CN114743136A - Abnormal behavior detection method, device and storage medium - Google Patents
Abnormal behavior detection method, device and storage medium Download PDFInfo
- Publication number
- CN114743136A CN114743136A CN202210325824.0A CN202210325824A CN114743136A CN 114743136 A CN114743136 A CN 114743136A CN 202210325824 A CN202210325824 A CN 202210325824A CN 114743136 A CN114743136 A CN 114743136A
- Authority
- CN
- China
- Prior art keywords
- abnormal behavior
- video
- detected
- information
- behavior detection
- 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
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The application relates to an abnormal behavior detection method, an abnormal behavior detection device and a storage medium, wherein the abnormal behavior detection method comprises the following steps: acquiring a video to be detected; coding the video to be detected to acquire intermediate coding information generated in the coding process, and taking the intermediate coding information as the intermediate coding information to be detected; and judging whether the video to be detected meets a preset abnormal behavior detection standard or not according to the intermediate coding information to be detected, and if so, judging that the video to be detected has abnormal behavior. The method and the device have the effects of reducing the complexity of work, reducing the workload of equipment and avoiding time delay caused by complex encoding and decoding processes.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a method and an apparatus for detecting abnormal behavior, a computer device, and a readable storage medium.
Background
In order to guarantee social order and public safety, video monitoring equipment in public places is increased year by year. In the face of massive monitoring video data, a traditional video monitoring system usually monitors a monitoring video in real time through manual work or calls the monitoring video after an abnormality occurs. The monitoring mode generally causes the problems of low monitoring efficiency, poor accuracy, insufficient response real-time performance and the like.
The existing automatic detection method for monitoring abnormal behaviors improves the detection efficiency of security monitoring abnormal events to a great extent. However, the method needs to encode the video data acquired by the monitoring device, and then performs complex processing and analysis of a spatial domain and a time domain on the decoded video, resulting in low real-time performance, and thus high real-time performance detection in an unexpected abnormal scene cannot be satisfied. Therefore, the inventors considered that the following drawbacks exist in the above population abnormal behavior detection method.
In the first aspect, there are a plurality of delay links in the process from video acquisition to acquisition of video data that can be processed and analyzed. In the existing method for automatically detecting abnormal monitoring behaviors, in order to obtain video data which can be processed and analyzed, links such as camera video acquisition, video coding, coded code stream transmission, video code stream decoding of monitoring center equipment and the like are needed to finally obtain the video data which can be processed and analyzed, and the coding delay, the transmission delay and the decoding delay generated in the intermediate links greatly influence the real-time performance of abnormal behavior detection.
And in the second aspect, abnormal behavior detection delay of the monitoring video is carried out in a space domain and a time domain. The existing group abnormal behavior detection method takes monitoring video data as a data base for moving target detection and abnormal behavior analysis, and the moving target detection needs to carry out pixel-by-pixel or pixel-by-pixel block operation on the video data, so that the operation complexity is high, and larger time delay is brought. And abnormal behavior analysis based on clustering and classification, reasoning models or coefficient expression requires training and analysis of a large amount of characteristic information of spatial domains and time domains, thereby further bringing about large delay.
In the third aspect, the moving object detection and the abnormal behavior analysis are high in complexity and need to be performed in an image domain, so that the real-time performance is poor.
Disclosure of Invention
In order to solve at least the above problems, the present application provides an abnormal behavior detection method, apparatus, computer device, and readable storage medium. According to the technical scheme, the moving target detection and the abnormal behavior detection are carried out by adopting the intermediate coding information generated in the process of coding the video, complex coding and decoding of the monitoring video are not needed, so that the complexity of work is reduced, the workload of equipment is reduced, and meanwhile, the time delay caused by the complex coding and decoding processes is also avoided.
The application adopts the following technical scheme to realize the purpose:
in a first aspect, the present application provides an abnormal behavior detection method, including the following steps: acquiring a video to be detected; coding the video to be detected to acquire intermediate coding information generated in the coding process, and taking the intermediate coding information as the intermediate coding information to be detected; and judging whether the video to be detected meets a preset abnormal behavior detection standard or not according to the intermediate coding information to be detected, and if so, judging that the video to be detected has abnormal behavior.
By adopting the technical scheme, the intermediate coding information generated in the process of coding the video to be detected is adopted to judge whether the video to be detected meets the preset abnormal behavior detection standard. The complex encoding and decoding of the monitoring video and the pixel-by-pixel or pixel block-by-pixel block operation of the video to be detected can be avoided, so that the complexity of the work is reduced, the workload of the equipment is reduced, and the time delay caused by the complex encoding and decoding process is avoided.
Optionally, the constructing of the abnormal behavior detection criterion includes the following steps: acquiring a training video; coding the training video to acquire intermediate coding information generated in the coding process, and taking the intermediate coding information as training intermediate coding information; and constructing an abnormal behavior detection standard according to the training intermediate coding information.
By adopting the technical scheme, the abnormal behavior detection standard can be constructed by utilizing the intermediate coding information generated in the process of coding the training video without carrying out complicated coding and decoding on the training video, so that the complexity of constructing an abnormal behavior detection model is reduced, and the workload of equipment is reduced.
Optionally, the obtaining of the intermediate coding information generated in the coding process includes the following steps: acquiring local characteristic information generated in the encoding process of each encoding block in each frame of an encoded video; and carrying out statistical calculation on the local characteristic information to obtain global characteristic information of each frame, and taking the global characteristic information as intermediate coding information.
By adopting the technical scheme, the local characteristic information of each coding block of each frame in the frame sequence of the video can be selected in the process of coding the video, the global characteristic information of each frame can be obtained according to the local characteristic information, the global motion mode of each frame in the video can be further analyzed according to the local characteristic information or the global characteristic information, and the ingenious selection, processing and utilization of the intermediate coding information in the video coding process can be realized.
Optionally, the local feature information includes one or more of the following information: motion vector information, block partition information, and code rate information.
By adopting the technical scheme, in the process of coding the video, the information of each coding block in the frame sequence of the video, which is related to the motion mode, is selected to analyze and determine the motion mode of the object in the video, so that the analysis result can be more accurate.
Optionally, the constructing the abnormal behavior detection standard according to the training intermediate coding information includes the following steps: clustering the training intermediate coding information in the corresponding feature space; determining abnormal behavior motion modes according to the clustering result, wherein the types of the abnormal behavior motion modes are at least one; and taking whether at least one of the abnormal behavior motion patterns is included as an abnormal behavior detection standard.
By adopting the technical scheme, the global characteristic information in the intermediate coding information is processed and analyzed in a clustering mode so as to determine the abnormal behavior movement mode. Because clustering is a mature and common method, abnormal behavior motion patterns can be determined more quickly and accurately.
Optionally, the determining, according to the intermediate coding information to be detected, whether the video to be detected meets a preset abnormal behavior detection standard includes the following steps: determining a global motion mode of each frame in the video to be detected according to the intermediate coding information to be detected; and judging whether the global motion mode belongs to the abnormal behavior motion mode, and if so, judging that the video to be detected meets the abnormal behavior detection standard.
By adopting the technical scheme, the global motion mode and the abnormal behavior motion mode of each frame are compared, so that the condition that the video to be detected meets the abnormal behavior detection standard can be conveniently and quickly judged.
Optionally, the abnormal behavior detection method further includes: and responding to the video to be detected to accord with the abnormal behavior detection standard, and outputting a corresponding abnormal behavior motion mode.
By adopting the technical scheme, the abnormal behaviors can be found in time, and the movement mode of the abnormal behaviors can be output to the user.
In a second aspect, the present application provides an abnormal behavior detection apparatus, including: the moving target detection module is used for acquiring a video to be detected, encoding the video to be detected to acquire intermediate encoding information generated in the encoding process, and taking the intermediate encoding information as the intermediate encoding information to be detected; and the abnormal behavior detection module comprises a preset abnormal behavior detection standard and is used for judging whether the video to be detected meets the preset abnormal behavior detection standard or not according to the intermediate coded information to be detected, and if so, judging that the video to be detected has abnormal behavior.
By adopting the technical scheme, the intermediate coding information generated in the process of coding the video to be detected is adopted to judge whether the video to be detected meets the preset abnormal behavior detection standard. The decoder is not needed, so that the equipment cost is reduced, the complexity of work is reduced, the workload is reduced, and in addition, the time delay caused by the complicated coding and decoding processes is also avoided.
In a third aspect, the present application provides a computer device, including a memory and a processor, where the memory stores a computer program, and the processor implements any of the steps of the above abnormal behavior detection method when executing the computer program.
By adopting the technical scheme, when a computer program in the processor is executed, the abnormal behavior detection method is realized, and whether the video to be detected meets the preset abnormal behavior detection standard or not is judged by adopting intermediate coding information generated in the process of coding the video to be detected. The complex encoding and decoding of the monitoring video and the pixel-by-pixel or pixel block-by-pixel block operation of the video to be detected can be avoided, so that the complexity of the work is reduced, the workload of the equipment is reduced, and the time delay caused by the complex encoding and decoding process is avoided.
In a fourth aspect, the present application provides a computer-readable storage medium storing a computer program, which when executed by a processor implements the steps of any of the above-described abnormal behavior detection methods.
By adopting the technical scheme, when the computer program is executed, the abnormal behavior detection method is realized, and whether the video to be detected meets the preset abnormal behavior detection standard or not is judged by adopting intermediate coding information generated in the process of coding the video to be detected. The complex encoding and decoding of the monitoring video and the pixel-by-pixel or pixel block-by-pixel block operation of the video to be detected can be avoided, so that the complexity of the work is reduced, the workload of the equipment is reduced, and the time delay caused by the complex encoding and decoding process is avoided.
In summary, the present application includes at least one of the following beneficial technical effects:
1. and judging whether the video to be detected meets the preset abnormal behavior detection standard or not by adopting intermediate coding information generated in the process of coding the video to be detected. The complex encoding and decoding of the monitoring video and the pixel-by-pixel or pixel block-by-pixel block operation of the video to be detected can be avoided, so that the working complexity is reduced, and the working load of the equipment is reduced.
2. The time delay caused by complicated encoding and decoding processes is avoided.
3. The method can be carried out in a parameter domain without a complex video data processing link, and does not need to enter an image domain for processing.
Drawings
Fig. 1 is a schematic flow chart of an abnormal behavior detection method according to an embodiment of the present application;
FIG. 2 is a schematic flow chart illustrating the construction of abnormal behavior detection criteria according to an embodiment of the present application;
FIG. 3 is a flowchart illustrating an example of obtaining intermediate encoded information generated during an encoding process according to the present application;
FIG. 4 is a diagram illustrating an embodiment of the present application for obtaining global feature information of each frame according to local feature information;
FIG. 5 is a schematic flow chart illustrating the construction of abnormal behavior detection criteria according to training intermediate coding information according to an embodiment of the present application;
FIG. 6 is a schematic diagram of feature space clustering based on one embodiment of the present application;
fig. 7 is a schematic flowchart illustrating an embodiment of determining whether a video to be detected meets an abnormal behavior detection standard according to intermediate coding information to be detected;
fig. 8 is a structural framework diagram of an abnormal behavior detection apparatus according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is further described in detail below with reference to fig. 1-8 and the embodiments. It should be understood that the specific embodiments described herein are merely illustrative of and not restrictive on the broad application.
The embodiment of the application discloses an abnormal behavior detection method. Referring to fig. 1, the method includes steps S101-S103. At step S101, a video to be detected is acquired. The video to be detected can be a surveillance video, and in different implementation scenarios, the video to be detected can be directly acquired from a camera or acquired from a storage device (e.g., a usb disk, a network storage server, etc.). In step S102, the video to be detected is input into the encoder, and the encoder may first obtain a frame sequence of the video to be encoded, and then encode the coding block of each frame in the frame sequence, so as to obtain intermediate coding information generated in the encoding process, and use the intermediate coding information as the intermediate coding information to be detected. In step S103, it is determined whether the video to be detected meets a preset abnormal behavior detection standard according to the intermediate coding information to be detected, and if so, it is determined that the video to be detected has an abnormal behavior.
In one embodiment, referring to fig. 2, the construction of the preset abnormal behavior detection criteria may include steps S201 to S203. At step S201, a training video is acquired. In one implementation scenario, the training video herein may be a surveillance video that has been identified as having abnormal behavior. In step S202, the training video is input into an encoder, where the encoder may first obtain a frame sequence of the training video, then encode a coding block of each frame in the frame sequence, respectively, to obtain intermediate coding information generated in an encoding process, and use the intermediate coding information as training intermediate coding information; at step S203, an abnormal behavior detection criterion is constructed from the training intermediate encoding information.
In an application scenario, the acquiring of the intermediate encoding information generated in the encoding process in steps S102 and S202 may include steps S301 to S302. Referring to fig. 3, in step S301, local feature information generated during encoding of each encoding block in each frame of an encoded video is acquired. In different application scenarios, the local feature information includes one or more of the following information: motion vector information, block division information, and code rate information. For example, in one embodiment, the local feature information may be motion vector information, and the motion vector information may reflect a motion direction of the object. In another embodiment, the local feature information may be block division information, and the block division information may reflect the intensity of the motion of the object. And may reflect the extent of region non-translational motion, such as complex motions like rotational telescoping. In another embodiment, the local feature information may be code rate information, and the code rate information may also reflect the intensity of the motion of the object, and may reflect the increase of the bit number of the coded motion information due to the large-amplitude translational motion. In addition, in an embodiment, the local feature information may further include motion vector information and block division information, or motion vector information and code rate information, or motion vector information, block division information, and code rate information.
Referring to fig. 4, the local feature information may form a local motion pattern of each coding block and form a local motion pattern feature vector. Therefore, the motion vector information, the block division information and the code rate information can provide more sufficient local motion change degree judgment for the monitored scene (because the direction of the motion of the object can be reflected, and the intensity of the motion of the object can be reflected).
In step S302, statistical calculation is performed on the local feature information to obtain global feature information of each frame, and the global feature information is used as intermediate coding information. In an application scenario, the above statistical calculation process may be as follows: the local motion mode feature vectors of all coding blocks included in each video frame are respectively statistically calculated, for example, the variance of the motion vector information of all coding blocks, the second-order central moment of block division information, and the variance of code rate information are respectively calculated. Accordingly, the global motion statistical information, the global block division statistical information and the global code rate statistical information of each frame are respectively obtained. These three pieces of information form a video frame global motion pattern feature vector. The global motion pattern feature vector reflects the global motion pattern of each frame. Therefore, the motion vector information, the block division information, and the code rate information can also provide a more sufficient overall motion change degree discrimination for the monitored scene.
In one embodiment, referring to FIG. 5, the above construction of abnormal behavior detection criteria from training intermediate encoded information may include steps S501-S503. In step S501, global feature information of the training intermediate coding information is clustered in feature spaces of corresponding dimensions. When the intermediate coded information contains one kind of information, it can be clustered in one-dimensional space; when the intermediate coding information contains two kinds of information, the intermediate coding information can be clustered in a two-dimensional space; when the intermediate coding information contains three kinds of information, the intermediate coding information can be clustered in a three-dimensional space; the higher the dimension, the more accurate the motion pattern it reflects. The clustering method can be realized by a clustering algorithm (such as a K-pototypes algorithm, a CLARANS algorithm and the like).
At step S502, an abnormal behavior motion pattern is determined according to the above clustering result. The type of the abnormal behavior motion pattern is at least one, for example, when the global feature information is motion vector information, the abnormal behavior motion pattern may be a reverse movement and a deviation from an original track, etc.; when the above-described global feature information is block division information, the abnormal behavior motion pattern may be an overspeed movement or the like. When the above global feature information is motion vector information and block division information, the abnormal behavior motion pattern may include reverse movement, deviation from an original track, overspeed movement, overspeed reverse movement, and the like.
Specifically, referring to fig. 6, a cluster of a plurality of center points may be formed by clustering, and different clusters reflect different motion patterns. By setting a clustering algorithm, the cluster closest to the coordinate starting point (cluster center) can be a normal behavior motion mode, other clusters reflect various abnormal behavior motion modes, and the types of the abnormal behavior motion modes are at least one. And then calculating the distance between each cluster and the cluster center, judging the behavior category under the monitoring scene according to the cluster distance, and judging whether abnormal behaviors exist. The farther away from the center of the cluster of normal behavior movement patterns, the more severe the abnormal behavior is.
In step S503, whether at least one of the above abnormal behavior motion patterns is included as an abnormal behavior detection criterion, and if so, it is determined that the video to be detected meets the abnormal behavior detection criterion. In an application scenario, when abnormal behaviors of a video to be detected are detected, the abnormal behavior motion patterns contained in the video to be detected can be output to a user at the same time, so that the user can implement corresponding countermeasures aiming at different abnormal behaviors in time. The abnormal behavior motion pattern may also be trained in advance and summarized in other ways, but its dimensions (for example, motion vector dimensions, block division dimensions, and code rate dimensions) include dimensions included in the inter-coded information to be detected.
In an embodiment, referring to fig. 7, the above determining whether the video to be detected meets the preset abnormal behavior detection standard according to the intermediate coding information to be detected may include steps S701 to S702. In step S701, a global motion pattern of each frame in the video to be detected is determined according to the inter-coded information to be detected (including the global feature information of each frame in the video to be detected). In step S702, it is determined that the video to be detected meets the abnormal behavior detection criterion according to the global motion pattern, that is, it is determined whether the global motion pattern matches one of the abnormal behavior motion patterns, and if so, it is determined that the video to be detected meets the abnormal behavior detection criterion.
The following takes a square monitoring video as an example, and an overall flow of the abnormal behavior detection method of the present application is exemplarily described. Firstly, constructing an abnormal behavior detection standard. Specifically, a video with a resolution of 1280x720 is continuously acquired by a square monitoring camera, then a video with abnormal behaviors is screened out to be used as a training video, or the existing training video with abnormal behaviors is passed through, then the training video is input into an H.265 (an encoder model) encoder to be encoded, basic encoding blocks are 64x64 blocks, and local feature information of 225 encoding blocks, including motion vector information, block division information and block-by-block code rate information, is output by each frame of encoding. And clustering feature spaces under three dimensions of motion vector information, block division information and code rate information, clustering training samples to form a cluster of a plurality of central points, wherein the cluster is not matched with the cluster of normal behaviors, namely, an abnormal behavior motion mode is formed, and the type of the abnormal behavior motion mode is at least one. And calculating the distance between each cluster and the cluster center, wherein the farther the cluster center is away from the normal behavior, the more violent the abnormal behavior is. Finally, whether at least one abnormal behavior motion mode is contained or not is used as an abnormal behavior detection standard,
and secondly, detecting the monitoring video serving as the video to be detected. Specifically, firstly, a video with a resolution of 1280x720 is continuously acquired by a square monitoring camera, and the video is input into an h.265 (an encoder model) encoder as a video to be detected for encoding, a basic encoding block is 64x64 blocks, and local feature information of 225 encoding blocks, including motion vector information, block division information, and block-by-block code rate information, is output by each frame encoding. Then, the local motion mode feature vectors of all the coding blocks contained in each frame are respectively calculated statistically. The method comprises the steps of calculating statistical characteristics of motion vector information, calculating statistical characteristics of block partition information and calculating statistical characteristics of code rate information to obtain global motion statistical information GlobalmotionInfo, global block partition statistical information GlobalPartitionInfo and global code rate statistical information GlobalbiteInfo, wherein the global motion mode characteristic vectors [ GlobalmotionInfo, GlobalbistritionInfo and GlobalbiteInfo ] of each frame in a video are formed by the three statistical information. And then, determining the global motion mode of each frame in the video to be detected according to the global motion mode feature vector. And finally, judging whether the global motion mode contains the abnormal behavior motion mode, if so, judging that the video to be detected accords with the abnormal behavior detection standard, specifically, respectively comparing the global motion mode with the abnormal behavior motion mode, and if the global motion mode is the same as at least one of the abnormal behavior motion modes, judging that the video to be detected has abnormal behavior, namely, the video to be detected accords with the abnormal behavior detection standard.
The implementation principle of the abnormal behavior detection method disclosed by the embodiment of the application is as follows: since the current video coding is performed on a block basis, a frame of video is divided into different blocks, and then each block is subjected to a coding process. The existing video coding architecture based on predictive coding and transform coding generates rich intermediate coding information in the coding process, such as motion vector information of a coding block, block division information of the coding block, and code rate information of a frame level and a block level. Compared with the simple motion information, the intermediate information provided by the encoder reflects the motion mode of the object in the video, and can provide more sufficient overall and local motion change degree judgment for the monitored scene.
Based on the reasons, at least one abnormal behavior motion mode can be formed by collecting one or more of the information of the training videos which are judged to contain abnormal behaviors, counting and clustering the information, and finally whether the abnormal behavior motion mode is contained or not is used as an abnormal behavior detection standard. Similarly, one or more of the above information of the video to be detected is collected, and the information is counted and clustered to form a global motion mode of each frame contained in the video to be detected, and finally, the global motion mode of each frame in the video to be detected is compared and matched with an abnormal behavior motion mode (at least one), and if the global motion mode of a certain frame in the video to be detected is the abnormal behavior motion mode, the video to be detected is judged to have abnormal behavior. The embodiment of the application makes full use of the intermediate information output by the encoder in the encoding process, does not need to perform complex encoding and decoding work, and simultaneously avoids the time delay caused by the complex encoding and decoding process.
In addition, an abnormal behavior detection device is further disclosed in an embodiment of the present application, and with reference to fig. 8, the device may include a moving object detection module and an abnormal behavior detection module, where the moving object detection module is configured to acquire a video to be detected, encode the video to be detected, acquire intermediate coding information generated in an encoding process, and use the intermediate coding information as the intermediate coding information to be detected. The abnormal behavior detection module comprises a preset abnormal behavior detection standard and is used for judging whether the video to be detected meets the preset abnormal behavior detection standard or not according to the intermediate coded information to be detected, and if so, judging that the video to be detected has abnormal behavior. Meanwhile, the abnormal behavior detection module can also acquire the video to be detected in real time so as to take the video to be detected as a training video, so that the abnormal behavior detection model can be trained in real time, and the abnormal behavior detection standard can be updated in real time.
Meanwhile, the embodiment of the application also discloses computer equipment which comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes the steps of the abnormal behavior detection method when executing the computer program.
Finally, the embodiment of the application also discloses a computer readable storage medium, which stores a computer program capable of being loaded by a processor and executing the abnormal behavior detection method. The computer-readable storage medium includes, for example: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The foregoing is a preferred embodiment of the present application and is not intended to limit the scope of the application in any way, and any features disclosed in this specification (including the abstract and drawings) may be replaced by alternative features serving equivalent or similar purposes, unless expressly stated otherwise. That is, unless expressly stated otherwise, each feature is only an example of a generic series of equivalent or similar features.
Claims (10)
1. An abnormal behavior detection method is characterized by comprising the following steps:
acquiring a video to be detected;
coding the video to be detected to acquire intermediate coding information generated in the coding process, and taking the intermediate coding information as the intermediate coding information to be detected;
and judging whether the video to be detected meets a preset abnormal behavior detection standard or not according to the intermediate coding information to be detected, and if so, judging that the video to be detected has abnormal behavior.
2. The abnormal behavior detection method according to claim 1, wherein the construction of the abnormal behavior detection criterion comprises the steps of:
acquiring a training video;
coding the training video to acquire intermediate coding information generated in the coding process, and taking the intermediate coding information as training intermediate coding information;
and constructing an abnormal behavior detection standard according to the training intermediate coding information.
3. The abnormal behavior detection method according to claim 1 or 2, wherein the obtaining of the intermediate coding information generated in the coding process comprises the following steps:
acquiring local characteristic information generated in the encoding process of each encoding block in each frame of an encoded video;
and carrying out statistical calculation on the local characteristic information to obtain global characteristic information of each frame, and taking the global characteristic information as intermediate coding information.
4. The abnormal behavior detection method according to claim 3, wherein the local feature information includes one or more of the following information: motion vector information, block division information, and code rate information.
5. The abnormal behavior detection method of claim 4, wherein the constructing of the abnormal behavior detection criteria based on the training intermediate coded information comprises the steps of:
clustering the training intermediate coding information in the corresponding feature space;
determining abnormal behavior motion modes according to the clustering result, wherein the types of the abnormal behavior motion modes are at least one;
and determining whether at least one of the abnormal behavior motion patterns is included as an abnormal behavior detection standard.
6. The abnormal behavior detection method according to claim 5, wherein the step of determining whether the video to be detected meets the preset abnormal behavior detection standard according to the intermediate coded information to be detected comprises the following steps:
determining a global motion mode of each frame in the video to be detected according to the intermediate coding information to be detected;
and judging whether the global motion mode belongs to the abnormal behavior motion mode, and if so, judging that the video to be detected meets the abnormal behavior detection standard.
7. The abnormal behavior detection method according to claim 6, further comprising: and responding to the video to be detected to accord with the abnormal behavior detection standard, and outputting a corresponding abnormal behavior motion mode.
8. An abnormal behavior detection apparatus, comprising:
the moving target detection module is used for acquiring a video to be detected, encoding the video to be detected to acquire intermediate coding information generated in the encoding process, and taking the intermediate coding information as the intermediate coding information to be detected; and
and the abnormal behavior detection module comprises a preset abnormal behavior detection standard and is used for judging whether the video to be detected accords with the preset abnormal behavior detection standard or not according to the intermediate coding information to be detected, and if so, judging that the video to be detected has abnormal behavior.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 7.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210325824.0A CN114743136A (en) | 2022-03-30 | 2022-03-30 | Abnormal behavior detection method, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210325824.0A CN114743136A (en) | 2022-03-30 | 2022-03-30 | Abnormal behavior detection method, device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114743136A true CN114743136A (en) | 2022-07-12 |
Family
ID=82279067
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210325824.0A Pending CN114743136A (en) | 2022-03-30 | 2022-03-30 | Abnormal behavior detection method, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114743136A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116052047A (en) * | 2023-01-29 | 2023-05-02 | 荣耀终端有限公司 | Moving object detection method and related equipment thereof |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101227599A (en) * | 2008-01-24 | 2008-07-23 | 苏州科技学院 | Intelligent video monitor terminal |
| KR20190046018A (en) * | 2017-10-25 | 2019-05-07 | 한국전자통신연구원 | Method of detecting abnormal behavior on the network and apparatus using the same |
| CN111144375A (en) * | 2019-12-31 | 2020-05-12 | 中国民用航空总局第二研究所 | Abnormal behavior detection method and device based on feature coding and electronic equipment |
| CN111178523A (en) * | 2019-08-02 | 2020-05-19 | 腾讯科技（深圳）有限公司 | Behavior detection method and device, electronic equipment and storage medium |
| CN111915015A (en) * | 2020-06-30 | 2020-11-10 | 上海金仕达软件科技有限公司 | Abnormal value detection method and device, terminal equipment and storage medium |
| CN112016500A (en) * | 2020-09-04 | 2020-12-01 | 山东大学 | Group abnormal behavior identification method and system based on multi-scale time information fusion |
| CN112597864A (en) * | 2020-12-16 | 2021-04-02 | 佳都新太科技股份有限公司 | Monitoring video abnormity detection method and device |
| CN113033424A (en) * | 2021-03-29 | 2021-06-25 | 广东众聚人工智能科技有限公司 | Multi-branch video anomaly detection method and system |
-
2022
- 2022-03-30 CN CN202210325824.0A patent/CN114743136A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101227599A (en) * | 2008-01-24 | 2008-07-23 | 苏州科技学院 | Intelligent video monitor terminal |
| KR20190046018A (en) * | 2017-10-25 | 2019-05-07 | 한국전자통신연구원 | Method of detecting abnormal behavior on the network and apparatus using the same |
| CN111178523A (en) * | 2019-08-02 | 2020-05-19 | 腾讯科技（深圳）有限公司 | Behavior detection method and device, electronic equipment and storage medium |
| CN111144375A (en) * | 2019-12-31 | 2020-05-12 | 中国民用航空总局第二研究所 | Abnormal behavior detection method and device based on feature coding and electronic equipment |
| CN111915015A (en) * | 2020-06-30 | 2020-11-10 | 上海金仕达软件科技有限公司 | Abnormal value detection method and device, terminal equipment and storage medium |
| CN112016500A (en) * | 2020-09-04 | 2020-12-01 | 山东大学 | Group abnormal behavior identification method and system based on multi-scale time information fusion |
| CN112597864A (en) * | 2020-12-16 | 2021-04-02 | 佳都新太科技股份有限公司 | Monitoring video abnormity detection method and device |
| CN113033424A (en) * | 2021-03-29 | 2021-06-25 | 广东众聚人工智能科技有限公司 | Multi-branch video anomaly detection method and system |
Non-Patent Citations (2)
| Title |
|---|
| XIMIN CAI ET AL: "Detecting Abnormal Behavior in Examination Surveillance Video with 3D Convolutional Neural Networks", 《2016 6TH INTERNATIONAL CONFERENCE ON DIGITAL HOME (ICDH)》 * |
| 卢博文: "基于深度学习的监控视频中的异常行为的检测算法研究", 《中国优秀硕士学位论文全文数据库（电子期刊）》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116052047A (en) * | 2023-01-29 | 2023-05-02 | 荣耀终端有限公司 | Moving object detection method and related equipment thereof |
| CN116052047B (en) * | 2023-01-29 | 2023-10-03 | 荣耀终端有限公司 | Moving object detection method and related equipment thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111553259B (en) | Image duplicate removal method and system | |
| CN109685097B (en) | Image detection method and device based on GAN | |
| JP6016332B2 (en) | Image processing apparatus and image processing method | |
| CN116188821A (en) | Copyright detection method, system, electronic device and storage medium | |
| CN107229710A (en) | A kind of video analysis method accorded with based on local feature description | |
| CN114743136A (en) | Abnormal behavior detection method, device and storage medium | |
| CN116258991A (en) | Abnormality detection apparatus, abnormality detection system, abnormality detection method, and computer program product | |
| US11704840B2 (en) | Attribute information prediction method, encoder, decoder and storage medium | |
| CN112597964A (en) | Method for counting layered multi-scale crowd | |
| CN117220920A (en) | Firewall policy management method based on artificial intelligence | |
| CN109101884B (en) | Pulse array prediction method | |
| CN112149596A (en) | Abnormal behavior detection method, terminal device and storage medium | |
| KR102177494B1 (en) | method of identifying abnormal-motion objects in compressed video by use of trajectory and pattern of motion vectors | |
| CN115909215A (en) | Edge intrusion early warning method and system based on target detection | |
| KR20200067682A (en) | syntax-based method of providing object classification in compressed video by use of neural network which is learned by cooperation with an external commercial classifier | |
| CN115376037A (en) | Station key area safety state monitoring method based on video | |
| CN112767438B (en) | Multi-target tracking method combining space-time motion | |
| Nascimento et al. | Public dataset of parking lot videos for computational vision applied to surveillance | |
| CN112149683A (en) | Method and device for detecting living objects in night vision environment | |
| KR20200068102A (en) | method of providing object classification for compressed video by use of syntax-based MRPN-CNN | |
| CN117354495B (en) | Video monitoring quality diagnosis method and system based on deep learning | |
| AU2021104124A4 (en) | A system and method for object extraction and encoding for video monitoring through low-bandwidth networks | |
| Yang et al. | 3D U-Net for Video Anomaly Detection | |
| CN110738692A (en) | spark cluster-based intelligent video identification method | |
| Rajarapollu et al. | Extraction of Visual Features from Video Sequences for Better Visual Analysis |
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 |