CN103413321A - Crowd behavior model analysis and abnormal behavior detection method under geographical environment - Google Patents
Crowd behavior model analysis and abnormal behavior detection method under geographical environment Download PDFInfo
- Publication number
- CN103413321A CN103413321A CN2013102980847A CN201310298084A CN103413321A CN 103413321 A CN103413321 A CN 103413321A CN 2013102980847 A CN2013102980847 A CN 2013102980847A CN 201310298084 A CN201310298084 A CN 201310298084A CN 103413321 A CN103413321 A CN 103413321A
- Authority
- CN
- China
- Prior art keywords
- crowd
- movement
- under
- sports ground
- group movement
- 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.)
- Pending
Links
Images
Abstract
The invention discloses a crowd behavior model analysis and abnormal behavior detection method under a geographical environment. The method comprises the following steps that: video monitoring signals can be captured, and crowd movement regions in a monitoring scene are set, and video monitoring crowd images are obtained, and geographic spatial mapping processing is performed on the crowd movement regions; measurable crowd movement fields are calculated through using an optical flow method under geographic reference, and the crowd movement fields are converted and mapped to polar coordinate reference; according to the distribution situation of the crowd movement fields under the polar coordinate reference, statistical analysis is performed on the crowd movement fields at each main direction of the polar coordinate reference, and then, crowd movement models and crowd movement trends under the geographical environment can be judged, and crowd movement rate at each main direction can be estimated; and based on the analysis results of the crowd movement models, movement trends and movement rate, detection on crowd abnormal behaviors such as movement rate mutation, movement trend mutation, reverse walking, sudden aggregation and sudden scatter is performed. The crowd behavior model analysis and abnormal behavior detection method of the invention can be widely used in areas where crowds are prone to aggregation.
Description
Technical field
The present invention relates to behavior mode of population analysis and anomaly detection method under a kind of geographical environment, specifically, be that a kind of video data that utilizes is analyzed crowd's motion state under geographical environment, and can under geographical environment, detect the method for Reference Group's abnormal behaviour.
Background technology
Along with socioeconomic fast development, the large-scale crowds such as recreation, exhibition activity, competitive sports are assembled frequent activity and are occurred, the crowd too assembles and often causes the crowded accident such as trample.Video itself has space-time concurrently, it is directly perceived to express, abundant information, dynamic characteristics such as real-time, and in recent years, monitoring probe has spread all over each corner in city, and utilizing video monitoring to carry out group behavior understanding has become study hotspot.Therefore, utilize video data to monitor in real time crowd's behavior pattern, can better hold crowd's flow development situation, for security protection department effectively the management crowd scientific basis is provided.
Group behavior understanding refers to by population analysis, crowd's motor pattern and rule be analyzed and identification, has become by the study hotspot of extensive concern in recent years.Crowd behaviour is understood study general and is followed Motion feature extraction and basic procedure (the Video understanding framework for automatic behavior recognition such as description, behavior identification, high-rise behavior and scene understanding; Behavior Research Methods Journal; 2006,38 (3): 416-426; Based on the automatic identification framework of the behavior of video, behavioral study method magazine, 2006, the 38th the 3rd phase of volume, 416-426).Motion feature extraction is dynamic object is detected, on the basis of classification and tracking with describing, and utilizes the correlated characteristic of image to describe the motion feature information of target; Behavior identification is to utilize image sequence to extract the motion feature of target, and its feature with reference image sequence is mated, according to the behavior pattern of matching result analysis dynamic object; It is that the relevant knowledge of behavior pattern is combined with scene information that high-rise behavior is understood with scene, judgement crowd's complex behavior pattern, thus realize the understanding to time and scene.For the crowd under specific environment, usually utilize the information such as main direction, speed, abnormal motion to detect crowd's abnormal behaviour.In recent years, Chinese scholars has proposed a lot of methods for population analysis and understanding, generally speaking, it can be divided into based on crowd's ontoanalysis and two kinds of methods of crowd's holistic approach.
Based on the analytical approach of crowd's individuality be by cut apart or the detection crowd in individuality, and the motor pattern between individuality is analyzed to the understanding realized crowd behaviour.If exist certain pedestrian's direction of motion opposite with crowd's motion principal direction, can judge and have potential danger.The propositions such as Bobick utilize template matching method identification human motion, at first template matching method carries out feature extraction to input image sequence, and the template of the feature of extraction and training stage pre-save is carried out to similarity relatively, will be with cycle tests apart from classification under minimum template, recognition result (The recognition of human movement using temporal templates as tested sequence, IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001,23 (3): 257-267; The human motion identification of time-based template, pattern analysis and machine intelligence (IEEE Transactions), calendar year 2001, the 23rd the 3rd phase of volume, 257-267).Jacques etc. have proposed a kind of computer vision technique active/passive formula crowd detection and classification algorithm that utilizes; adopt Voronoi figure to carry out crowd's volume tracing to the monitoring scene of overlooking video camera; this sociology concept of quantitative description individual space (Understanding people motion in video sequences using voronoi diagrams, Pattern Analysis& Applications, 2007,10 (4): 321-332; In sequence video, utilize the human Motion Understanding of Voronoi figure, pattern analysis and application, 2007, the 10th the 4th phase of volume, 321-332).Cheriyadat etc. utilize optic flow technique to extract the crowd's sports ground in scene, by cluster analysis, movement locus and crowd's main body direction of motion have been excavated, and realized with the inconsistent abnormal behaviour of main body direction of motion, detecting (Detecting dominant motions in dense crowds, IEEE Journal of Selected Topics in Signal Processing, 2008,2 (4): 568-581; Dense crowd's main body motion detection, it is selected that signal is processed the IEEE magazine, and 2008, the 2nd the 4th phase of volume, 568-581).Wang etc. have proposed, based on crowd behaviour in the complex scene of unsupervised learning and interaction modeling method, to can be used for detecting the abnormal behaviour in monitoring scene, can be dissimilar behavior state etc. by crowd's motion segmentation.Based on individual analytical approach, be only applicable to the low density crowd scene; for the higher monitoring scene of crowd density; owing to blocking and the phenomenon such as overlapping; employing based on the ontoanalysis method can't realize to the analysis of crowd behaviour with understand (Unsupervised activity perception in crowded and complicated scenes using hierarchical Bayesian models; IEEE Transactions on Pattern Analysis and Machine Intelligence; 2009,31 (3): 539-555; Non-supervisory behavior based on the level Bayesian model under crowded complex scene is identified, pattern analysis and machine intelligence (IEEETransactions), and 2009, the 31st the 3rd phase of volume, 539-555).
Based on the analytical approach of crowd's integral body, be crowd in scene to be done as a whole, analyze and the behavior pattern of understanding the crowd from whole angle.These class methods, without the individuality of cutting apart in the crowd, are suitable for the Dense crowd of crowded complexity.Davies etc. combine discrete cosine transform with linear transformation; judgement crowd's Stillness and motion; and by the moving characteristic of pixel or image block, crowd's overall movement speed (comprising direction and size) (Crowd monitoring using image processing, Electronics& are described; Communication Engineering Journal, 1995,7 (1): 37-47; Based on the population surveillance that image is processed, electronics and communication engineering magazine, nineteen ninety-five, the 7th the 1st phase of volume, 37-47).Boghossian and Velastin adopt the block matching motion estimation technique, crowd's track in video monitoring and motion general direction are estimated, and detect abnormal behaviour (the Motion-based machine vision technique for the management of large crowds in monitoring scene by crowd's flow trace and direction, IEEE Conference on Electronics, Circuits and Systems, Pafos, Cyprus, 5-8Semptember1999; The crowd of based on motion machine vision technique management, electronics in 1999, Circuits and Systems IEEE International Academic Conference collection of thesis, Paphos, 5-8 day in September, 1999).Ali and Shah have proposed a kind of based on the dynamic (dynamical) crowd behaviour analytical approach of Lagrangian particle, by crowd's optical flow field cut apart to detect group abnormality behavior (A lagrangian particle dynamics approach for crowd flow segmentation and stability analysis, IEEE Conference on Computer Vision and Pattern Recognition, Minnesota, 18-23June2007; A kind ofly based on the dynamic (dynamical) stream of people of Lagrangian particle, cut apart and method for analyzing stability computer vision in 2007 and pattern-recognition IEEE International Academic Conference collection of thesis, Minnesota, 18-23 day in June, 2007).Yang Lin etc. utilize Block Matching Algorithm to detect crowd's motion vector field, the motion vector proper vector is inputted to classification (a kind of design and realization of crowd's abnormal behaviour detection system, railway computer utility, 2010 that support vector machine classifier completes group behavior, the 19th the 7th phase of volume, 37-41).Zhu Hailong etc. propose the abnormal state detection of a kind of figure analysis method for the dynamic crowd scene, by the dispersion degree between analysis chart summit space distribution and limit weight matrix Dynamic System Forecast value and observed reading, anomalous event in dynamic scene is detected and locates (the figure analysis method of crowd's abnormal state detection, the robotization journal, 2012, the 38th the 5th phase of volume, 742-750).
From the existing analysis and research of group behavior based on video, mainly concentrate on the safety monitoring aspect of specific application area, as public gathering places such as subway, square, stations.To the extraction of crowd's motion feature only for image space, image coordinate with reference under analyze behavior mode of population, can only take image space as with reference to describing group behavior, can't real motion state and the behavior pattern of perception crowd under geographical environment.For group movement speed, research in the past be take the image pixel number and is dimension, if expect real movement rate size, needs the conversion scale factor between further computed image pixel and actual value.
Summary of the invention
Key issue to be solved by this invention is under geographical environment, to carry out the behavior mode of population analysis, utilize video data, under Geographic Reference, extract crowd's motion feature, crowd's motion feature is analyzed and then obtained the behavior mode of population under geographical environment, comprise group movement pattern, group movement trend, group movement speed.On this basis, detect the sudden change of group movement trend, the sudden change of group movement speed, reverse walking, poly-and abnormal behaviour that suddenly fall apart suddenly.Therefore, the present invention proposes a kind of video data that utilizes, under geographical environment, carry out the method for behavior mode of population analysis the behavior of detection group abnormality.
Basic ideas of the present invention: the geographical space mapping method that utilizes video data by the video data unification to Geographic Reference; Under Geographic Reference, utilize optical flow method to ask and calculate measurable crowd's sports ground, the conversion of crowd's sports ground is mapped to the polar coordinates reference; According to the distribution situation of crowd's sports ground under the polar coordinates reference, in the statistical study of polar coordinates with reference to the enterprising pedestrian group's sports ground of each main direction, and then the group movement pattern under the judgement geographical environment, group movement trend, and estimate the group movement speed of each main direction; On this basis, carry out movement rate sudden change, movement tendency sudden change, reverse walking, poly-and loose group abnormality behavior detection suddenly suddenly.
The basic step of the behavior mode of population analytical approach under geographical environment of the present invention is:
The first step, capturing video pilot signal, set the crowd activity zone in monitoring scene, obtains video monitoring crowd image, and the geographical space mapping is carried out in the crowd activity zone and process;
Second step, under Geographic Reference, utilize optical flow method to calculate crowd's sports ground, obtain having mensurable crowd's sports ground of Geographic Reference;
The 3rd step, based on measurable crowd's sports ground under Geographic Reference, its conversion is mapped under the polar coordinates reference, according to crowd's motion vector polar coordinates with reference under distribution, analyze group movement pattern, group movement trend under geographical environment, and the group movement speed of all directions under the quantitative estimation geographical environment, detect on this basis the sudden change of group movement speed, the sudden change of group movement trend, reverse walking, poly-and rapid loose group abnormality behavior suddenly.
The geographical space mapping concrete steps of the described video monitoring image of the first step are:
(1) choose the crowd activity zone in image;
(2) utilize two vanishing point perspective models (Luo Xiaohui, based on the image perspective transform method of two vanishing points, computer engineering, 2009,35 (15): 212-214.) image-region of choosing is carried out to perspective correction;
(3) choose three groups of image coordinate and corresponding geographic coordinates thereof after above perspective correction, according to the corresponding relation of coordinate, ask the mapping transformation matrix of nomogram image space to geographical space, complete the geographical space mapping of monitoring crowd image.
The described concrete construction step of optical flow method calculating crowd's sports ground that utilizes of second step is:
(1) Real-time Obtaining population surveillance image;
(2) utilize Lucas-Kanade(LK) optical flow algorithm (An iterative image registration technique with an application to stereo vision; Proceedings of the1981DARPA Imaging Understanding Workshop; Washington, 18-21April1981; A kind of iterative image registration technology and the application in stereoscopic vision thereof, ARPA's image understanding symposium collection of thesis in 1981, Washington, 18-21 day in April, 1981) crowd's sports ground in computed image space;
(3) utilize in the first step described (3) and ask the mapping transformation matrix of the image space of calculation to geographical space, crowd's sports ground Mapping and Converting of image space, to geographical space, is obtained to measurable crowd's sports ground under Geographic Reference.
The described group movement pattern of the 3rd step comprises that one-way movement, bidirectional-movement, center are gathered, surrounding is dispersed and careless and sloppy unordered five kinds.Concrete steps to the group movement pattern analysis are:
(1) by the mensurable crowd's sports ground of the described geographical space of second step transformed mappings coordinate reference extremely;
(2) according to the distribution of crowd's sports ground under the polar coordinates reference, the judgement crowd belongs to certain group movement pattern of certain direction.
The concrete steps of the described group movement trend analysis of the 3rd step are:
(1) based on the mensurable crowd's sports ground under Geographic Reference, it is mapped to the polar coordinates reference frame;
(2) as required polar coordinate system is divided into to some main directions, according to the main direction standard that polar coordinate system delimited, calculates the affiliated main direction of each motion vector;
(3) according to wind rose map principle and method, generate crowd's movement tendency rose diagram, obtain the cumulative frequency of each main direction crowd motion vector;
(4) the crowd's sports ground rose diagram that utilizes statistical study to obtain, can judge the crowd movement tendency of monitoring scene under Geographic Reference.
The concrete steps of the described group movement rate estimation of the 3rd step are:
(1) Real-time Obtaining processing video data, extract measurable crowd's sports ground under Geographic Reference;
(2) as required the polar coordinates reference is divided into to some main directions;
(3) measurable crowd's sports ground under Geographic Reference is converted to the polar coordinates reference, judges the direction under each motion vector according to the main direction of setting under polar coordinate system;
(4) the motion vector displacement of all directions is accumulated to calculating, and add up the motion vector number that all directions scope has, ask the average of calculating the displacement of all directions motion vector;
In the time interval of (5) calculating according to crowd's sports ground, carry out all directions crowd movement rate and solve.
The described group abnormality behavior of the 3rd step detects, and mainly comprises the sudden change of group movement trend, the sudden change of group movement speed, reverse walking, suddenly gathers and the 5 kind of groups abnormal behaviours of suddenly faling apart, and concrete steps are:
(1) utilize the described group movement pattern of the 3rd step, group movement trend and group movement rate estimation concrete steps, Real-Time Monitoring group movement pattern, group movement trend, group movement speed;
(2) as required the group movement trend of monitoring scene all directions is changed with movement rate and changes setting threshold;
(3), when the group movement trend monitored, group movement rate variation surpass setting threshold, judge that movement tendency sudden change, movement rate Sudden Anomalies occur;
(4) in regulation can only the monitoring scene of unidirectional walking, if, when the movement tendency probability distribution appears in the reverse direction of crowd's motion, judge that reverse walking occurs extremely;
(5) based on crowd's sports ground of geographical space, the divergence of calculating sports ground in real time (thanks to arboriculture, vector analysis and field theory (the 1st edition), Beijing: Higher Education Publishing House, 2005,21-80) distribute, utilize space distribution judgement crowd motion rapid of divergence in the crowd activity zone to gather with rapid loose abnormal.
The present invention relates to behavior mode of population analysis and anomaly detection method under a kind of geographical environment.The user can utilize the behavior mode of population of CCTV camera analysis crowd motion under geographical environment, and detects the group abnormality behavior.Can be widely used in the population surveillance of the easy aggregation zones of crowd such as square, subway, tourist attraction, shopping mall, market potential is huge.
The accompanying drawing explanation
Fig. 1 is general technical schematic flow sheet of the present invention;
Fig. 2 is the geographical space mapping method of video data;
Fig. 3 is mensurable crowd's sports ground computing method schematic flow sheet under Geographic Reference of the present invention;
Fig. 4 is behavior mode of population analytical approach schematic flow sheet under geographical environment of the present invention;
Fig. 5 is group abnormality behavior detection method schematic flow sheet under geographical environment of the present invention.
Embodiment
Below in conjunction with drawings and Examples, the present invention is described in further detail.
The first step, relevant device are prepared.Prepare a ThinkPad X201i type portable notebook computer, first of modern HYC-S200 Multifunctional camera, one of Rikaline GPS-6033 type bluetooth gps satellite receiver.
Second step, the mapping of video data geographical space.
Fig. 2 has described the geometric relationship between video data and spatial data.Figure mid point C is camera position, and the image mapped that video camera is taken is to the picture planar I, and plane T is the image after perspective correction, and G is GIS(Geographic Information System) the Geographic Reference plane in space.Any point P (x in geographical space (G plane)
g, y
g) as Zhong De position, plane, be p (u, v), the position after the image perspective correction in the T of plane is P
t(x
t, y
t), the video data spatial mappings is sets up some p and the transformation relation of putting P, realizes the mapping of image space I to geographical space G.
(1) camera is placed on to square or the crossing of crowd massing, be used to monitoring certain regional crowd, and is connected to notebook computer;
(2) open the video monitoring program based on Aforge.NET storehouse (a kind of computer vision storehouse of increasing income) exploitation, be used to obtaining the monitoring image of each camera, described to the monitoring image perspective correction according to Fig. 1;
(3) in the monitoring image of camera, choose 4 reference mark, utilize the GPS receiver to obtain its corresponding geographic coordinate, ask and calculate the regional homograph matrix to geographical space of crowd activity in the camera head monitor image.
The 3rd step, the mensurable crowd's sports ground of geographical space calculate, and specifically see accompanying drawing 3.
(1) Real-time Obtaining video data, carry out the geographical space mapping to crowd activity zone in monitoring image;
(2) utilize LK optical flow method real-time calculating crowd motion optical flow field under Geographic Reference, obtain measurable crowd's sports ground under Geographic Reference.
Behavior mode of population analysis under the 4th step, geographical environment, specifically be shown in accompanying drawing 4.
(1) by the polar coordinates reference frame from 11.25 ° of east by north, by the interval with 22.5 ° counterclockwise, coordinate system is divided into to 16 parts;
(2) the mensurable crowd's sports ground of geographical space the 3rd step calculated is converted to the polar coordinates reference;
(3) according to the distribution situation of crowd's sports ground under utmost point reference, crowd's group movement pattern in the judgement scene;
(4), according to the main direction of delimiting in the 4th step described (1), judge the affiliated main direction of each motion vector;
(5) add up the motion vector quantity that each main direction has, calculate the ratio that all directions motion vector quantity accounts for the amount of movement sum, obtain the probability distribution of all directions motion vector, and then obtain the group movement trend of crowd in monitoring scene;
(6) calculate the accumulated value of all directions crowd motion vector mould, the statistics all directions have the quantity of crowd's motion vector, ask the average of calculating all directions crowd motion vector, obtain crowd's movement rate of all directions.
Group abnormality behavior under the 5th step, geographical environment detects, and specifically sees accompanying drawing 5.
(1) set in this monitoring scene the threshold value that the sudden change of group movement trend and group movement speed Sudden Anomalies occur;
(2) Real-Time Monitoring all directions group movement trend probability, this monitoring scene is one-way movement if stipulate, when reciprocal movement tendency is greater than 0, can judge that the reverse walking of generation is abnormal;
(3), if the regulation monitoring scene is non-one-way movement, when the movement tendency probability variation of certain direction is greater than setting threshold, can judges and move the trend sudden change;
Group movement speed Sudden Anomalies, when the movement rate variation is greater than setting threshold, occur in (4) movement rate of Real-Time Monitoring all directions;
(5) calculate in real time the divergence space distribution of mensurable crowd's sports ground, according to the threshold value of setting, monitoring is poly-, loose abnormal generation suddenly suddenly.
Claims (3)
1. behavior mode of population analysis and the anomaly detection method under geographical environment, the steps include:
The first step, capturing video pilot signal, set the crowd activity zone in monitoring scene, obtains video monitoring crowd image, and the geographical space mapping is carried out in the crowd activity zone and process;
Second step, under Geographic Reference, utilize optical flow method to calculate crowd's sports ground, obtain having mensurable crowd's sports ground of Geographic Reference;
The 3rd step, based on measurable crowd's sports ground under Geographic Reference, its conversion is mapped under the polar coordinates reference, according to crowd's motion vector polar coordinates with reference under distribution, analyze group movement pattern, group movement trend under geographical environment, and the group movement speed of all directions under the quantitative estimation geographical environment, detect on this basis the sudden change of group movement speed, the sudden change of group movement trend, reverse walking, poly-and rapid loose group abnormality behavior suddenly.
2. behavior mode of population analysis and the anomaly detection method under geographical environment according to claim 1, is characterized in that, the concrete steps that the described first step is carried out geographical space mapping processing are:
(1) choose the crowd activity zone in image;
(2) utilize two vanishing point perspective models to carry out perspective correction to the image-region of choosing;
(3) choose three groups of image coordinate and corresponding geographic coordinates thereof after above perspective correction, according to the corresponding relation of coordinate, ask the mapping transformation matrix of nomogram image space to geographical space, complete the geographical space mapping of monitoring crowd image.
3. behavior mode of population analysis and the anomaly detection method under geographical environment according to claim 1 and 2, is characterized in that, the concrete steps of described the 3rd step are respectively:
(a) analyze the group movement pattern: crowd's sports ground of geographical space is converted to the polar coordinates reference, by analyzing crowd's motion vector field distribution characteristics of polar coordinate space, which kind of group movement pattern the judgement crowd belongs to, and the group movement pattern comprises that one-way movement, bidirectional-movement, center are gathered, surrounding is dispersed and the undiscipline is unordered;
(b) analyze group movement trend: the mensurable crowd's sports ground based under Geographic Reference maps to the polar coordinates reference frame by it; According to the main direction standard of delimiting in polar coordinate system, calculate the affiliated main direction of each motion vector, and generate crowd's movement tendency rose diagram according to wind rose map principle and method, obtain crowd's motion vector cumulative frequency of each main direction; The crowd's sports ground rose diagram that utilizes statistical study to obtain, can judge the crowd movement tendency of this monitoring scene under Geographic Reference;
(c) estimation group movement speed: real-time processing video data, under Geographic Reference, extract measurable crowd's sports ground; Crowd's sports ground under Geographic Reference is converted to the polar coordinates reference, judges the direction under each motion vector according to the main direction of setting under polar coordinate system; Calculating is accumulated in motion vector displacement to all directions, and adds up the motion vector number that all directions scope has, and asks the average of calculating the displacement of all directions motion vector; According to the time interval that crowd's sports ground calculates, carry out all directions crowd movement rate and solve;
(d) the group abnormality behavior detects: according to above-mentioned (a) and (b), (c) described method, Real-Time Monitoring group movement pattern, group movement trend and group movement speed; Utilize the Real-Time Monitoring result, when the acceleration change of crowd's motion is greater than setting threshold, judges and move the speed Sudden Anomalies; At the guarded region that is defined as the one-way movement pattern, if detect, there is rightabout crowd's motion, judge that reverse walking abnormal behaviour occurs; Based on crowd's sports ground of geographical space, the divergence of calculating in real time sports ground distributes, and utilizes space distribution judgement crowd motion rapid of divergence in the crowd activity zone to gather with rapid loose abnormal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013102980847A CN103413321A (en) | 2013-07-16 | 2013-07-16 | Crowd behavior model analysis and abnormal behavior detection method under geographical environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013102980847A CN103413321A (en) | 2013-07-16 | 2013-07-16 | Crowd behavior model analysis and abnormal behavior detection method under geographical environment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103413321A true CN103413321A (en) | 2013-11-27 |
Family
ID=49606326
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013102980847A Pending CN103413321A (en) | 2013-07-16 | 2013-07-16 | Crowd behavior model analysis and abnormal behavior detection method under geographical environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103413321A (en) |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103869367A (en) * | 2014-03-25 | 2014-06-18 | 中南大学 | Seismic event time and space gathering mode extraction method based on shared density |
| CN105913008A (en) * | 2016-04-07 | 2016-08-31 | 杭州电子科技大学 | Crowd exceptional event detection method based on hypothesis examination |
| CN106157325A (en) * | 2015-04-07 | 2016-11-23 | 中国科学院深圳先进技术研究院 | Group abnormality behavioral value method and system |
| CN106157326A (en) * | 2015-04-07 | 2016-11-23 | 中国科学院深圳先进技术研究院 | Group abnormality behavioral value method and system |
| CN106813677A (en) * | 2016-12-28 | 2017-06-09 | 华中科技大学 | It is a kind of can many multi-dimensional movement data measurement units of concurrent working |
| CN107330414A (en) * | 2017-07-07 | 2017-11-07 | 郑州轻工业学院 | Act of violence monitoring method |
| CN108230305A (en) * | 2017-12-27 | 2018-06-29 | 浙江新再灵科技股份有限公司 | Method based on the detection of video analysis staircase abnormal operating condition |
| CN108537204A (en) * | 2018-04-20 | 2018-09-14 | 广州林邦信息科技有限公司 | Mankind's activity monitoring method, device and server |
| CN109063578A (en) * | 2018-07-06 | 2018-12-21 | 江西洪都航空工业集团有限责任公司 | A kind of crowd massing detection method |
| CN109087478A (en) * | 2018-08-22 | 2018-12-25 | 徐自远 | A kind of early warning of the anti-swarm and jostlement of intelligence and method of river diversion and system |
| CN109697438A (en) * | 2018-03-08 | 2019-04-30 | 中国科学院大学 | A kind of specific group's Assembling Behavior early detection and aggregation ground prediction technique and system |
| CN110009866A (en) * | 2019-04-03 | 2019-07-12 | 天津中航亿达科技有限公司 | A kind of method of video detection temperature anomaly |
| CN110427855A (en) * | 2019-07-25 | 2019-11-08 | 浙江树人学院（浙江树人大学） | A kind of algorithm that the identification of crowd's abnormal behaviour is positioned using accumulative optical flow |
| CN110826726A (en) * | 2019-11-08 | 2020-02-21 | 腾讯科技（深圳）有限公司 | Object processing method, object processing apparatus, object processing device, and medium |
| CN111368650A (en) * | 2020-02-18 | 2020-07-03 | 北京交通大学 | Crowd crowding behavior identification method based on track data |
| CN111611938A (en) * | 2020-05-22 | 2020-09-01 | 浙江大华技术股份有限公司 | Method and device for determining reverse direction |
| CN111753651A (en) * | 2020-05-14 | 2020-10-09 | 南京熊猫电子股份有限公司 | Subway group abnormal behavior detection method based on station two-dimensional crowd density analysis |
| CN111931713A (en) * | 2020-09-21 | 2020-11-13 | 成都睿沿科技有限公司 | Abnormal behavior detection method and device, electronic equipment and storage medium |
| CN112215083A (en) * | 2020-09-17 | 2021-01-12 | 中国科学院沈阳应用生态研究所 | Multi-geographic-video self-adaptive event detection method based on abnormal change modeling |
| CN112446122A (en) * | 2019-08-27 | 2021-03-05 | 北京理工大学 | Analysis method and device for motion process and storage medium |
| US10972860B2 (en) | 2017-05-25 | 2021-04-06 | International Business Machines Corporation | Responding to changes in social traffic in a geofenced area |
| CN112712548A (en) * | 2020-12-31 | 2021-04-27 | 大连海事大学 | Underwater fish swarm motion mode analysis method |
| CN112850436A (en) * | 2019-11-28 | 2021-05-28 | 宁波微科光电股份有限公司 | Pedestrian trend detection method and system of elevator intelligent light curtain |
| CN113239236A (en) * | 2021-07-13 | 2021-08-10 | 北京达佳互联信息技术有限公司 | Video processing method and device, electronic equipment and storage medium |
| CN113570546A (en) * | 2021-06-16 | 2021-10-29 | 北京农业信息技术研究中心 | Fan running state detection method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050025341A1 (en) * | 2003-06-12 | 2005-02-03 | Gonzalez-Banos Hector H. | Systems and methods for using visual hulls to determine the number of people in a crowd |
| CN101872418A (en) * | 2010-05-28 | 2010-10-27 | 电子科技大学 | Detection method based on group environment abnormal behavior |
| JP4646933B2 (en) * | 2007-03-01 | 2011-03-09 | 技研トラステム株式会社 | Congestion level discrimination system |
| CN102982341A (en) * | 2012-11-01 | 2013-03-20 | 南京师范大学 | Self-intended crowd density estimation method for camera capable of straddling |
-
2013
- 2013-07-16 CN CN2013102980847A patent/CN103413321A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050025341A1 (en) * | 2003-06-12 | 2005-02-03 | Gonzalez-Banos Hector H. | Systems and methods for using visual hulls to determine the number of people in a crowd |
| JP4646933B2 (en) * | 2007-03-01 | 2011-03-09 | 技研トラステム株式会社 | Congestion level discrimination system |
| CN101872418A (en) * | 2010-05-28 | 2010-10-27 | 电子科技大学 | Detection method based on group environment abnormal behavior |
| CN102982341A (en) * | 2012-11-01 | 2013-03-20 | 南京师范大学 | Self-intended crowd density estimation method for camera capable of straddling |
Non-Patent Citations (5)
| Title |
|---|
| HONGQUAN SONG 等: "Real-time Monitoring for Crowd Counting using Video Surveillance and GIS", 《2012 2ND INTERNATIONAL CONFERENCE ON REMOTE SENSING,ENVIRONMENT AND TRANSPORTATION ENGINEERING》 * |
| 宋宏权 等: "区域人群状态的实时感知监控", 《地球信息科学学报》 * |
| 宋宏权 等: "区域人群状态的实时感知监控", 《地球信息科学学报》, vol. 14, no. 6, 31 December 2012 (2012-12-31) * |
| 曹书凯: "人群运动的流体表达及异常行为检测", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 杨琳，苗振江: "一种人群异常行为检测系统的设计与实现", 《铁路计算机应用》 * |
Cited By (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103869367A (en) * | 2014-03-25 | 2014-06-18 | 中南大学 | Seismic event time and space gathering mode extraction method based on shared density |
| CN106157325A (en) * | 2015-04-07 | 2016-11-23 | 中国科学院深圳先进技术研究院 | Group abnormality behavioral value method and system |
| CN106157326A (en) * | 2015-04-07 | 2016-11-23 | 中国科学院深圳先进技术研究院 | Group abnormality behavioral value method and system |
| CN106157326B (en) * | 2015-04-07 | 2020-10-27 | 中国科学院深圳先进技术研究院 | Group abnormal behavior detection method and system |
| CN105913008B (en) * | 2016-04-07 | 2019-03-01 | 杭州电子科技大学 | Based on the assumption that the crowd's accident detection method examined |
| CN105913008A (en) * | 2016-04-07 | 2016-08-31 | 杭州电子科技大学 | Crowd exceptional event detection method based on hypothesis examination |
| CN106813677B (en) * | 2016-12-28 | 2018-06-29 | 华中科技大学 | It is a kind of can more concurrent workings multi-dimensional movement data measurement unit |
| CN106813677A (en) * | 2016-12-28 | 2017-06-09 | 华中科技大学 | It is a kind of can many multi-dimensional movement data measurement units of concurrent working |
| US10972860B2 (en) | 2017-05-25 | 2021-04-06 | International Business Machines Corporation | Responding to changes in social traffic in a geofenced area |
| CN107330414A (en) * | 2017-07-07 | 2017-11-07 | 郑州轻工业学院 | Act of violence monitoring method |
| CN108230305A (en) * | 2017-12-27 | 2018-06-29 | 浙江新再灵科技股份有限公司 | Method based on the detection of video analysis staircase abnormal operating condition |
| CN109697438A (en) * | 2018-03-08 | 2019-04-30 | 中国科学院大学 | A kind of specific group's Assembling Behavior early detection and aggregation ground prediction technique and system |
| CN108537204A (en) * | 2018-04-20 | 2018-09-14 | 广州林邦信息科技有限公司 | Mankind's activity monitoring method, device and server |
| CN109063578B (en) * | 2018-07-06 | 2022-05-06 | 江西洪都航空工业集团有限责任公司 | Crowd gathering detection method |
| CN109063578A (en) * | 2018-07-06 | 2018-12-21 | 江西洪都航空工业集团有限责任公司 | A kind of crowd massing detection method |
| CN109087478A (en) * | 2018-08-22 | 2018-12-25 | 徐自远 | A kind of early warning of the anti-swarm and jostlement of intelligence and method of river diversion and system |
| CN110009866A (en) * | 2019-04-03 | 2019-07-12 | 天津中航亿达科技有限公司 | A kind of method of video detection temperature anomaly |
| CN110427855A (en) * | 2019-07-25 | 2019-11-08 | 浙江树人学院（浙江树人大学） | A kind of algorithm that the identification of crowd's abnormal behaviour is positioned using accumulative optical flow |
| CN112446122A (en) * | 2019-08-27 | 2021-03-05 | 北京理工大学 | Analysis method and device for motion process and storage medium |
| CN110826726A (en) * | 2019-11-08 | 2020-02-21 | 腾讯科技（深圳）有限公司 | Object processing method, object processing apparatus, object processing device, and medium |
| CN110826726B (en) * | 2019-11-08 | 2023-09-08 | 腾讯科技（深圳）有限公司 | Target processing method, target processing device, target processing apparatus, and medium |
| CN112850436A (en) * | 2019-11-28 | 2021-05-28 | 宁波微科光电股份有限公司 | Pedestrian trend detection method and system of elevator intelligent light curtain |
| CN111368650B (en) * | 2020-02-18 | 2024-03-12 | 北京交通大学 | Crowd crowding behavior identification method based on track data |
| CN111368650A (en) * | 2020-02-18 | 2020-07-03 | 北京交通大学 | Crowd crowding behavior identification method based on track data |
| CN111753651A (en) * | 2020-05-14 | 2020-10-09 | 南京熊猫电子股份有限公司 | Subway group abnormal behavior detection method based on station two-dimensional crowd density analysis |
| CN111611938B (en) * | 2020-05-22 | 2023-08-29 | 浙江大华技术股份有限公司 | Retrograde direction determining method and device |
| CN111611938A (en) * | 2020-05-22 | 2020-09-01 | 浙江大华技术股份有限公司 | Method and device for determining reverse direction |
| CN112215083B (en) * | 2020-09-17 | 2021-11-09 | 中国科学院沈阳应用生态研究所 | Multi-geographic-video self-adaptive event detection method based on abnormal change modeling |
| CN112215083A (en) * | 2020-09-17 | 2021-01-12 | 中国科学院沈阳应用生态研究所 | Multi-geographic-video self-adaptive event detection method based on abnormal change modeling |
| CN111931713B (en) * | 2020-09-21 | 2021-01-29 | 成都睿沿科技有限公司 | Abnormal behavior detection method and device, electronic equipment and storage medium |
| CN111931713A (en) * | 2020-09-21 | 2020-11-13 | 成都睿沿科技有限公司 | Abnormal behavior detection method and device, electronic equipment and storage medium |
| CN112712548A (en) * | 2020-12-31 | 2021-04-27 | 大连海事大学 | Underwater fish swarm motion mode analysis method |
| CN112712548B (en) * | 2020-12-31 | 2023-07-07 | 大连海事大学 | Underwater fish school movement pattern analysis method |
| CN113570546A (en) * | 2021-06-16 | 2021-10-29 | 北京农业信息技术研究中心 | Fan running state detection method and device |
| CN113570546B (en) * | 2021-06-16 | 2023-12-05 | 北京农业信息技术研究中心 | Fan running state detection method and device |
| CN113239236A (en) * | 2021-07-13 | 2021-08-10 | 北京达佳互联信息技术有限公司 | Video processing method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103413321A (en) | Crowd behavior model analysis and abnormal behavior detection method under geographical environment | |
| Zhang et al. | Physics inspired methods for crowd video surveillance and analysis: a survey | |
| Xie et al. | Video crowd detection and abnormal behavior model detection based on machine learning method | |
| CN102982341B (en) | Self-intended crowd density estimation method for camera capable of straddling | |
| Seer et al. | Kinects and human kinetics: A new approach for studying pedestrian behavior | |
| Chen et al. | Fall detection system based on real-time pose estimation and SVM | |
| CN103020606A (en) | Pedestrian detection method based on spatio-temporal context information | |
| CN104159088B (en) | A kind of long-distance intelligent vehicle monitoring system and method | |
| CN103530601A (en) | Monitoring blind area crowd state deduction method based on Bayesian network | |
| Juyal et al. | Locating people in real-world for assisting crowd behaviour analysis using SSD and deep SORT algorithm | |
| Chauhan et al. | Study of moving object detection and tracking for video surveillance | |
| US11315257B2 (en) | Method for real time surface tracking in unstructured environments | |
| Yoon et al. | Tracking System for mobile user Based on CCTV | |
| Jung et al. | Real-time estimation of 3D scene geometry from a single image | |
| CN104123569A (en) | Video person number information statistics method based on supervised learning | |
| Ruget et al. | Real-time, low-cost multi-person 3D pose estimation | |
| Zhao et al. | An alphapose-based pedestrian fall detection algorithm | |
| Parvathy et al. | Anomaly detection using motion patterns computed from optical flow | |
| Wu et al. | Video-Based Fall Detection Using Human Pose and Constrained Generative Adversarial Network | |
| Lee et al. | Modeling crowd motions for abnormal activity detection | |
| Aramvith et al. | Video processing and analysis for surveillance applications | |
| Murayama et al. | Deep pedestrian density estimation for smart city monitoring | |
| Lao et al. | Human running detection: Benchmark and baseline | |
| Hao et al. | A Detection Method of Abnormal Event in Crowds Based on Image Entropy | |
| Zhang et al. | Motion Attitude Recognition and Behavior Prediction Algorithm Based on Wireless Sensor Network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20131127 |