CN106157325B - Group abnormal behavior detection method and system - Google Patents
Group abnormal behavior detection method and system Download PDFInfo
- Publication number
- CN106157325B CN106157325B CN201510160419.8A CN201510160419A CN106157325B CN 106157325 B CN106157325 B CN 106157325B CN 201510160419 A CN201510160419 A CN 201510160419A CN 106157325 B CN106157325 B CN 106157325B
- Authority
- CN
- China
- Prior art keywords
- group
- tracking
- value
- cooperativity
- feature points
- 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
- 206010000117 Abnormal behaviour Diseases 0.000 title claims abstract description 37
- 238000001514 detection method Methods 0.000 title claims abstract description 22
- 238000000034 method Methods 0.000 claims abstract description 47
- 230000008569 process Effects 0.000 claims abstract description 18
- 230000033001 locomotion Effects 0.000 claims description 63
- 230000003287 optical effect Effects 0.000 claims description 6
- 230000005856 abnormality Effects 0.000 claims description 5
- 230000006870 function Effects 0.000 description 8
- 230000007423 decrease Effects 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000001427 coherent effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Landscapes
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
本发明涉及一种群体异常行为检测方法，包括：采用群体结构动态演化的群体跟踪方法，对视频帧中的群体进行群体跟踪；在群体跟踪的过程中，对视频帧中跟踪到群体的每一帧的群体协同性值进行记录存储，直到群体跟踪结束；根据跟踪的群体的每一帧的群体协同性值，计算该群体d帧内的群体协同性值，并根据预设的阈值判断有无异常行为发生。本发明还涉及一种群体异常行为检测系统。本发明直接利用群体的基本特性来检测异常行为，非常简单方便，避免了复杂的模型学习过程，提高了查找造成安全问题原因的效率。
The invention relates to a group abnormal behavior detection method, comprising: adopting a group tracking method of group structure dynamic evolution to perform group tracking on a group in a video frame; The group synergy value of the frame is recorded and stored until the group tracking ends; according to the group synergy value of each frame of the tracked group, the group synergy value in the group d frame is calculated, and the presence or absence is determined according to the preset threshold. Abnormal behavior occurs. The invention also relates to a group abnormal behavior detection system. The invention directly utilizes the basic characteristics of the group to detect abnormal behavior, which is very simple and convenient, avoids the complex model learning process, and improves the efficiency of finding the cause of the security problem.
Description
技术领域technical field
本发明涉及一种群体异常行为检测方法及系统。The invention relates to a group abnormal behavior detection method and system.
背景技术Background technique
在节奏快速发展的现代社会中，人口的增长速度越来越快，一系列人口引发的问题越显突出。在人群密度比较密集的场所，比如火车站、汽车站等，安全问题尤为突出。通过对异常行为的检测，可以将监控视频中大量的对安防无用的信息过滤掉，节约了大量的人力。在人员复杂的公共场所中，通过对监控视频进行分析进而对一些造成安全问题的事件的发生进行判定并及时做出相应的响应，不仅能够对突发事件进行有效处理，而且对维护公共场所的安全和人民的生命财产安全等方面都有突出的贡献。In the fast-paced modern society, the population growth rate is getting faster and faster, and a series of problems caused by the population are becoming more and more prominent. In places with dense crowds, such as train stations and bus stations, safety issues are particularly prominent. Through the detection of abnormal behavior, a large amount of information that is useless to security can be filtered out in the surveillance video, saving a lot of manpower. In a public place with complex personnel, by analyzing the surveillance video to determine the occurrence of some events that cause safety problems and make corresponding responses in a timely manner, it can not only effectively deal with emergencies, but also maintain the safety of public places. It has made outstanding contributions to safety and the safety of people's lives and property.
目前对于群体的异常行为，由于人群规模和密度较大，所以大多以宏观的角度进行研究，即将群体当做一个整体进行研究。主要有以下步骤：对视频运动目标检测、跟踪；根据群体的运动特性进行监测；通过模型对群体轨迹建模，识别群体的异常行为。At present, due to the large size and density of the crowd, the abnormal behavior of the group is mostly studied from a macro perspective, that is, the group is studied as a whole. The main steps are as follows: detecting and tracking video moving objects; monitoring according to the movement characteristics of the group; modeling the group trajectory through the model to identify the abnormal behavior of the group.
可见，目前的群体异常行为检测方式大都需要建立模型，然后进行模型学习，效率较低而且过程复杂。It can be seen that most of the current group abnormal behavior detection methods need to establish a model and then perform model learning, which is inefficient and complicated.
发明内容SUMMARY OF THE INVENTION
有鉴于此，有必要提供一种群体异常行为检测方法及系统。In view of this, it is necessary to provide a group abnormal behavior detection method and system.
本发明提供一种群体异常行为检测方法，该方法包括如下步骤：a.采用群体结构动态演化的群体跟踪方法，对视频帧中的群体进行群体跟踪；b.在群体跟踪的过程中，对视频帧中跟踪到群体的每一帧的群体协同性值进行记录存储，直到群体跟踪结束；c.根据跟踪的群体的每一帧的群体协同性值，计算该群体d帧内的群体协同性值，并根据预设的阈值判断有无异常行为发生。The present invention provides a method for detecting abnormal group behavior, which comprises the following steps: a. using a group tracking method of dynamic evolution of group structure to perform group tracking on groups in video frames; b. in the process of group tracking, The group synergy value of each frame of the group tracked in the frame is recorded and stored until the group tracking ends; c. Calculate the group synergy value in the group d frame according to the group synergy value of each frame of the tracked group , and determine whether abnormal behavior occurs according to the preset threshold.
其中，所述的步骤a具体包括：利用光流法跟踪视频帧中提取到的特征点，并得到所述特征点的运动信息；根据得到的所述特征点的运动信息，计算特征点的运动模式是否一致，将运动模式一致的特征点按照密度进行聚类，使生成的块中所包含的特征点的密度较大；采用群体合并方法，根据上述生成的块，检测得到所述视频帧中的群体；采用分层的动态树结构，对上述检测得到的所述视频帧中的群体进行群体跟踪。Wherein, the step a specifically includes: using the optical flow method to track the feature points extracted from the video frame, and obtaining motion information of the feature points; calculating the motion information of the feature points according to the obtained motion information of the feature points Whether the patterns are consistent, the feature points with consistent motion patterns are clustered according to the density, so that the density of the feature points contained in the generated blocks is larger; using the group merging method, according to the above generated blocks, it is detected that the group; using a hierarchical dynamic tree structure, group tracking is performed on the group in the video frame obtained by the above detection.
所述群体协同性值通过如下方式获得：检测出由特征点组成的块；获得每一个特征点的群体协同性的运动方式；根据所述块中包含的绝大多数特征点的群体协同性的运动方式得到所述块的群体协同性的运动方式；根据上述块的群体协同性的运动方式得到所述群体的群体协同性值。The group synergy value is obtained by: detecting a block composed of feature points; obtaining the movement mode of the group synergy of each feature point; The movement mode of the block is used to obtain the movement mode of the group coordination of the block; the group coordination value of the group is obtained according to the movement mode of the group coordination of the above-mentioned block.
所述的步骤c中计算该群体d帧内的群体协同性值具体包括：每间隔d帧，通过函数φ对该群体在d帧内的群体协同性值进行计算，所述函数φ可以对群体在d帧内的群体协同性值进行平均计算，或者进行方差计算。In the step c, calculating the group synergy value in the group d frame specifically includes: every d frame, calculating the group synergy value of the group in the d frame by a function φ, the function φ can be used for the group. Population cooperativity values within d-frames are averaged, or variances are calculated.
所述的步骤c中根据预设的阈值判断有无异常行为发生具体包括：与前一个d帧的φ值做差；当差值大于预先设置的阈值T的时，则判断有异常发生；若差值小于阈值T，则重复上述过程，直到群体跟踪结束，如果直到群体跟踪结束，所述差值仍小于阈值T，则判断群体无异常行为。In the step c, judging whether there is abnormal behavior according to the preset threshold specifically includes: making a difference with the φ value of the previous d frame; when the difference is greater than the preset threshold T, then judging that there is an abnormality; if If the difference is less than the threshold T, the above process is repeated until the group tracking ends. If the difference is still smaller than the threshold T until the group tracking ends, it is determined that the group has no abnormal behavior.
本发明还提供一种群体异常行为检测系统，该系统包括跟踪模块、存储模块及判断模块，其中：所述跟踪模块用于采用群体结构动态演化的群体跟踪方法，对视频帧中的群体进行群体跟踪；所述存储模块用于在群体跟踪的过程中，对视频帧中跟踪到群体的每一帧的群体协同性值进行记录存储，直到群体跟踪结束；所述判断模块用于根据跟踪的群体的每一帧的群体协同性值，计算该群体d帧内的群体协同性值，并根据预设的阈值判断有无异常行为发生。The present invention also provides a group abnormal behavior detection system, the system includes a tracking module, a storage module and a judgment module, wherein: the tracking module is used to adopt a group tracking method of dynamic evolution of the group structure to perform group analysis on the group in the video frame. Tracking; the storage module is used to record and store the group synergy value of each frame of the group tracked in the video frame during the group tracking process, until the group tracking ends; the judgment module is used to track the group according to the group. The group synergy value of each frame of d is calculated, the group synergy value in the group d frame is calculated, and whether abnormal behavior occurs or not is judged according to the preset threshold.
其中，所述跟踪模块具体用于：利用光流法跟踪视频帧中提取到的特征点，并得到所述特征点的运动信息；根据得到的所述特征点的运动信息，计算特征点的运动模式是否一致，将运动模式一致的特征点按照密度进行聚类，使生成的块中所包含的特征点的密度较大；采用群体合并方法，根据上述生成的块，检测得到所述视频帧中的群体；采用分层的动态树结构，对上述检测得到的所述视频帧中的群体进行群体跟踪。Wherein, the tracking module is specifically used for: tracking the feature points extracted from the video frame by using the optical flow method, and obtaining the motion information of the feature points; calculating the motion information of the feature points according to the obtained motion information of the feature points Whether the patterns are consistent, the feature points with consistent motion patterns are clustered according to the density, so that the density of the feature points contained in the generated blocks is larger; using the group merging method, according to the above generated blocks, it is detected that the group; using a hierarchical dynamic tree structure, group tracking is performed on the group in the video frame obtained by the above detection.
所述的群体协同性值通过如下方式获得：检测出由特征点组成的块；获得每一个特征点的群体协同性的运动方式；根据所述块中包含的绝大多数特征点的群体协同性的运动方式得到所述块的群体协同性的运动方式；根据上述块的群体协同性的运动方式得到所述群体的群体协同性值。The group synergy value is obtained by: detecting a block composed of feature points; obtaining the movement mode of the group synergy of each feature point; according to the group synergy of most of the feature points contained in the block The movement mode of the block is obtained according to the movement mode of the group coordination of the block; the group coordination value of the group is obtained according to the movement mode of the group coordination of the above-mentioned block.
所述判断模块具体用于：每间隔d帧，通过函数φ对该群体在d帧内的群体协同性值进行计算；与前一个d帧的φ值做差；当差值大于预先设置的阈值T的时，则判断有异常发生；若差值小于阈值T，则重复上述过程，直到群体跟踪结束，如果直到群体跟踪结束，所述差值仍小于阈值T，则判断群体无异常行为。The judging module is specifically used for: calculating the group synergy value of the group in the d frame by the function φ every d frame; making a difference with the φ value of the previous d frame; when the difference value is greater than the preset threshold If the difference is less than the threshold T, the above process is repeated until the group tracking ends, and if the difference is still less than the threshold T until the group tracking ends, it is determined that the group has no abnormal behavior.
所述的函数φ可以对群体在d帧内的群体协同性值进行平均计算，或者进行方差计算。The function φ may perform an average calculation on the population cooperativity value of the population in the d frame, or perform a variance calculation.
本发明群体异常行为检测方法及系统，直接利用群体的一些基本特性来检测异常行为，非常简单方便，而且避免了复杂的模型学习的过程。本发明能够检测出群体运动突然变得杂乱无章的事件，不仅可以提高查找造成安全问题原因的效率，而且也可以使相关人员及时做出相应的响应,对突发事件进行有效处理。The method and system for detecting abnormal behavior of a group of the present invention directly utilize some basic characteristics of the group to detect abnormal behavior, which is very simple and convenient, and avoids the complex model learning process. The invention can detect the event that the group movement suddenly becomes disordered, which can not only improve the efficiency of finding the cause of the safety problem, but also enable the relevant personnel to respond in time and effectively deal with the emergency.
附图说明Description of drawings
图1为本发明一种群体异常行为检测方法的流程图；Fig. 1 is the flow chart of a kind of group abnormal behavior detection method of the present invention;
图2为本发明一种群体异常行为检测系统的硬件架构图。FIG. 2 is a hardware architecture diagram of a group abnormal behavior detection system according to the present invention.
具体实施方式Detailed ways
下面结合附图及具体实施例对本发明作进一步详细的说明。The present invention will be described in further detail below with reference to the accompanying drawings and specific embodiments.
参阅图1所示，是本发明一种群体异常行为检测方法较佳实施例的作业流程图。Referring to FIG. 1 , it is a working flow chart of a preferred embodiment of a method for detecting abnormal group behavior of the present invention.
步骤S1，采用群体结构动态演化的群体跟踪方法，对视频帧中的群体进行群体跟踪。具体步骤如下：In step S1, the group tracking method of the group structure dynamic evolution is used to perform group tracking on the group in the video frame. Specific steps are as follows:
(1)利用光流法跟踪视频帧中提取到的特征点，并得到所述特征点的运动信息。(1) Using the optical flow method to track the feature points extracted from the video frame, and obtain the motion information of the feature points.
(2)根据得到的所述特征点的运动信息，计算特征点的运动模式是否一致，将运动模式一致的特征点按照密度进行聚类，即将特征点密度大的类生成块(patch)。(2) According to the obtained motion information of the feature points, calculate whether the motion patterns of the feature points are consistent, and cluster the feature points with consistent motion patterns according to the density, that is, generate patches for classes with high feature point density.
(3)采用群体合并(Collective Merging)方法，根据上述生成的块，检测得到所述视频帧中的群体。(3) Using a collective merging method, the groups in the video frame are detected according to the above generated blocks.
需要说明的是，本实施例在检测群体时，通过群体协同性的运动模式进行群体检测。具体来说，因为每一个群体都具有一种群体协同性运动模式，所以可以通过群体协同性运动模式区分不同的群体。本发明就是以此为基础进行异常行为检测。It should be noted that, when detecting a group in this embodiment, group detection is performed through a group coordinated movement pattern. Specifically, because each group has a group cooperative movement pattern, different groups can be distinguished by the group cooperative movement pattern. The present invention performs abnormal behavior detection based on this.
(4)采用分层的动态树结构，对上述检测得到的所述视频帧中的群体进行群体跟踪。(4) Using a hierarchical dynamic tree structure, group tracking is performed on the groups in the video frames obtained by the above detection.
步骤S2，在群体跟踪的过程中，对视频帧中跟踪到群体的每一帧的群体协同性值进行记录存储，直到群体跟踪结束。具体如下：Step S2, in the process of group tracking, record and store the group synergy value of each frame in which the group is tracked in the video frame until the group tracking ends. details as follows:
在进行存储时，为视频帧中跟踪到的每一个群体分配一个ID号，然后记录该ID号对应的群体在每一帧的群体协同性值，并进行存储。其中，所述群体分配的ID号在整个跟踪过程保持不变。During storage, an ID number is assigned to each group tracked in the video frame, and then the group synergy value of the group corresponding to the ID number in each frame is recorded and stored. Wherein, the ID number assigned by the group remains unchanged throughout the tracking process.
群体协同性(crowd collectiveness)：群体协同性衡量的是群体中每一个个体在运动模式上协同一致的程度。若群体的协同一致的程度越大，群体协同性值就相对比较高，反之则比较低。所述群体协同性值通过如下方式获得：Crowd collectiveness: Crowd collectiveness measures the degree to which each individual in a group is coherent in movement patterns. If the degree of group synergy is greater, the group synergy value is relatively high, and vice versa. The population synergy value is obtained by:
本实施例在检测群体的时候，首先检测出由特征点组成的块；由于特征点的跟踪可以获得特征点的运动信息，进而获得每一个特征点的群体协同性的运动方式；块的群体协同性的运动方式，由块中所包含的绝大多数的特征点的群体协同性的运动方式决定，最后，采用群体合并方法，根据上述生成的块，检测得到所述视频帧中的群体，根据上述块的群体协同性的运动方式得到所述群体的群体协同性值。When detecting a group in this embodiment, firstly, a block composed of feature points is detected; the motion information of the feature point can be obtained due to the tracking of the feature point, and then the group cooperative movement mode of each feature point can be obtained; the group coordination of the block The characteristic motion mode is determined by the group cooperative motion mode of the vast majority of the feature points contained in the block. Finally, the group merging method is used to detect the group in the video frame according to the above generated block. The movement mode of the group synergy of the above-mentioned blocks obtains the group synergy value of the group.
步骤S3，根据跟踪的群体的每一帧的群体协同性值，计算该群体一定时间段内的群体协同性值，并根据预设的阈值判断有无异常行为发生。具体而言：Step S3, according to the group synergy value of each frame of the tracked group, calculate the group synergy value of the group within a certain period of time, and judge whether abnormal behavior occurs according to a preset threshold. in particular:
所述跟踪到的每一个群体都会有一个群体的协同性(Crowd Collectiveness)值，在群体中的行人运动比较一致、稳定时，群体协同性值会相对稳定，而且数值较高。当跟踪的群体和其它群体发生冲突或者因为其他的原因而导致群体中行人的运动杂乱无章，那么群体协同性值就会突然降低。当检测到该群体协同性值突然降低，则可以判断该群体发生异常行为。Each group that is tracked will have a group Crowd Collectiveness value. When the pedestrian movements in the group are relatively consistent and stable, the group cooperativeness value will be relatively stable and higher. When the tracked group collides with other groups or the movement of pedestrians in the group is disordered for other reasons, the group coordination value will suddenly decrease. When it is detected that the group's synergy value suddenly decreases, it can be determined that the group has abnormal behavior.
具体步骤如下：Specific steps are as follows:
每间隔d帧，通过函数φ对该群体在d帧内的群体协同性值进行计算，具体计算时，函数φ可以对群体在d帧内的群体协同性值进行平均计算，或者进行方差计算。然后与前一个d帧的φ值做差，当差值大于预先设置的阈值T的时候，即φn-φn-1>T时，则判断有异常发生。若差值小于预先设置的阈值T，则重复上述过程，直到群体跟踪结束。如果直到群体整个跟踪过程完成，差值仍小于预先设置的阈值，则判断群体无异常行为。Every d frame, the group synergy value of the group in the d frame is calculated by the function φ. In the specific calculation, the function φ can average the group synergy value of the group in the d frame, or perform variance calculation. Then make a difference with the φ value of the previous d frame. When the difference value is greater than the preset threshold T, that is, when φ n -φ n-1 >T, it is judged that an abnormality has occurred. If the difference is smaller than the preset threshold T, the above process is repeated until the group tracking ends. If the difference is still smaller than the preset threshold until the whole tracking process of the group is completed, it is judged that the group has no abnormal behavior.
参阅图2所示，是本发明一种群体异常行为检测系统的硬件架构图。该系统包括跟踪模块、存储模块及判断模块。Referring to FIG. 2 , it is a hardware architecture diagram of a group abnormal behavior detection system of the present invention. The system includes a tracking module, a storage module and a judgment module.
所述跟踪模块用于采用群体结构动态演化的群体跟踪方法，对视频帧中的群体进行群体跟踪。具体步骤如下：The tracking module is used for group tracking the group in the video frame by adopting the group tracking method of group structure dynamic evolution. Specific steps are as follows:
(1)利用光流法跟踪视频帧中提取到的特征点，并得到所述特征点的运动信息。(1) Using the optical flow method to track the feature points extracted from the video frame, and obtain the motion information of the feature points.
(2)根据得到的所述特征点的运动信息，计算特征点的运动模式是否一致，将运动模式一致的特征点按照密度进行聚类，即将特征点密度大的类生成块(patch)。(2) According to the obtained motion information of the feature points, calculate whether the motion patterns of the feature points are consistent, and cluster the feature points with consistent motion patterns according to the density, that is, generate patches for classes with high feature point density.
(3)采用群体合并(Collective Merging)方法，根据上述生成的块，检测得到所述视频帧中的群体。(3) Using a collective merging method, the groups in the video frame are detected according to the above generated blocks.
需要说明的是，本实施例在检测群体时，通过群体协同性的运动模式进行群体检测。具体来说，因为每一个群体都具有一种群体协同性运动模式，所以可以通过群体协同性运动模式区分不同的群体。本发明就是以此为基础进行异常行为检测。It should be noted that, when detecting a group in this embodiment, group detection is performed through a group coordinated movement pattern. Specifically, because each group has a group cooperative movement pattern, different groups can be distinguished by the group cooperative movement pattern. The present invention performs abnormal behavior detection based on this.
(4)采用分层的动态树结构，对上述检测得到的所述视频帧中的群体进行群体跟踪。(4) Using a hierarchical dynamic tree structure, group tracking is performed on the groups in the video frames obtained by the above detection.
所述存储模块用于在群体跟踪的过程中，对视频帧中跟踪到群体的每一帧的群体协同性值进行记录存储，直到群体跟踪结束。具体如下：The storage module is used for recording and storing the group synergy value of each frame of the group tracked in the video frame during the group tracking process until the group tracking ends. details as follows:
在进行存储时，为视频帧中跟踪到的每一个群体分配一个ID号，然后记录该ID号对应的群体在每一帧的群体协同性值，并进行存储。其中，所述群体分配的ID号在整个跟踪过程保持不变。During storage, an ID number is assigned to each group tracked in the video frame, and then the group synergy value of the group corresponding to the ID number in each frame is recorded and stored. Wherein, the ID number assigned by the group remains unchanged throughout the tracking process.
所述群体协同性值通过如下方式获得：The population synergy value is obtained by:
本实施例在检测群体的时候，首先检测出由特征点组成的块；由于特征点的跟踪可以获得特征点的运动信息，进而获得每一个特征点的群体协同性的运动方式；块的群体协同性的运动方式，由块中所包含的绝大多数的特征点的群体协同性的运动方式决定，最后，采用群体合并方法，根据上述生成的块，检测得到所述视频帧中的群体，根据上述块的群体协同性的运动方式得到所述群体的群体协同性值。When detecting a group in this embodiment, firstly, a block composed of feature points is detected; the motion information of the feature point can be obtained due to the tracking of the feature point, and then the group cooperative movement mode of each feature point can be obtained; the group coordination of the block The characteristic motion mode is determined by the group cooperative motion mode of the vast majority of the feature points contained in the block. Finally, the group merging method is used to detect the group in the video frame according to the above generated block. The movement mode of the group synergy of the above-mentioned blocks obtains the group synergy value of the group.
所述判断模块用于根据跟踪的群体的每一帧的群体协同性值，计算该群体一定时间段内的群体协同性值，并根据预设的阈值判断有无异常行为发生。具体而言：The judging module is used to calculate the group synergy value of the group within a certain period of time according to the group synergy value of each frame of the tracked group, and judge whether there is abnormal behavior according to a preset threshold. in particular:
所述跟踪到的每一个群体都会有一个群体的协同性(Crowd Collectiveness)值，在群体中的行人运动比较一致稳定时，群体协同性值会相对稳定，而且数值较高。当跟踪的群体和其它群体发生冲突或者因为其他的原因而导致群体中行人的运动杂乱无章，那么群体协同性值就会突然降低。当检测到该群体协同性值突然降低，则可以判断该群体发生异常行为。Each group that is tracked will have a group Crowd Collectiveness value. When the pedestrian movements in the group are relatively consistent and stable, the group cooperativeness value will be relatively stable and the value is high. When the tracked group collides with other groups or the movement of pedestrians in the group is disordered for other reasons, the group coordination value will suddenly decrease. When it is detected that the group's synergy value suddenly decreases, it can be determined that the group has abnormal behavior.
具体步骤如下：Specific steps are as follows:
每间隔d帧，通过函数φ对该群体在d帧内的群体协同性值进行计算，具体计算时，函数φ可以对群体在d帧内的群体协同性值进行平均计算，或者进行方差计算。然后与前一个d帧的φ值做差，当差值大于预先设置的阈值T的时候，即φn-φn-1>T时，则判断有异常发生。若差值小于预先设置的阈值T，则重复上述过程，直到群体跟踪结束。如果直到群体整个跟踪过程完成，差值仍小于预先设置的阈值，则判断群体无异常行为。Every d frame, the group synergy value of the group in the d frame is calculated by the function φ. In the specific calculation, the function φ can average the group synergy value of the group in the d frame, or perform variance calculation. Then make a difference with the φ value of the previous d frame. When the difference value is greater than the preset threshold T, that is, when φ n -φ n-1 >T, it is judged that an abnormality has occurred. If the difference is smaller than the preset threshold T, the above process is repeated until the group tracking ends. If the difference is still smaller than the preset threshold until the whole tracking process of the group is completed, it is judged that the group has no abnormal behavior.
虽然本发明参照当前的较佳实施方式进行了描述，但本领域的技术人员应能理解，上述较佳实施方式仅用来说明本发明，并非用来限定本发明的保护范围，任何在本发明的精神和原则范围之内，所做的任何修饰、等效替换、改进等，均应包含在本发明的权利保护范围之内。Although the present invention has been described with reference to the current preferred embodiments, those skilled in the art should understand that the above preferred embodiments are only used to illustrate the present invention, not to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the scope of the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims (4)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510160419.8A CN106157325B (en) | 2015-04-07 | 2015-04-07 | Group abnormal behavior detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510160419.8A CN106157325B (en) | 2015-04-07 | 2015-04-07 | Group abnormal behavior detection method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106157325A CN106157325A (en) | 2016-11-23 |
| CN106157325B true CN106157325B (en) | 2020-10-27 |
Family
ID=57337853
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510160419.8A Active CN106157325B (en) | 2015-04-07 | 2015-04-07 | Group abnormal behavior detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106157325B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106355605B (en) * | 2016-08-25 | 2018-12-28 | 深圳先进技术研究院 | Group movement consistency filter method |
| CN111401220B (en) * | 2020-03-12 | 2021-03-19 | 重庆特斯联智慧科技股份有限公司 | Crowd aggregation characteristic analysis method and system for intelligent security |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7386420B2 (en) * | 2006-06-02 | 2008-06-10 | United Microelectronics Corp. | Data analysis method for integrated circuit process and semiconductor process |
| CN101872418B (en) * | 2010-05-28 | 2012-09-12 | 电子科技大学 | Detection method based on group environment abnormal behavior |
| CN102799863B (en) * | 2012-07-02 | 2014-12-17 | 中国计量学院 | Method for detecting group crowd abnormal behaviors in video monitoring |
| CN103258193B (en) * | 2013-05-21 | 2016-06-22 | 西南科技大学 | A kind of group abnormality Activity recognition method based on KOD energy feature |
| CN103413321A (en) * | 2013-07-16 | 2013-11-27 | 南京师范大学 | Crowd behavior model analysis and abnormal behavior detection method under geographical environment |
| CN103473533B (en) * | 2013-09-10 | 2017-03-15 | 上海大学 | Moving Objects in Video Sequences abnormal behaviour automatic testing method |
| CN103593646A (en) * | 2013-10-16 | 2014-02-19 | 中国计量学院 | Dense crowd abnormal behavior detection method based on micro-behavior analysis |
| CN103745230B (en) * | 2014-01-14 | 2017-05-10 | 四川大学 | Adaptive abnormal crowd behavior analysis method |
| CN103996051A (en) * | 2014-05-12 | 2014-08-20 | 上海大学 | Method for automatically detecting abnormal behaviors of video moving object based on change of movement features |
| CN104123544B (en) * | 2014-07-23 | 2018-03-13 | 通号通信信息集团有限公司 | Anomaly detection method and system based on video analysis |
-
2015
- 2015-04-07 CN CN201510160419.8A patent/CN106157325B/en active Active
Non-Patent Citations (1)
| Title |
|---|
| Visual abnormal behavior detection based on trajectory sparse reconstruction analysis;CeLi 等;《Neurocomputing》;20131231;第94-100页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106157325A (en) | 2016-11-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110689054B (en) | Worker violation monitoring method | |
| CN103678299B (en) | A kind of method and device of monitor video summary | |
| CN106156706B (en) | Pedestrian abnormal behavior detection method | |
| CN103761748A (en) | Method and device for detecting abnormal behaviors | |
| CN103198296A (en) | Method and device of video abnormal behavior detection based on Bayes surprise degree calculation | |
| CN113011322B (en) | Detection model training method and detection method for specific abnormal behavior of surveillance video | |
| Arif et al. | Counting of people in the extremely dense crowd using genetic algorithm and blobs counting | |
| CN113568774A (en) | A real-time anomaly detection method for multidimensional time series data using unsupervised deep neural networks | |
| CN106157326B (en) | Group abnormal behavior detection method and system | |
| CN106157325B (en) | Group abnormal behavior detection method and system | |
| CN110674753A (en) | Theft early warning method, terminal device and storage medium | |
| An et al. | VFP290k: A large-scale benchmark dataset for vision-based fallen person detection | |
| CN104077571B (en) | A kind of crowd's anomaly detection method that model is serialized using single class | |
| CN111739651A (en) | A multi-body distance detection system and method based on group identification | |
| CN114500574A (en) | Monitoring method, device and medium for improving grain depot safety based on block chain | |
| CN117354469B (en) | District monitoring video target tracking method and system based on security precaution | |
| CN107920067A (en) | A kind of intrusion detection method in active objects storage system | |
| Mahmood et al. | Contextual anomaly detection based video surveillance system | |
| JP2014182585A (en) | Image analysis device, image analysis method and image analysis program | |
| Xu et al. | Detecting pedestrian crossing events in large video data from traffic monitoring cameras | |
| Xiao | Abnormal Event Detection and Localization Based on Crowd Analysis in Video Surveillance | |
| Wang et al. | Video data mining based on K-Means algorithm for surveillance video | |
| US12039781B2 (en) | Methods and apparatus for targeted analysis of video images and detection of persistent changes in a camera view | |
| Wang et al. | Anomaly detection in crowd scene using historical information | |
| CN116757336B (en) | Track traffic risk prediction method and system based on data driving |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |