CN115311735A - Intelligent recognition early warning method for abnormal behaviors - Google Patents
Intelligent recognition early warning method for abnormal behaviors Download PDFInfo
- Publication number
- CN115311735A CN115311735A CN202210543585.6A CN202210543585A CN115311735A CN 115311735 A CN115311735 A CN 115311735A CN 202210543585 A CN202210543585 A CN 202210543585A CN 115311735 A CN115311735 A CN 115311735A
- Authority
- CN
- China
- Prior art keywords
- recognition
- early warning
- abnormal behavior
- identification
- target
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Social Psychology (AREA)
- Psychiatry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Human Computer Interaction (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an intelligent abnormal behavior recognition early warning method, and relates to the technical field of human behavior recognition, target tracking and face recognition. According to the invention, by applying leading-edge information technologies such as artificial intelligence, internet of things, big data, 5G mobile internet and the like, an edge computing visual perception platform is built through a high-altitude camera network covering a wide target area, and the abnormal behavior place of the student is monitored and accurately positioned in real time; the abnormal agent is automatically identified and positioned through the video AI intelligent identification and early warning management system, and the system automatically sends out warning information after identifying and positioning in the video image. According to the invention, by collecting abnormal behavior event data of students, automatic acousto-optic early warning, real-time information transmission and data statistics and mining of abnormal events are realized, so that a system for judging abnormal behaviors of students and quickly early warning and responding is provided, campus safety guarantee is enhanced to the maximum extent, early warning is carried out in advance, and the occurrence degree of tragedy events is reduced.
Description
Technical Field
The invention belongs to the technical field of human behavior recognition, target tracking and face recognition, and particularly relates to an intelligent recognition early warning method for abnormal behaviors.
Background
With the continuous development of the times, the quality of life of people is continuously improved, but along with the high requirements of parents on children, the psychological stress of students is continuously increased and the anxiety is continuously amplified, so that the tragic events in the campus are continuously frequently sent; the school reduces the occurrence of tragedy events, helps the children enhance the psychological bearing capacity, cultivates sound self-understanding and social understanding, and necessary preventive measures are also an important ring for campus safety management; the current campus safety management means is usually through campus monitoring system, and artifical guard, the artificial student action of looking over are hardly accomplished intelligent supervision, early warning in advance to the tragedy event that probably takes place, often treat that the tragedy event takes place the back, look over the control record and can only regard as the means of following up the responsibility, but the life is passed through this moment, in no matter mend.
Disclosure of Invention
The invention aims to provide an intelligent abnormal behavior recognition and early warning method to solve the problems in the background technology.
In order to solve the technical problems, the invention is realized by the following technical scheme:
the invention relates to an intelligent recognition early warning method for abnormal behaviors.
The method comprises the following steps:
s1: by using leading-edge information technologies such as artificial intelligence, internet of things, big data, 5G mobile internet and the like, an edge computing visual perception platform is built through a high-altitude camera network widely covering a target area, and abnormal behavior occurrence places of students are monitored and accurately positioned in real time;
s2: the abnormal behavior is automatically identified and positioned by the video AI intelligent identification and early warning management system, the system automatically sends out warning information after identifying and positioning in a video image, and the information is sent to safety management personnel in real time, so that the aim of automatically discovering the abnormal behavior of students by the system is fulfilled.
Further, the system comprises a hardware structure consisting of a high-altitude networking camera, an edge computing visual perception platform, an intelligent cloud box, a server for realizing information interaction and a real-time and rapid 5G transmission network, and a software structure consisting of target tracking, pedestrian re-recognition, trajectory analysis, face recognition and posture detection and recognition.
Further, the steps S1 and S2 are realized through an edge computing video sensing node hardware module, a lightweight deep neural network pedestrian detection module, a multi-target tracking basic module, a behavior recognition basic module, a pedestrian re-recognition module and a face recognition module;
the edge computing video perception node hardware module is used for adding edge equipment at a camera deployment point to detect and identify an agent; the communication transmission among the camera, the edge device and the server is carried out in a wireless network connection mode, so that a real-time monitoring and early warning target is achieved;
the lightweight deep neural network pedestrian detection module is used for enhancing the robustness of the model by increasing relevant characteristics and enhancing the recognition rate of people in various complex environments;
the multi-target tracking basic module is used for detecting a plurality of targets in the video and giving IDs (identification) for track tracking under the condition that the number of the targets is not known in advance; different targets have different IDs so as to realize subsequent track prediction, accurate search and other work.
Furthermore, the behavior recognition basic module is used for classifying and distinguishing behaviors through a series of convolution, pooling and full-connection operations on the input student behavior video;
the pedestrian re-identification module is used for matching the pedestrian images or videos crossing the equipment by using a computer vision algorithm, namely, giving an inquiry image and searching the same pedestrian in image libraries of different monitoring equipment;
the face recognition module is used for rapidly verifying the identity through the face.
Further, the multi-target tracking basic module comprises the following steps:
a. given the original frame of the video, the detector is run to obtain a bounding box;
b. for each detected object, different characteristics, typically visual and motion characteristics, are calculated
c. And calculating the probability of two objects belonging to the same target through the similarity calculation step, and finally distributing a digital ID to each object, so as to facilitate later inquiry.
Further, the face recognition module is composed of face detection, face alignment, face feature extraction and face recognition;
the face feature matching method is used for comparing one face feature with features corresponding to N identities registered in a library one by one to find out one feature with the highest similarity to an input feature; and comparing the highest similarity value with a preset threshold value, and returning the identity corresponding to the feature if the highest similarity value is greater than the threshold value.
Further, the step of sending the alarm information is as follows:
(1) judging the identity of a person and whether abnormal behaviors exist or not by face recognition, posture detection and recognition and combining with the psychological evaluation result of students and the behavior history records of the students, generating primary response information in real time for the abnormal events of the identity and the behaviors, and sending a specified target in a preset mode to remind the attention;
(2) judging whether the locked target has upgrading abnormal behavior through AI identification and posture detection and identification, and automatically executing a preset second-stage response scheme according to a judgment result, wherein the preset second-stage response scheme comprises alarming, informing a teacher or security personnel to go to check and the like;
(3) judging whether the abnormal behavior of the target upgrading is interfered or not through information feedback, AI identification and posture detection and identification, and automatically entering and executing a preset third-stage response scheme or a successful interference follow-up scheme according to whether the abnormal behavior of the target upgrading is interfered or not and an interference result;
(4) and evaluating whether the intervention behavior conflicts with the student position change or not through information feedback, AI identification and posture detection and identification, feeding back to the specified target in real time, and enabling the platform to automatically enter a highest-level response state.
The invention has the following beneficial effects:
according to the invention, by collecting abnormal behavior event data of students, automatic acousto-optic early warning, real-time information transmission and data statistics and mining of abnormal events are realized, so that a system for judging abnormal behaviors of students and quickly early warning response can be realized, the campus safety guarantee strength is enhanced to the maximum extent, early warning is carried out in advance, and the occurrence degree of tragedy events is reduced.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic structural view of the present invention;
FIG. 2 is a diagram of a lightweight pedestrian detection model architecture of the present invention;
FIG. 3 is a schematic diagram of multi-target tracking in accordance with the present invention;
FIG. 4 is a diagram of a multi-target tracking network architecture of the present invention;
FIG. 5 is a schematic diagram of an embodiment of a behavior recognition network according to the present invention;
FIG. 6 is a schematic diagram of pedestrian re-identification according to the present invention;
FIG. 7 is a flow chart of a face recognition scheme of the present invention;
FIG. 8 is a schematic diagram of a face recognition application of the present invention;
FIG. 9 is a schematic illustration of the operation of the present invention;
FIG. 10 is a schematic diagram of the identification of abnormal behavior segment 1 according to the present invention;
FIG. 11 is a schematic diagram of the abnormal behavior segment 2 identification according to the present invention;
FIG. 12 is a schematic diagram of the abnormal behavior segment 3 identification according to the present invention;
fig. 13 is a schematic diagram of the abnormal behavior segment 4 according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1-13, the present invention is an intelligent abnormal behavior recognition and early warning method.
In order to reduce campus tragedy events, the invention comprehensively applies the leading-edge information technologies such as artificial intelligence, internet of things, big data, 5G mobile internet and the like, develops a low-cost edge computing visual perception platform through a high-altitude camera network widely covering a target area, and monitors and accurately positions the abnormal behavior occurrence places of students in real time;
the invention mainly comprises two aspects of hardware and software, as shown in figure 1; the hardware mainly comprises a high-altitude networking camera which widely covers a target area, an edge computing visual perception platform facing low cost, namely an intelligent cloud box, a server for realizing information interaction and a real-time and rapid 5G transmission network;
the software is mainly in the aspect of AI algorithm, and the involved algorithm technology mainly comprises target tracking, pedestrian re-recognition, trajectory analysis, face recognition, posture detection and recognition and the like;
the invention has the advantages that a set of efficient real-time scheme is designed, abnormal actors are automatically identified and positioned by a video AI intelligent identification and early warning management system, alarm information is automatically sent out after the system identifies and positions in a video image, and the information is sent to safety management personnel in real time, so that the aim of automatically discovering abnormal behaviors of students by the system is fulfilled, and the system comprises the following specific points;
1. the response speed is high; the algorithm is deployed at the edge end, so that the quick response to the abnormal behavior at the first time is realized;
2. leading the algorithm; the multi-target tracking algorithm has high accuracy and efficiency, can quickly identify abnormal behaviors, and has high accuracy;
3. the deployment cost is low; the monitoring of a camera does not need to be newly arranged, and the prior monitoring equipment in the campus is utilized for real-time monitoring and early warning;
4. the method has the advantages that the method is robust in recognition of various weather and light conditions, can effectively discriminate behaviors under various conditions, and is suitable for various conditions;
edge computing video sensing node hardware module
The invention adopts edge equipment for deployment, and the edge equipment is added at a camera deployment point to carry out detection and identification support on an agent; the communication transmission among the camera, the edge device and the server is carried out in a wireless network connection mode, so that a real-time monitoring and early warning target is achieved;
(II) lightweight deep neural network pedestrian detection module
Aiming at students in a campus, a deep neural network with better effect is designed for pedestrian target detection; the model needs to solve pedestrian detection under long distance and various weather conditions, the robustness of the model is enhanced by increasing relevant characteristics, and the recognition rate of people under various complex environments is enhanced; the network is shown in FIG. 2;
(III) Multi-target tracking basic Module
Multi-target tracking, namely detecting a plurality of targets in a video and giving IDs (identity) for track tracking under the condition that the number of the targets is not known in advance; different targets have different IDs so as to realize subsequent operations of track prediction, accurate search and the like, as shown in FIG. 3; the multi-target tracking step is generally that, given the original frame of the video, the detector is operated to obtain a bounding box, then for each detected object, different features, usually visual and motion features, are calculated, then the probability that two objects belong to the same target is calculated by the similarity calculation step, and finally a digital ID is assigned to each object; the multi-target tracking network structure is shown in FIG. 4; by 5 months in 2020, the team multi-target tracking technology ranks first on the multi-target tracking data set MOT17 comprehensive target detection and tracking accuracy rate and speed index;
(IV) behavior recognition basic module
Aiming at the behaviors of students, a deep network is designed for identification, classification and judgment are carried out on the behaviors of input student behavior videos through a series of convolution, pooling and full-connection operations, and the design scheme of the behavior identification network is shown in figure 5;
(V) pedestrian re-identification module
Pedestrian re-identification, also known as pedestrian re-identification, is now viewed as a class of key sub-problems for image retrieval; the method comprises the steps that a computer vision algorithm is utilized to match images or videos of pedestrians across equipment, namely, an inquiry image is given, and the same pedestrian is searched in image libraries of different monitoring equipment, as shown in figure 6, namely, the same person needs to be found in different videos;
(VI) face recognition module
An integrated face recognition system comprises four main components, namely face detection, face alignment, face feature extraction and face recognition; at present, the face recognition technology is mature, and the process is as shown in fig. 7, wherein a face feature is input, and the feature with the highest similarity to the input feature is found out by comparing the face feature with the features corresponding to the N identities registered in the database one by one; comparing the highest similarity value with a preset threshold value, and if the highest similarity value is greater than the preset threshold value, returning the identity corresponding to the feature; at present, teams have successfully applied face recognition technology to face recognition attendance machine systems, as shown in fig. 8.
See fig. 9-13; inputting a section of video, the invention can automatically identify and position abnormal behavior (such as the abnormal behavior shown in figure 10 and the abnormal behavior shown in figure 12) through the system shown in figure 9, the system automatically sends out alarm information after identifying and positioning in a video image, and the information is sent to a safety manager in real time, thereby realizing the aim of automatically finding the abnormal behavior of students through the system; the specific process is as follows;
judging the identity of a person and the existence of abnormal behaviors by a face recognition algorithm, a posture detection algorithm and a recognition algorithm in combination with the psychological evaluation result of students and the historical record of the behaviors of the students, generating primary response information for the abnormal events of the identity and the behaviors in real time, and sending a specified target by a preset mode to remind the attention;
2. judging whether the locked target has upgrading abnormal behavior or not through an AI algorithm and a posture detection and recognition algorithm, and automatically executing a preset second-stage response scheme according to a judgment result, wherein the preset second-stage response scheme comprises an alarm, a teacher or a security guard is informed to go to check, and the like;
3. judging whether the target upgrading abnormal behavior is interfered or not through information feedback, an AI algorithm and a posture detection and recognition algorithm, and automatically entering a preset third-level response scheme or a subsequent scheme of successful interference according to whether the target upgrading abnormal behavior is interfered or not and an interference result;
4. and evaluating whether the intervention behavior conflicts with the position change of the student through information feedback, an AI algorithm and a posture detection and recognition algorithm, feeding back to a specified target in real time, and automatically entering a highest-level response state by the platform.
In the description herein, references to the description of "one embodiment," "an example," "a specific example," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The preferred embodiments of the invention disclosed above are intended to be illustrative only. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise embodiments disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best utilize the invention. The invention is limited only by the claims and their full scope and equivalents.
Claims (7)
1. An intelligent abnormal behavior identification early warning method is characterized by comprising the following steps:
s1: by applying leading-edge information technologies such as artificial intelligence, internet of things, big data, 5G mobile internet and the like, an edge computing visual perception platform is built through a high-altitude camera network covering a wide target area, and abnormal behavior places of students are monitored and accurately positioned in real time;
s2: the abnormal behavior is automatically identified and positioned by the video AI intelligent identification and early warning management system, the system automatically sends out warning information after identifying and positioning in a video image, and the information is sent to safety management personnel in real time, so that the aim of automatically discovering the abnormal behavior of students by the system is fulfilled.
2. The intelligent abnormal behavior recognition and early warning method as claimed in claim 1, wherein the method comprises a hardware structure consisting of a high-altitude networking camera, an edge computing visual perception platform, an intelligent cloud box, a server for realizing information interaction and a real-time and rapid 5G transmission network, and a software structure consisting of target tracking, pedestrian re-recognition, trajectory analysis, face recognition and gesture detection and recognition.
3. The intelligent abnormal behavior recognition and early warning method according to claim 1, wherein the steps S1 and S2 are realized by an edge computing video sensing node hardware module, a lightweight deep neural network pedestrian detection module, a multi-target tracking basic module, a behavior recognition basic module, a pedestrian re-recognition module and a face recognition module;
the edge computing video perception node hardware module is used for adding edge equipment at a camera deployment point to detect and identify an agent; the communication transmission among the camera, the edge device and the server is carried out in a wireless network connection mode, so that a real-time monitoring and early warning target is achieved;
the lightweight deep neural network pedestrian detection module is used for enhancing the robustness of the model by increasing relevant characteristics and enhancing the recognition rate of pedestrians under various complex environments;
the multi-target tracking basic module is used for detecting a plurality of targets in the video and giving IDs (identification) for track tracking under the condition that the number of the targets is not known in advance; different targets have different IDs so as to realize subsequent track prediction, accurate search and other work.
4. The intelligent abnormal behavior recognition and early warning method according to claim 3, wherein the behavior recognition basic module is used for classifying and judging behaviors through a series of convolution, pooling and full connection operations on input student behavior videos;
the pedestrian re-identification module is used for matching the pedestrian images or videos crossing the equipment by using a computer vision algorithm, namely, giving an inquiry image and searching the same pedestrian in image libraries of different monitoring equipment;
the face recognition module is used for rapidly verifying the identity through the face.
5. The intelligent abnormal behavior recognition and early warning method according to claim 3, wherein the multi-target tracking base module comprises the following steps:
a. given the original frame of the video, the detector is run to obtain a bounding box;
b. for each detected object, different characteristics, typically visual and motion characteristics, are calculated
c. And calculating the probability of two objects belonging to the same target through the similarity calculation step, and finally distributing a digital ID to each object, so as to facilitate later inquiry.
6. The intelligent abnormal behavior recognition and early warning method according to claim 4, wherein the face recognition module comprises face detection, face alignment, face feature extraction and face recognition;
the face feature recognition method is used for comparing features corresponding to N identities registered in a library one by one when inputting a face feature to find out a feature with the highest similarity to the input feature; and comparing the highest similarity value with a preset threshold value, and if the highest similarity value is greater than the preset threshold value, returning the identity corresponding to the feature.
7. The intelligent abnormal behavior identification and early warning method according to claim 2, wherein the step of sending the warning information is as follows:
(1) judging the identity of a person and whether abnormal behaviors exist or not by face recognition, posture detection and recognition and combining with the psychological evaluation result of students and the behavior history records of the students, generating primary response information in real time for the abnormal events of the identity and the behaviors, and sending a specified target in a preset mode to remind the attention;
(2) judging whether the locked target has upgrading abnormal behavior or not through AI identification and posture detection and identification, and automatically executing a preset second-stage response scheme according to a judgment result, wherein the preset second-stage response scheme comprises an alarm, a teacher or a security guard is informed to go to check, and the like;
(3) judging whether the target upgrading abnormal behavior is interfered or not through information feedback, AI identification and posture detection and identification, and automatically entering and executing a preset third-level response scheme or a subsequent successful intervention scheme according to whether the target upgrading abnormal behavior is interfered or not and the interference result;
(4) and evaluating whether the intervention behavior conflicts with the student position change or not through information feedback, AI identification and gesture detection and identification, feeding back to a specified target in real time, and enabling the platform to automatically enter a highest-level response state.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210543585.6A CN115311735A (en) | 2022-05-19 | 2022-05-19 | Intelligent recognition early warning method for abnormal behaviors |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210543585.6A CN115311735A (en) | 2022-05-19 | 2022-05-19 | Intelligent recognition early warning method for abnormal behaviors |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115311735A true CN115311735A (en) | 2022-11-08 |
Family
ID=83854815
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210543585.6A Pending CN115311735A (en) | 2022-05-19 | 2022-05-19 | Intelligent recognition early warning method for abnormal behaviors |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115311735A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116503814A (en) * | 2023-05-24 | 2023-07-28 | 北京安录国际技术有限公司 | Personnel tracking method and system for analysis |
| CN117354468A (en) * | 2023-12-04 | 2024-01-05 | 南京海汇装备科技有限公司 | Intelligent state sensing system and method based on big data |
-
2022
- 2022-05-19 CN CN202210543585.6A patent/CN115311735A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116503814A (en) * | 2023-05-24 | 2023-07-28 | 北京安录国际技术有限公司 | Personnel tracking method and system for analysis |
| CN116503814B (en) * | 2023-05-24 | 2023-10-24 | 北京安录国际技术有限公司 | Personnel tracking method and system for analysis |
| CN117354468A (en) * | 2023-12-04 | 2024-01-05 | 南京海汇装备科技有限公司 | Intelligent state sensing system and method based on big data |
| CN117354468B (en) * | 2023-12-04 | 2024-02-13 | 南京海汇装备科技有限公司 | Intelligent state sensing system and method based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Aboah | A vision-based system for traffic anomaly detection using deep learning and decision trees | |
| CN108062349B (en) | Video monitoring method and system based on video structured data and deep learning | |
| US10964031B2 (en) | Video analysis methods and apparatus | |
| Saxena et al. | Crowd behavior recognition for video surveillance | |
| Bird et al. | Real time, online detection of abandoned objects in public areas | |
| Shah et al. | Automated visual surveillance in realistic scenarios | |
| CN105184258A (en) | Target tracking method and system and staff behavior analyzing method and system | |
| CN115311735A (en) | Intelligent recognition early warning method for abnormal behaviors | |
| US20100207762A1 (en) | System and method for predicting abnormal behavior | |
| JP2018173914A (en) | Image processing system, imaging apparatus, learning model creation method, and information processing device | |
| US11893727B2 (en) | Rail feature identification system | |
| KR101900237B1 (en) | On-site judgment method using situation judgment data on acquired images | |
| CN115527340A (en) | Intelligent construction site safety monitoring system and method based on unmanned aerial vehicle and surveillance camera | |
| CN104933542B (en) | A kind of logistic storage monitoring method based on computer vision | |
| KR20200017594A (en) | Method for Recognizing and Tracking Large-scale Object using Deep learning and Multi-Agent | |
| CN110728252A (en) | Face detection method applied to regional personnel motion trail monitoring | |
| CN111553305B (en) | System and method for identifying illegal videos | |
| Brax et al. | Finding behavioural anomalies in public areas using video surveillance data | |
| CN116993265A (en) | Intelligent warehouse safety management system based on Internet of things | |
| Saluky et al. | Abandoned Object Detection Method Using Convolutional Neural Network | |
| Siddique et al. | Analysis of real-time hostile activitiy detection from spatiotemporal features using time distributed deep CNNs, RNNs and attention-based mechanisms | |
| Shuoyan et al. | Abnormal behavior detection based on the motion-changed rules | |
| KR102286229B1 (en) | A feature vector-based fight event recognition method | |
| Ho et al. | Public space behavior modeling with video and sensor analytics | |
| Brax et al. | An ensemble approach for increased anomaly detection performance in video surveillance data |
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 |