) carrying out affine transformation processing on the video frame image in the S2, and modeling the internal structure of the bus. The detection device for the abnormal aggregation behavior in the bus comprises a video acquisition module, a video analysis module, an abnormal judgment module and an abnormal alarm module, wherein the video acquisition module is electrically connected with the video analysis module, the video analysis module is electrically connected with the abnormal judgment module, and the abnormal judgment module is electrically connected with the abnormal alarm module. Aiming at the problem that the detection algorithm in the bus lacks consideration of the actual background of the bus, the detection algorithm actually considers the actual scene in the bus and accurately judges the abnormal aggregation behavior.

Description

Method for detecting abnormal aggregation behavior in bus
Technical Field
The invention relates to a detection method for abnormal aggregation behavior in a bus.
Background
In recent years, the country advocates green traffic through great force, so that public travel becomes the preferred mode of many citizens. As the most common green travel mode, the bus has the characteristics of high passenger flow mobility, high density, complex crowd and the like, and is a high-occurrence scene of group events. Meanwhile, the problem that the system is relatively closed and has no other safety monitoring means is solved, so that the dangerous degree of occurrence of group events is more prominent.
In traditional research, only the use of surveillance video for driver boarding and disembarking or for location tracking of criminal agents is emphasized. While abnormal behavior of passengers has not been of great concern. Due to the characteristic of strong influence of the swarm in the buses, the detection of abnormal aggregation behavior can prevent major violent events and trampling events. If the danger in the crowd is found in time, the early warning of sudden crowd events can be performed in time, and the life safety of more people is guaranteed. Therefore, the method has important practical significance in detecting and early warning abnormal aggregation behaviors in the bus.
At present, the research on the detection of abnormal behaviors in buses has certain limitations. Firstly, the traditional processing mainly relies on manpower to carry out later tracking and analysis, lacks of real-time analysis and identification of a monitoring scene, and does not have the function of providing warning for an abnormal event; secondly, the detection algorithm used in the existing buses does not consider the actual background of the buses, but mostly uses an algorithm model trained by a public action data set, and does not consider the characteristics of a closed structure in the buses, so that the actual effect cannot be well ensured.
In summary, in the aspect of practical application, the research on detecting abnormal aggregation behavior in a bus needs to actually consider the design of the method for performing the method in the actual scene in the bus.
Disclosure of Invention
Technical problems to be solved by the invention
Aiming at the technical problem that the detection algorithm in the bus in the prior art lacks consideration of the actual background of the bus, the invention provides the detection method and the device for the abnormal aggregation behavior in the bus, which actually consider the actual scene in the bus and accurately judge the abnormal aggregation behavior.
Technical proposal
In order to solve the problems, the technical scheme provided by the invention is as follows:
the method for detecting the abnormal aggregation behavior in the bus comprises the following steps:
s1: arranging infrared cameras for shooting videos at a plurality of positions in a bus;
s2: the method comprises the steps that an infrared camera in a bus in S1 is utilized to obtain video segments in the bus, video frame image preprocessing is carried out on the obtained video segments, and video frame images are converted into a data set;
s3: carrying out affine transformation processing on the video frame image in the S2, modeling the internal structure of the bus, and dividing the internal space of the bus into four spaces of the front part, the tail part, the left side and the right side of a carriage;
s4: performing head target detection algorithm training on the data set in the S2 by using a yolov5-S algorithm, and performing coding identification on heads in video frequency bands;
s5: the method comprises the steps that a bus has two motion states, namely a running state between two stations and a stop-to-stop on-off state, a time value of the two motion states of the bus is set, the running state time of the bus between the two stations is set to be t1, and the stop-to-stop on-off state time of the bus is set to be t2;
s6: according to the results of S3, S4 and S5, four motion characteristic indexes including crowd average kinetic energy, crowd motion direction entropy, crowd individual interval potential energy and individual average acceleration of different spaces in the bus at t1 time and different spaces in the bus at t2 time are calculated;
s7: setting an adjustable time threshold according to the division of the internal space of the bus and the motion state of the bus;
s8: calculating comprehensive weight a according to the four motion characteristic indexes in the step S6, and judging potential aggregation sets of abnormal aggregation;
s9: triggering prompt information to an information platform at a bus driver when the potential aggregation set with abnormal aggregation is judged, lighting a yellow lamp and carrying out voice prompt, and continuing to execute from S2 when the potential aggregation set without abnormal aggregation is judged;
s10: if the duration of the potential aggregation set of the abnormal aggregation in the S9 exceeds the adjustable time threshold set in the S7, the red light is turned on and voice prompt is carried out, and meanwhile, the abnormal video segment and the dangerous signal are uploaded to the cloud platform and the traffic safety department for real-time information transmission.
Optionally, when preprocessing each frame of image in S2, mosaicing is performed in a manner of mosaics data enhancement, random scaling, random clipping and random arrangement.
Optionally, the head code identifier in S4 includes a head code ID, location information, and time information, where the location information is obtained by using yolov5 algorithm.
Optionally, in S3, modeling is performed on bus space information by using infrared emission thermography imaging, and structural modeling of a carriage is performed by using 3dmax, and calibration is performed on the position of facilities such as a seat.
Alternatively, the bus internal space in S3 is defined as: the front door to the rear door of the bus is set to be the front part of the carriage, the rear door to the last row of seats of the bus is set to be the tail part of the carriage, the left vertical handrail of the bus is set to the left window of the left side of the carriage, and the right vertical handrail of the bus is set to the right window of the right side of the carriage.
Optionally, for the abnormal aggregation behavior determination in S8, the specific steps are as follows:
s81: extracting information data of the current head real-time position, time and head code ID from the S4 according to the time value set in the S5;
s82: calculating the distance between the current head real-time positions in S81;
s83: setting the level of the head gathering distance, and setting head gathering distance thresholds of different levels;
s84: dividing the current real-time position data of the head into a plurality of head position sets according to the comparison of the values of S82 and S83;
s85: determining the area distribution of the positions of the position sets of the heads of the people in the front part, the tail part, the left side and the right side of the carriage of the bus in S84;
s86: setting a threshold value of the number of people gathering in the head position set;
s87: screening potential people gathering collections of which the personnel number of each collection is greater than the threshold of the personnel gathering number of the S86;
s88: according to the spatial position information and the time information of the potential personnel gathering sets screened in the S87, matching thresholds under different conditions to further screen the potential personnel gathering sets;
s89: and (3) calculating the comprehensive weights of four motion characteristic indexes including the crowd average kinetic energy, the crowd motion direction entropy, the inter-individual distance potential energy and the individual average acceleration of the crowd of the person gathering set screened in the step S88 to judge the potential gathering set of abnormal gathering.
Optionally, the threshold of the number of people gathering set in S86 is defined as:
when the bus is in a driving state between two stations: the front personnel aggregation number threshold of the carriage is 5, the rear personnel aggregation number threshold of the carriage is 4, and the left and right personnel aggregation number thresholds of the carriage are 3 respectively;
when the bus is in a state of arriving at a stop to get on or off the bus: the front personnel gathering quantity threshold of the carriage is 8, the rear personnel gathering quantity threshold of the carriage is 6, and the left and right personnel gathering quantity thresholds of the carriage are 5.
Optionally, the average kinetic energy of the population in S6 or S89 is a ratio of a sum of the kinetic energy of the population to the number of the population, the sum of the kinetic energy of the population being represented by calculating optical flow energy within the region of interest;
the crowd motion direction entropy in the S6 or the S89 comprises an optical flow vector direction histogram, a direction probability distribution map and a direction entropy, and the larger the crowd motion direction entropy is, the larger the degree of confusion of crowd motion directions is;
the inter-population individual distance potential energy in S6 or S89 reflects the dispersion degree between two individuals of the population, and if the distance potential energy is suddenly increased or suddenly reduced, the possibility of occurrence of an abnormal situation is indicated;
the average acceleration of the individual in the S6 or the S89 reflects the intensity of the movement of the crowd, and the higher the numerical value is, the greater the abnormality is.
Beneficial effects of
Compared with the prior art, the technical scheme provided by the invention has the following beneficial effects:
the method and the device for detecting the abnormal aggregation behavior in the bus actually consider the actual scene in the bus and accurately judge the abnormal aggregation behavior.
Drawings
FIG. 1 is a logic flow diagram of a method for detecting abnormal aggregation behavior in a bus according to the present invention;
fig. 2 is a logic block diagram of a detection device for abnormal aggregation behavior in a bus according to the present invention.
Detailed Description
For a further understanding of the present invention, the present invention will be described in detail with reference to fig. 1 and the examples.
Referring to fig. 1, the method for detecting abnormal aggregation behavior in a bus provided by the invention comprises the following steps:
s1: arranging infrared cameras for shooting videos at a plurality of positions in a bus;
s2: the method comprises the steps that an infrared camera in a bus in S1 is utilized to obtain video segments in the bus, video frame image preprocessing is carried out on the obtained video segments, and video frame images are converted into a data set;
s3: carrying out affine transformation processing on the video frame image in the S2, modeling the internal structure of the bus, and dividing the internal space of the bus into four spaces of the front part, the tail part, the left side and the right side of a carriage;
s4: performing head target detection algorithm training on the data set in the S2 by using a yolov5-S algorithm, and performing coding identification on heads in video frequency bands;
s5: the method comprises the steps that a bus has two motion states, namely a running state between two stations and a stop-to-stop on-off state, a time value of the two motion states of the bus is set, the running state time of the bus between the two stations is set to be t1, and the stop-to-stop on-off state time of the bus is set to be t2;
s6: according to the results of S3, S4 and S5, four motion characteristic indexes including crowd average kinetic energy, crowd motion direction entropy, crowd individual interval potential energy and individual average acceleration of different spaces in the bus at t1 time and different spaces in the bus at t2 time are calculated;
s7: setting an adjustable time threshold according to the division of the internal space of the bus and the motion state of the bus;
s8: calculating comprehensive weight a according to the four motion characteristic indexes in the step S6, and judging potential aggregation sets of abnormal aggregation;
s9: triggering prompt information to an information platform at a bus driver when the potential aggregation set with abnormal aggregation is judged, lighting a yellow lamp and carrying out voice prompt, and continuing to execute from S2 when the potential aggregation set without abnormal aggregation is judged;
s10: if the duration of the potential aggregation set of the abnormal aggregation in the S9 exceeds the adjustable time threshold set in the S7, the red light is turned on and voice prompt is carried out, and meanwhile, the abnormal video segment and the dangerous signal are uploaded to the cloud platform and the traffic safety department for real-time information transmission.
The detection steps S3, S4 and S5 of the method for detecting abnormal aggregation behavior in the bus are not time-sequential. And in the step S3, modeling is carried out on bus space information by adopting infrared emission thermography imaging, structural modeling of a carriage is carried out by utilizing 3dmax, and the positions of facilities such as seats and the like are calibrated. The bus internal space in S3 is defined as: the front door to the rear door of the bus is set to be the front part of the carriage, the rear door to the last row of seats of the bus is set to be the tail part of the carriage, the left vertical handrail of the bus is set to the left window of the left side of the carriage, and the right vertical handrail of the bus is set to the right window of the right side of the carriage. And S3, carrying out affine transformation processing on the video frame data, and carrying out mapping transformation on the position of the head and the actual distance position of the carriage space structure.
The head code identification in the S4 comprises a head code ID, position information and time information, wherein the position information is obtained by utilizing a yolov5-S algorithm.
Setting an abnormal aggregation threshold in the algorithm, judging the magnitudes of the comprehensive weight a and a first aggregation threshold when judging the potential aggregation set of the abnormal aggregation of the comprehensive weight a in the step S8, and judging that the potential aggregation of the abnormal aggregation exists in the current bus if the comprehensive weight a is larger than the set abnormal aggregation threshold; if the judgment result a is smaller than the set abnormal aggregation threshold value, judging that potential aggregation of abnormal aggregation exists in the current bus.
Further, each frame of image shot by the video camera is preprocessed, so that the problems of image blurring, deformation and the like caused by environmental or shooting azimuth angle and the like of the image are solved. When each frame of image in the S2 is preprocessed, accurate identification of a small target is realized by splicing in a mode of enhancing the Mosaic data, randomly scaling, randomly cutting and randomly arranging.
Further, training the human head target detection algorithm in the step S4, and finally training to obtain a pyrach model; embedding the human head target detection algorithm into a xavier NX algorithm board, wherein the xavier NX algorithm board is installed in a bus, and model conversion is required to be carried out when the pyrach model is deployed on the xavier NX, and the conversion steps are as follows:
s41: converting the pytoch model to an ONNX model;
s42: converting the ONNX model in S41 into a tensorRT model;
s43: invoking the tensrrt model in S42 with c++;
s44: c++ in S43 is embedded into the xavier NX algorithm board.
Further, for the abnormal aggregation behavior determination in S8, the specific steps are as follows:
s81: extracting information data of the current head real-time position, time and head code ID from the S4 according to the time value set in the S5;
s82: calculating the distance between the current head real-time positions in S81;
s83: setting the level of the head gathering distance, and setting head gathering distance thresholds of different levels;
s84: dividing the current real-time position data of the head into a plurality of head position sets according to the comparison of the values of S82 and S83;
s85: determining the area distribution of the positions of the position sets of the heads of the people in the front part, the tail part, the left side and the right side of the carriage of the bus in S84;
s86: setting a threshold value of the number of people gathering in the head position set;
s87: screening potential people gathering collections of which the personnel number of each collection is greater than the threshold of the personnel gathering number of the S86;
s88: according to the spatial position information and the time information of the potential personnel gathering sets screened in the S87, matching thresholds under different conditions to further screen the potential personnel gathering sets;
s89: and (3) calculating the comprehensive weights of four motion characteristic indexes including the crowd average kinetic energy, the crowd motion direction entropy, the inter-individual distance potential energy and the individual average acceleration of the crowd of the person gathering set screened in the step S88 to judge the potential gathering set of abnormal gathering.
Further, the threshold of the number of people gathering set in S86 is defined as:
when the bus is in a driving state between two stations: the front personnel aggregation number threshold of the carriage is 5, the rear personnel aggregation number threshold of the carriage is 4, and the left and right personnel aggregation number thresholds of the carriage are 3 respectively;
when the bus is in a state of arriving at a stop to get on or off the bus: the front personnel gathering quantity threshold of the carriage is 8, the rear personnel gathering quantity threshold of the carriage is 6, and the left and right personnel gathering quantity thresholds of the carriage are 5.
Further, the crowd average kinetic energy in S6 or S89 is a ratio of a sum of crowd motion energy to the crowd quantity, wherein the sum of crowd motion energy is represented by calculating optical flow energy in the region of interest, and the crowd average kinetic energy reflects the speed and intensity of crowd motion speed. The crowd motion direction entropy in S6 or S89 includes an optical flow vector direction histogram, a direction probability distribution map, and a direction entropy, and the larger the crowd motion direction entropy is, the greater the degree of confusion of crowd motion directions is. In a bus, when the bus is in a running state between two stations, most of the motion direction entropy of the crowd is in a stable value, which indicates that the crowd moves in the same direction or does not move; when the bus is in a state of stopping to get on or off, the entropy of the crowd movement direction is in a floating value, which indicates that the crowd movement is concentrated to flow in or flow out from the front door and the rear door.
And the inter-population individual distance potential energy in S6 or S89 reflects the dispersion degree between two individuals of the population, and is calculated by combining the head position information obtained by the yolov5-S algorithm. If the distance potential energy among individuals of the crowd suddenly increases or suddenly decreases, the possibility of occurrence of abnormal conditions is indicated. The average acceleration of the individual in the S6 or the S89 reflects the intensity of the movement of the crowd, and the higher the numerical value is, the greater the abnormality is.
Referring to fig. 2, the detection device for the abnormal aggregation behavior in the bus provided by the invention comprises a video acquisition module, a video analysis module, an abnormal judgment module and an abnormal alarm module, wherein the video acquisition module is electrically connected with the video analysis module, the video analysis module is electrically connected with the abnormal judgment module, and the abnormal judgment module is electrically connected with the abnormal alarm module.
The video acquisition module is used for acquiring video data in the bus, preprocessing the video data and modeling the internal space of the bus; the video acquisition module transmits the preprocessed video data to the video analysis module, and the video analysis module carries out coding identification on the head of a person in the received video data; the video analysis module transmits the head code identification information to the abnormality judgment module, and the abnormality judgment module calculates abnormality judgment indexes, namely four motion characteristic indexes of crowd average kinetic energy, crowd motion direction entropy, crowd individual interval potential energy and individual average acceleration according to the received head identification information, calculates comprehensive weight a according to the four motion characteristic indexes, and judges potential aggregation sets of abnormal aggregation; and if the potential aggregation set is judged, the abnormality judgment module transmits an alarm signal to the abnormality alarm module, and the abnormality alarm module alarms after receiving the alarm signal.
The invention and its embodiments have been described above by way of illustration and not limitation, and the invention is illustrated in the accompanying drawings and described in the drawings in which the actual structure is not limited thereto. Therefore, if one of ordinary skill in the art is informed by this disclosure, the structural mode and the embodiments similar to the technical scheme are not creatively designed without departing from the gist of the present invention.

Claims (6)

1. The method for detecting the abnormal aggregation behavior in the bus is characterized by comprising the following steps of:
s1: arranging infrared cameras for shooting videos at a plurality of positions in a bus;
s2: the method comprises the steps that an infrared camera in a bus in S1 is utilized to obtain video segments in the bus, video frame image preprocessing is carried out on the obtained video segments, and video frame images are converted into a data set;
s3: carrying out affine transformation processing on the video frame image in the S2, modeling the internal structure of the bus, and dividing the internal space of the bus into four spaces of the front part, the tail part, the left side and the right side of a carriage;
s4: performing head target detection algorithm training on the data set in the S2 by using a yolov5-S algorithm, and performing coding identification on heads in video frequency bands;
the coding identifier in the S4 comprises a head code ID, position information and time information, wherein the position information is obtained by utilizing a yolov5-S algorithm;
s5: the method comprises the steps that a bus has two motion states, namely a running state between two stations and a stop-to-stop on-off state, a time value of the two motion states of the bus is set, the running state time of the bus between the two stations is set to be t1, and the stop-to-stop on-off state time of the bus is set to be t2;
s6: according to the results of S3, S4 and S5, four motion characteristic indexes including crowd average kinetic energy, crowd motion direction entropy, crowd individual interval potential energy and individual average acceleration of different spaces in the bus at t1 time and different spaces in the bus at t2 time are calculated;
s7: setting an adjustable time threshold according to the division of the internal space of the bus and the motion state of the bus;
s8: calculating comprehensive weight a according to the four motion characteristic indexes in the step S6, and judging potential aggregation sets of abnormal aggregation;
and for the abnormal aggregation behavior judgment in the S8, the specific steps are as follows:
s81: extracting information data of the current head real-time position, time and head code ID from the S4 according to the time value set in the S5;
s82: calculating the distance between the current head real-time positions in S81;
s83: setting the level of the head gathering distance, and setting head gathering distance thresholds of different levels;
s84: dividing the current real-time position data of the head into a plurality of head position sets according to the comparison of the values of S82 and S83;
s85: determining the area distribution of the positions of the position sets of the heads of the people in the front part, the tail part, the left side and the right side of the carriage of the bus in S84;
s86: setting a threshold value of the number of people gathering in the head position set;
s87: screening potential people gathering collections of which the personnel number of each collection is greater than the threshold of the personnel gathering number of the S86;
s88: according to the spatial position information and the time information of the potential personnel gathering sets screened in the S87, matching thresholds under different conditions to further screen the potential personnel gathering sets;
s89: calculating the comprehensive weights of four motion characteristic indexes of the crowd average kinetic energy, the crowd motion direction entropy, the inter-individual distance potential energy and the individual average acceleration of the crowd of the person gathering set screened in the S88 to perform abnormal gathering potential gathering set judgment;
s9: triggering prompt information to an information platform at a bus driver when the potential aggregation set with abnormal aggregation is judged, lighting a yellow lamp and carrying out voice prompt, and continuing to execute from S2 when the potential aggregation set without abnormal aggregation is judged;
s10: if the duration of the potential aggregation set of the abnormal aggregation in the S9 exceeds the adjustable time threshold set in the S7, the red light is turned on and voice prompt is carried out, and meanwhile, the abnormal video segment and the dangerous signal are uploaded to the cloud platform and the traffic safety department for real-time information transmission.
2. The method for detecting abnormal aggregation behavior in a bus according to claim 1, wherein when preprocessing each frame of image in S2, mosaics are performed in a manner of enhancing mosaics data, randomly scaling, randomly cutting, and randomly arranging.
3. The method for detecting abnormal aggregation behavior in a bus according to claim 1, wherein in S3, the infrared emission thermography is adopted to model the bus space information, and 3dmax is utilized to model the structure of the carriage, and the position of the seat facility is calibrated.
4. The method for detecting abnormal aggregation behavior in a bus according to claim 1, wherein for the bus internal space in S3, it is defined that: the front door to the rear door of the bus is set to be the front part of the carriage, the rear door to the last row of seats of the bus is set to be the tail part of the carriage, the left vertical handrail of the bus is set to the left window of the left side of the carriage, and the right vertical handrail of the bus is set to the right window of the right side of the carriage.
5. The method for detecting abnormal aggregation behavior in a bus according to claim 1, wherein the threshold value of the number of people aggregation set in S86 is defined as:
when the bus is in a driving state between two stations: the front personnel aggregation number threshold of the carriage is 5, the rear personnel aggregation number threshold of the carriage is 4, and the left and right personnel aggregation number thresholds of the carriage are 3 respectively;
when the bus is in a state of arriving at a stop to get on or off the bus: the front personnel gathering quantity threshold of the carriage is 8, the rear personnel gathering quantity threshold of the carriage is 6, and the left and right personnel gathering quantity thresholds of the carriage are 5.
6. The method for detecting abnormal aggregation behavior in a bus according to claim 1, wherein the crowd average kinetic energy in S6 or S89 is a ratio of a sum of crowd motion energy to a crowd number, the sum of crowd motion energy being represented by calculating optical flow energy in a region of interest;
the crowd motion direction entropy in the S6 or the S89 comprises an optical flow vector direction histogram, a direction probability distribution map and a direction entropy, and the larger the crowd motion direction entropy is, the larger the degree of confusion of crowd motion directions is;
the inter-population individual distance potential energy in S6 or S89 reflects the dispersion degree between two individuals of the population, and if the distance potential energy is suddenly increased or suddenly reduced, the possibility of occurrence of an abnormal situation is indicated;
the average acceleration of the individual in the S6 or the S89 reflects the intensity of the movement of the crowd, and the higher the numerical value is, the greater the abnormality is.
CN202011218749.5A 2020-11-04 2020-11-04 Method for detecting abnormal aggregation behavior in bus Active CN112347906B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011218749.5A CN112347906B (en) 2020-11-04 2020-11-04 Method for detecting abnormal aggregation behavior in bus

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011218749.5A CN112347906B (en) 2020-11-04 2020-11-04 Method for detecting abnormal aggregation behavior in bus

Publications (2)

Publication Number Publication Date
CN112347906A CN112347906A (en) 2021-02-09
CN112347906B true CN112347906B (en) 2023-06-27

Family

ID=74430018

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011218749.5A Active CN112347906B (en) 2020-11-04 2020-11-04 Method for detecting abnormal aggregation behavior in bus

Country Status (1)

Country Link
CN (1) CN112347906B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113420602A (en) * 2021-05-27 2021-09-21 南京四维向量科技有限公司 Atlas-based embedded human body detection edge vision computing system
CN115601714B (en) * 2022-12-16 2023-03-10 广东汇通信息科技股份有限公司 Campus violent behavior identification method based on multi-modal data analysis

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH10311402A (en) * 1997-05-07 1998-11-24 Daikin Ind Ltd Continuously variable transmission
CN106846801A (en) * 2017-02-06 2017-06-13 安徽新华博信息技术股份有限公司 A kind of region based on track of vehicle is hovered anomaly detection method
CN108257385A (en) * 2018-03-19 2018-07-06 北京工业大学 A kind of discriminating method of the anomalous event based on public transport
GB201902525D0 (en) * 2019-02-25 2019-04-10 Canon Kk Method and system for auto-setting of image acquisition and processing modules and of sharing resources in large scale video systems
WO2020092635A1 (en) * 2018-10-30 2020-05-07 Frazzoli Emilio Redundancy in autonomous vehicles
CN111325872A (en) * 2020-01-21 2020-06-23 和智信(山东)大数据科技有限公司 Driver driving abnormity detection equipment and detection method based on computer vision

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10296794B2 (en) * 2016-12-20 2019-05-21 Jayant Rtti On-demand artificial intelligence and roadway stewardship system
US11103773B2 (en) * 2018-07-27 2021-08-31 Yogesh Rathod Displaying virtual objects based on recognition of real world object and identification of real world object associated location or geofence

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH10311402A (en) * 1997-05-07 1998-11-24 Daikin Ind Ltd Continuously variable transmission
CN106846801A (en) * 2017-02-06 2017-06-13 安徽新华博信息技术股份有限公司 A kind of region based on track of vehicle is hovered anomaly detection method
CN108257385A (en) * 2018-03-19 2018-07-06 北京工业大学 A kind of discriminating method of the anomalous event based on public transport
WO2020092635A1 (en) * 2018-10-30 2020-05-07 Frazzoli Emilio Redundancy in autonomous vehicles
GB201902525D0 (en) * 2019-02-25 2019-04-10 Canon Kk Method and system for auto-setting of image acquisition and processing modules and of sharing resources in large scale video systems
CN111325872A (en) * 2020-01-21 2020-06-23 和智信(山东)大数据科技有限公司 Driver driving abnormity detection equipment and detection method based on computer vision

Also Published As

Publication number Publication date
CN112347906A (en) 2021-02-09

Similar Documents

Publication Publication Date Title
CN109670404B (en) Road ponding image detection early warning method based on hybrid model
CN105809679B (en) Mountain railway side slope rockfall detection method based on visual analysis
Jiansheng Vision-based real-time traffic accident detection
CN101303735B (en) Method for detecting moving objects in a blind spot region of a vehicle and blind spot detection device
CN112347906B (en) Method for detecting abnormal aggregation behavior in bus
CN112434564B (en) Detection system for abnormal aggregation behavior in bus
CN105799740A (en) Automatic detecting and early warning method for track foreign matter invasion based on Internet of Things technology
CN103366506A (en) Device and method for automatically monitoring telephone call behavior of driver when driving
CN103456024B (en) A kind of moving target gets over line determination methods
CN111046797A (en) Oil pipeline warning method based on personnel and vehicle behavior analysis
CN106570444A (en) On-board smart prompting method and system based on behavior identification
CN113484858A (en) Intrusion detection method and system
CN109508659A (en) A kind of face identification system and method for crossing
CN103152558B (en) Based on the intrusion detection method of scene Recognition
CN114875877A (en) Ship lockage safety detection method
CN111383248B (en) Pedestrian red light running judging method and device and electronic equipment
CN108986476A (en) Motor vehicle does not use according to regulations high beam recognition methods, system and storage medium
CN113706901B (en) Intelligent accident prevention and control and early warning system for entrance section of expressway tunnel
CN106372566A (en) Digital signage-based emergency evacuation system and method
CN113055651A (en) Artificial intelligence type vehicle security system and computer readable storage medium
Al Jarouf et al. A hybrid method to detect and verify vehicle crash with haar-like features and svm over the web
CN111932889B (en) Traffic dispersion system based on vehicle-road cooperation
CN112406700B (en) Blind area early warning system based on upper and lower binocular vision analysis range finding
CN112418000B (en) Bad driving behavior detection method and system based on monocular camera
KR102434154B1 (en) Method for tracking multi target in traffic image-monitoring-system

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