CN115100572A - System and method for analyzing abnormal behaviors of campus - Google Patents
System and method for analyzing abnormal behaviors of campus Download PDFInfo
- Publication number
- CN115100572A CN115100572A CN202210832159.4A CN202210832159A CN115100572A CN 115100572 A CN115100572 A CN 115100572A CN 202210832159 A CN202210832159 A CN 202210832159A CN 115100572 A CN115100572 A CN 115100572A
- Authority
- CN
- China
- Prior art keywords
- abnormal
- campus
- analysis
- behaviors
- video
- 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.)
- Withdrawn
Links
- 206010000117 Abnormal behaviour Diseases 0.000 title claims abstract description 68
- 238000000034 method Methods 0.000 title claims abstract description 16
- 230000002159 abnormal effect Effects 0.000 claims abstract description 122
- 238000004458 analytical method Methods 0.000 claims abstract description 103
- 238000001514 detection method Methods 0.000 claims abstract description 55
- 230000006399 behavior Effects 0.000 claims abstract description 48
- 238000012544 monitoring process Methods 0.000 claims abstract description 41
- 238000004364 calculation method Methods 0.000 claims abstract description 22
- 238000007781 pre-processing Methods 0.000 claims abstract description 17
- 230000000007 visual effect Effects 0.000 claims description 6
- 230000005856 abnormality Effects 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 2
- 238000012423 maintenance Methods 0.000 abstract description 6
- 238000013473 artificial intelligence Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 4
- 238000013475 authorization Methods 0.000 description 3
- 230000002860 competitive effect Effects 0.000 description 3
- 238000005265 energy consumption Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000007306 turnover Effects 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 1
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000009545 invasion Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000028838 turning behavior Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
-
- 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)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses a campus abnormal behavior analysis system and method, which belong to the technical field of computer vision and comprise monitoring acquisition equipment, edge computing equipment, a central storage system and a central server, wherein the central server runs an abnormal detection intelligent algorithm, analyzes campus abnormal behavior data by using the abnormal detection intelligent algorithm to obtain a specific abnormal type, and reports an abnormal behavior early warning. In order to solve the technical problems of huge cost overhead and calculation cost brought by full data detection and lack of customized strategy configuration to adapt to various complex conditions, the system and the method for analyzing abnormal campus behaviors filter most invalid data through edge calculation equipment preprocessing, reduce bandwidth occupation and calculation overhead of a central server, reduce deployment and operation and maintenance cost, and generate configuration of intelligent strategies, so that the central server can make flexible and targeted strategies for various scenes, and improve generalization and accuracy of abnormal analysis.
Description
Technical Field
The invention relates to the technical field of computer vision, in particular to a system and a method for analyzing abnormal campus behaviors.
Background
With the development of artificial intelligence technology, intelligent identification of abnormal behaviors under different scenes in various industries becomes possible, but an abnormal behavior intelligent control system facing some special scenes is still lacking nowadays, and a chinese patent with publication number CN113065500A discloses an abnormal behavior control system facing one class of special actions, which is composed of a software system and hardware equipment. The hardware equipment comprises audio and video collecting equipment, a server and a client; the software system consists of three modules of comprehensive situation and risk assessment, abnormal behavior detection and emergency plan management. The invention can complete abnormal behavior monitoring and identification, show comprehensive situation, evaluate risk and send out alarm, can rapidly cope with various emergency situations, and maintain social stability, and the patent introduces the software and hardware system flow of abnormal detection in a general way;
chinese patent publication No. CN109657626A discloses an analysis method for recognizing human body behaviors through a program, which includes a central processing unit, a camera module, a sensing module, an information transmission module, a receiving module, an information processing module, an information base, a simple behavior analysis module, a behavior comparison module, a behavior information feedback module, a terminal device, and an alarm module, wherein an output end of the camera module is electrically connected to an input end of the sensing module. The patent technology of the invention has the characteristics of high speed, stable performance, strong expandability and the like through the method, can detect and position various bad behaviors of government personnel in real time, such as leaving behind posts, sleeping, chatting, playing mobile phones and the like, and provides effective data for higher management departments in time, thereby standardizing window service, reducing lazy administration and messy administration, continuously improving the service quality of public departments, simultaneously monitoring driver behaviors in real time and reducing traffic accidents, and the patent combines human behavior analysis to provide a detection method of abnormal human body behaviors;
chinese patent publication No. CN202111453407.6 discloses a security monitoring method and system based on edge calculation, the method includes the following steps: step S10, the edge computing box acquires the encrypted authorization image set from the monitoring server and decrypts the authorization image set; step S20, the edge calculation box acquires the latest face image set which is acquired and encrypted in real time from the FTP server, and decrypts the latest face image set; step S30, the edge calculation box verifies the latest face image set by using the authorized image set based on the face algorithm and generates a verification result; step S40, the edge computing box encrypts the check result and uploads the encrypted check result to the monitoring server for safety warning; the invention has the advantages that: the security of monitoring is greatly improved, and the patent is used for decryption and authorization management of the edge device. The above patents suffer from the following drawbacks:
the huge cost and calculation cost brought by the full data detection is lack of customized strategy configuration to adapt to various complex conditions.
Disclosure of Invention
The invention aims to provide a campus abnormal behavior analysis system and method, which filter most invalid data through edge computing equipment preprocessing, reduce bandwidth occupation and central server computing overhead, reduce deployment and operation and maintenance costs, and generate configuration of an intelligent strategy, so that a central server can make flexible and targeted strategies for various scenes, and improve generalization and accuracy of abnormal analysis so as to solve the problems in the background art.
In order to achieve the purpose, the invention provides the following technical scheme:
the system for analyzing abnormal campus behaviors comprises a monitoring acquisition device, an edge computing device, a central storage system and a central server, wherein the central storage system comprises a plurality of storage units
The monitoring acquisition devices are used for acquiring audio and video data in a campus in real time, the monitoring acquisition devices are distributed in each area of the campus corresponding to the monitoring acquisition devices, the audio and video data of each area of the campus are acquired through the monitoring acquisition devices distributed in the campus, and the audio and video data acquired in real time are transmitted to the edge computing device for preprocessing;
the edge computing equipment is used for preprocessing the acquired audio and video data, running an abnormity pre-detection algorithm, pre-screening the acquired audio and video data by utilizing the abnormity pre-detection algorithm, filtering most invalid data, retaining effective data of abnormal behaviors of the campus, and transmitting the effective data of the abnormal behaviors of the campus to the central storage system;
the central storage system is used for compressing, integrating and storing the received effective data of the abnormal campus behaviors and transmitting the stored effective data of the abnormal campus behaviors to the central server for detailed analysis;
the central server is used for analyzing the effective data of the campus abnormal behaviors after the campus abnormal behaviors are detected and screened in detail, the central server runs an abnormal detection intelligent algorithm, the effective data of the campus abnormal behaviors are analyzed in detail by using the abnormal detection intelligent algorithm to obtain specific abnormal types, and the abnormal behavior early warning is reported.
Further, the edge computing device is a monitoring acquisition device itself or a micro embedded computing chip connected in the monitoring acquisition device.
Further, the edge computing device runs an anomaly pre-detection algorithm and performs the following operations:
combining campus scene, time, place and target information, and obtaining abnormal distribution behaviors by using a video frame data difference calculation method;
the abnormal frame analysis needs to be carried out on the basis of a section of continuous video frame, the length of the continuous video frame is represented by using a sliding window S, video difference calculation is carried out according to the section of continuous video frame, and abnormal frames existing in the section of continuous video are calculated, wherein the video frame difference calculation method comprises but is not limited to difference, perceptual hash, feature points and encoder-decoder;
and performing video difference analysis on the abnormal frames, determining an abnormal behavior trend graph according to the video difference degree in each sliding window, analyzing according to the abnormal behavior trend graph, finding abnormal trends based on the mathematical characteristics of a threshold or a gradient, and performing early warning based on the abnormal trends.
Further, the central server runs an intelligent algorithm for detecting the abnormality, and executes the following operations:
the central server receives audio and video data sent by each monitoring acquisition device, and the audio and video data are pre-screened by the edge computing device and have the possibility of being an abnormal event;
the central server intelligently generates an analysis strategy by using an intelligent strategy generator according to the source location, the event and the equipment ID information of the audio and video data;
calling a specific intelligent analysis method according to the generated analysis strategy, and judging whether an abnormality exists;
and reporting and early warning according to the generated abnormal analysis result.
According to another aspect of the present invention, there is provided an analysis method for a campus abnormal behavior analysis system, including the following steps:
s10: acquiring audio and video data of each area in the campus through each monitoring acquisition device distributed in the campus, and transmitting the audio and video data acquired in real time to edge computing equipment for preprocessing;
s20: the edge computing equipment runs an abnormal pre-detection algorithm, pre-screens the collected audio and video data by using the abnormal pre-detection algorithm, filters most invalid data, retains effective data of abnormal behaviors of the campus, and transmits the effective data of the abnormal behaviors of the campus to a central storage system;
s30: the central storage system compresses and integrates the received effective data of the abnormal campus behaviors, classifies and stores the compressed and integrated effective data of the abnormal campus behaviors, stores the data according to campus scenes, time, places and purposes, and transmits the stored effective data of the abnormal campus behaviors to a central server for detailed analysis;
s40: the central server runs an abnormal detection intelligent algorithm, analyzes effective data of abnormal behaviors of the campus in detail by using the abnormal detection intelligent algorithm to obtain specific abnormal types, and reports early warning of the abnormal behaviors.
Further, when performing video difference analysis on the abnormal frame, the following operations are performed:
acquiring each video frame in a section of continuous video, and repeatedly comparing the front video frame and the rear video frame by using a video comparison unit, wherein the comparison content is the correlation comparison between the audio and the image of the front video frame and the rear video frame;
aiming at the condition that the comparison of the front video frame and the back video frame is consistent, the fact that the videos in the sliding windows are not different is shown, and campus abnormal behaviors do not exist in the obtained continuous videos;
and aiming at the situation that the comparison of the front video frame and the back video frame is inconsistent, the difference of the videos in each sliding window is indicated, and the acquired continuous videos have campus abnormal behaviors.
Further, when the intelligent analysis method is called according to the generated analysis strategy, the following operations are executed:
the central server intelligently generates an analysis strategy by using an intelligent strategy generator according to the information of the source place, the event and the equipment ID of the audio and video data, calls a specific intelligent analysis algorithm according to the generated analysis strategy, wherein the intelligent strategy generator intelligently generates the analysis strategy according to the generation modes of target detection, semantic analysis and time judgment, intelligently generates the analysis strategy by adopting different generation modes aiming at different possible abnormal campus behaviors, calls a specific intelligent analysis algorithm corresponding to the analysis strategy according to the intelligently generated analysis strategy, and judges whether the abnormal campus behaviors exist or not by using the intelligent analysis algorithm.
Further, campus abnormal behaviors are analyzed and judged according to a generation mode of target detection, semantic analysis and time judgment, multi-angle analysis is integrated, an abnormal analysis result is determined, and aiming at the abnormal analysis result of the campus abnormal behaviors, the central server transmits alarm information to the audible and visual alarm through the wireless network to perform audio and visual alarm in the campus, remotely transmits information to campus managers and informs the campus managers of the existence of the campus abnormal behaviors.
Compared with the prior art, the invention has the beneficial effects that:
the system and the method for analyzing the abnormal campus behaviors provide a scheme of edge device preprocessing and central server data processing, combine with computer vision and an artificial intelligence algorithm, can greatly reduce the calculated amount on the basis of ensuring the identification accuracy, reduce the cost in the aspects of hardware resources, bandwidth resources, calculation resources and energy consumption, provide a competitive solution for campus safety monitoring, filter out most invalid data through edge computing device preprocessing, reduce bandwidth occupation and central server calculation overhead, reduce deployment and operation and maintenance costs, and generate the configuration of an intelligent strategy, so that a central server can make flexible and targeted strategies for various scenes, and improve the generalization and accuracy of abnormal analysis.
Drawings
FIG. 1 is a block diagram of an analysis system for campus abnormal behavior according to the present invention;
FIG. 2 is a flow chart of the edge computing device operation anomaly pre-detection algorithm of the present invention;
FIG. 3 is a trend graph of an edge computing device operation anomaly pre-detection algorithm of the present invention;
FIG. 4 is a flow chart of the intelligent algorithm for detecting the abnormal operation of the central server according to the present invention;
FIG. 5 is a flow chart of wall turnover anomaly detection according to the present invention;
FIG. 6 is a flow chart of campus pedestrian anomaly detection according to the present invention;
fig. 7 is a flowchart of an analysis method of the campus abnormal behavior analysis system 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.
Example one
Referring to fig. 1, an analysis system for campus abnormal behavior includes a monitoring and collecting device, an edge computing device, a central storage system and a central server, wherein
The monitoring acquisition devices are used for acquiring audio and video data in a campus in real time, the monitoring acquisition devices are distributed in each area of the campus corresponding to the monitoring acquisition devices, the audio and video data of each area of the campus are acquired through the monitoring acquisition devices distributed in the campus, and the audio and video data acquired in real time are transmitted to the edge computing device for preprocessing;
the edge computing device is used for preprocessing the acquired audio and video data, runs an abnormal pre-detection algorithm, pre-screens the acquired audio and video data by using the abnormal pre-detection algorithm, filters most invalid data, retains effective data of abnormal behaviors of the campus, and transmits the effective data of the abnormal behaviors of the campus to the central storage system, and the edge computing device is a monitoring acquisition device or a micro embedded computing chip which is arranged and connected in the monitoring acquisition device;
the central storage system is used for compressing, integrating and storing the received effective data of the abnormal campus behaviors and transmitting the stored effective data of the abnormal campus behaviors to the central server for detailed analysis;
the central server is used for analyzing the effective data of the campus abnormal behaviors after the campus abnormal behaviors are detected and screened in detail, the central server runs an abnormal detection intelligent algorithm, the effective data of the campus abnormal behaviors are analyzed in detail by using the abnormal detection intelligent algorithm to obtain specific abnormal types, and the abnormal behavior early warning is reported.
The method comprises the steps that each monitoring acquisition device distributed in a campus acquires audio and video data of each region, the edge computing device preprocesses the audio and video data acquired by the monitoring acquisition devices in real time, an abnormal pre-detection algorithm calculates and analyzes the audio and video data acquired by the monitoring acquisition devices, suspicious abnormal videos are found and reported to a central server, the abnormal pre-detection algorithm is adopted to achieve small calculated amount and strong real-time performance, the purpose of finding abnormal data distribution characteristics is achieved, specific abnormal semantic analysis is not performed, the abnormal candidate data amount after pre-screening is greatly reduced compared with the original videos, the data amount is optimized to the maximum extent, the deployment and operation and maintenance costs are greatly reduced, the central server performs detailed semantic and abnormal detection on the data filtered by the edge computing device in combination with an artificial intelligence algorithm in combination with various characteristics, and obtaining abnormal events and abnormal types, such as behaviors of walking late at night, abnormal invasion, fighting, abnormal aggregation, wall turning and abnormal access and the like, and early warning and reporting.
Referring to fig. 2-3, it should be noted that the edge computing device runs the anomaly pre-detection algorithm, and performs the following operations:
combining campus scene, time, place and target information, and obtaining abnormal distribution behaviors by using a video frame data difference calculation method;
the abnormal frame analysis needs to be carried out on the basis of a section of continuous video frame, the length of the continuous video frame is represented by using a sliding window S, video difference calculation is carried out according to the section of continuous video frame, and abnormal frames existing in the section of continuous video are calculated, wherein the video frame difference calculation method comprises but is not limited to difference, perceptual hash, feature points and encoder-decoder;
and performing video difference analysis on the abnormal frames, determining an abnormal behavior trend graph according to the video difference degree in each sliding window, analyzing according to the abnormal behavior trend graph, finding abnormal trends based on the mathematical characteristics of a threshold or a gradient, and performing early warning based on the abnormal trends.
Referring to fig. 4-6, it should be noted that the central server runs the anomaly detection intelligent algorithm, and performs the following operations:
the central server receives audio and video data sent by each monitoring acquisition device, and the audio and video data are pre-screened by the edge computing device and have the possibility of being abnormal events;
the central server intelligently generates an analysis strategy by using an intelligent strategy generator according to the source location, the event and the equipment ID information of the audio and video data;
calling a specific intelligent analysis method according to the generated analysis strategy, and judging whether an abnormality exists;
and reporting and early warning according to the generated abnormal analysis result.
Abnormal detection is carried out on the wall turning behavior, and the detection conditions are shown in table 1:
table 1: wall turnover behavior anomaly detection
| Detecting wall-turning targets | Detecting a condition |
| Detecting the crossing of the enclosure is an animal | Non-exception event |
| Detecting the crossing of the enclosure is a student | Abnormal event |
Abnormal detection is carried out on the pedestrian behaviors in the campus, and the detection conditions are shown in table 2:
table 2: campus pedestrian behavior anomaly detection
| Detecting a temporal target | Detecting a condition |
| In daytime, pedestrians appear on the lawn | Non-exception event |
| At night, pedestrians appear on the lawn | Abnormal event |
And analyzing and judging campus abnormal behaviors according to the generation modes of target detection, semantic analysis and time judgment, and determining an abnormal analysis result by integrating multi-angle analysis.
Example two
Referring to fig. 7, in order to better show the analysis process of the analysis system for abnormal campus behaviors, the present embodiment provides an analysis method of the analysis system for abnormal campus behaviors, which includes the following steps:
s10: acquiring audio and video data of respective areas in a campus through monitoring acquisition equipment distributed in the campus, and transmitting the audio and video data acquired in real time to edge computing equipment for preprocessing;
s20: the edge computing equipment runs an abnormal pre-detection algorithm, pre-screens the collected audio and video data by using the abnormal pre-detection algorithm, filters most invalid data, retains effective data of abnormal behaviors of the campus, and transmits the effective data of the abnormal behaviors of the campus to a central storage system;
s30: the central storage system compresses and integrates the received effective data of the abnormal campus behaviors, classifies and stores the compressed and integrated effective data of the abnormal campus behaviors, stores the data according to campus scenes, time, places and purposes, and transmits the stored effective data of the abnormal campus behaviors to a central server for detailed analysis;
s40: the central server runs an abnormal detection intelligent algorithm, analyzes effective data of abnormal behaviors in the campus in detail by using the abnormal detection intelligent algorithm to obtain specific abnormal types, and reports the early warning of the abnormal behaviors.
The scheme of edge device preprocessing and central server data processing is provided, the computer vision and an artificial intelligence algorithm are combined, the calculated amount can be greatly reduced on the basis of ensuring the identification accuracy, the cost is reduced in terms of hardware resources, bandwidth resources, calculation resources and energy consumption, a competitive solution is provided for campus safety monitoring, most invalid data are filtered out through edge computing device preprocessing, the bandwidth occupation and the calculation overhead of a central server are reduced, the deployment and operation and maintenance costs are reduced, and the configuration generation of intelligent strategies is realized, so that the central server can make flexible and targeted strategies for various scenes, and the generalization and accuracy of anomaly analysis are improved.
It should be noted that, when performing video difference analysis on an abnormal frame, the following operations are performed:
acquiring each video frame in a section of continuous video, and repeatedly comparing the front video frame and the rear video frame by using a video comparison unit, wherein the comparison content is the correlation comparison between the audio and the image of the front video frame and the rear video frame;
aiming at the condition that the comparison of the front video frame and the back video frame is consistent, the fact that the videos in the sliding windows are not different is shown, and campus abnormal behaviors do not exist in the obtained continuous videos;
and aiming at the situation that the comparison of the front video frame and the back video frame is inconsistent, the difference of the videos in each sliding window is indicated, and the acquired continuous videos have campus abnormal behaviors.
It should be noted that, when the intelligent analysis method is called according to the generated analysis policy, the following operations are performed:
the central server intelligently generates an analysis strategy by using an intelligent strategy generator according to the information of the source place, the event and the equipment ID of the audio and video data, calls a specific intelligent analysis algorithm according to the generated analysis strategy, wherein the intelligent strategy generator intelligently generates the analysis strategy according to the generation mode of target detection, semantic analysis and time judgment, intelligently generates the analysis strategy by adopting different generation modes aiming at different possible abnormal campus behaviors, calls a specific intelligent analysis algorithm corresponding to the analysis strategy according to the intelligently generated analysis strategy, and judges whether the abnormal campus behaviors exist or not by using the intelligent analysis algorithm.
It should be noted that, campus abnormal behaviors are analyzed and judged according to a generation mode of target detection, semantic analysis and time judgment, multi-angle analysis is integrated to determine an abnormal analysis result, and aiming at the abnormal analysis result of the campus abnormal behaviors, the central server transmits alarm information to the audible and visual alarm through the wireless network to perform audible and visual alarm in the campus, and remotely transmits information to campus managers to inform the campus managers of the existence of the campus abnormal behaviors.
In summary, the system and the method for analyzing abnormal campus behaviors provide a scheme of edge device preprocessing and central server data processing, combine with computer vision and artificial intelligence algorithm, can greatly reduce the calculated amount on the basis of ensuring the identification accuracy, reduce the cost in terms of hardware resources, bandwidth resources, calculation resources and energy consumption, provide a competitive solution for campus security monitoring, filter out most invalid data through edge computing device preprocessing, reduce bandwidth occupation and central server calculation overhead, reduce deployment and operation and maintenance costs, and generate configuration of intelligent strategies, so that the central server can make flexible and targeted strategies for various scenes, and improve the generalization and accuracy of abnormal analysis.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be able to cover the technical solutions and the inventive concepts of the present invention within the technical scope of the present invention.
Claims (8)
1. The system for analyzing abnormal campus behaviors is characterized by comprising monitoring acquisition equipment, edge computing equipment, a central storage system and a central server, wherein the monitoring acquisition equipment, the edge computing equipment, the central storage system and the central server are arranged in the campus
The monitoring acquisition devices are used for acquiring audio and video data in a campus in real time, the monitoring acquisition devices are distributed in each area of the campus corresponding to the monitoring acquisition devices, the audio and video data of each area of the campus are acquired through the monitoring acquisition devices distributed in the campus, and the audio and video data acquired in real time are transmitted to the edge computing device for preprocessing;
the edge computing equipment is used for preprocessing the acquired audio and video data, running an abnormal pre-detection algorithm, pre-screening the acquired audio and video data by using the abnormal pre-detection algorithm, filtering most invalid data, retaining effective data of abnormal campus behaviors, and transmitting the effective data of the abnormal campus behaviors to the central storage system;
the central storage system is used for compressing, integrating and storing the received effective data of the abnormal campus behaviors and transmitting the stored effective data of the abnormal campus behaviors to the central server for detailed analysis;
the central server is used for analyzing the effective data of the campus abnormal behaviors after the campus abnormal behaviors are detected and screened in detail, the central server runs an abnormal detection intelligent algorithm, the effective data of the campus abnormal behaviors are analyzed in detail by using the abnormal detection intelligent algorithm to obtain specific abnormal types, and early warning and reporting of the abnormal behaviors are carried out.
2. The system for analyzing abnormal campus behavior as claimed in claim 1, wherein the edge computing device is a monitoring and collecting device itself or a micro embedded computing chip connected to the monitoring and collecting device.
3. The system for analyzing abnormal campus behavior as claimed in claim 1, wherein said edge computing device runs an abnormal pre-detection algorithm to perform the following operations:
combining campus scene, time, place and target information, and obtaining abnormal distribution behaviors by using a video frame data difference calculation method;
abnormal frame analysis is carried out on the basis of a section of continuous video frames, the length of the continuous video frames is expressed by using a sliding window S, video difference calculation is carried out according to the section of continuous video frames, and abnormal frames existing in the section of continuous video are calculated, wherein the video frame difference calculation method comprises but is not limited to difference, perceptual hash, feature points and encoder-decoder;
and performing video difference analysis on the abnormal frames, determining an abnormal behavior trend graph according to the video difference degree in each sliding window, analyzing according to the abnormal behavior trend graph, finding abnormal trends based on the mathematical characteristics of a threshold or a gradient, and performing early warning based on the abnormal trends.
4. The system for analyzing campus abnormal behavior as claimed in claim 1, wherein said central server runs an intelligent algorithm for detecting abnormality, and performs the following operations:
the central server receives audio and video data sent by each monitoring acquisition device, and the audio and video data are pre-screened by the edge computing device and have the possibility of being an abnormal event;
the central server intelligently generates an analysis strategy by using an intelligent strategy generator according to the source location, the event and the equipment ID information of the audio and video data;
calling a specific intelligent analysis method according to the generated analysis strategy, and judging whether an abnormality exists;
and reporting and early warning according to the generated abnormal analysis result.
5. An analysis method of the campus abnormal behavior analysis system as claimed in any one of claims 1 to 4, comprising the steps of:
s10: acquiring audio and video data of each area in the campus through each monitoring acquisition device distributed in the campus, and transmitting the audio and video data acquired in real time to edge computing equipment for preprocessing;
s20: the edge computing equipment runs an abnormal pre-detection algorithm, pre-screens the collected audio and video data by using the abnormal pre-detection algorithm, filters most invalid data, retains effective data of abnormal behaviors of the campus, and transmits the effective data of the abnormal behaviors of the campus to a central storage system;
s30: the central storage system compresses and integrates the received effective data of the abnormal campus behaviors, classifies and stores the compressed and integrated effective data of the abnormal campus behaviors, stores the data according to campus scenes, time, places and purposes, and transmits the stored effective data of the abnormal campus behaviors to a central server for detailed analysis;
s40: the central server runs an abnormal detection intelligent algorithm, analyzes effective data of abnormal behaviors in the campus in detail by using the abnormal detection intelligent algorithm to obtain specific abnormal types, and reports the early warning of the abnormal behaviors.
6. The method as claimed in claim 5, wherein the following operations are performed when performing the video difference analysis on the abnormal frames:
acquiring each video frame in a section of continuous video, and repeatedly comparing the front video frame and the rear video frame by using a video comparison unit, wherein the comparison content is the correlation comparison between the audio and the image of the front video frame and the rear video frame;
aiming at the condition that the comparison of the front video frame and the back video frame is consistent, the fact that the videos in the sliding windows are not different is shown, and campus abnormal behaviors do not exist in the obtained continuous videos;
and aiming at the situation that the comparison of the front video frame and the back video frame is inconsistent, the difference of the videos in each sliding window is indicated, and the acquired continuous videos have campus abnormal behaviors.
7. The analysis method of the campus abnormal behavior analysis system as claimed in claim 6, wherein when the intelligent analysis method is called according to the generated analysis policy, the following operations are performed:
the central server intelligently generates an analysis strategy by using an intelligent strategy generator according to the information of the source place, the event and the equipment ID of the audio and video data, calls a specific intelligent analysis algorithm according to the generated analysis strategy, wherein the intelligent strategy generator intelligently generates the analysis strategy according to the generation mode of target detection, semantic analysis and time judgment, intelligently generates the analysis strategy by adopting different generation modes aiming at different possible abnormal campus behaviors, calls a specific intelligent analysis algorithm corresponding to the analysis strategy according to the intelligently generated analysis strategy, and judges whether the abnormal campus behaviors exist or not by using the intelligent analysis algorithm.
8. The method as claimed in claim 7, wherein the campus abnormal behavior is analyzed and determined according to a generation method of target detection, semantic analysis and time determination, multi-angle analysis is integrated to determine an abnormal analysis result, and for the abnormal analysis result of the campus abnormal behavior, the central server transmits alarm information to the audible and visual alarm through the wireless network, performs audible and visual alarm in the campus, and remotely transmits information to the campus administrator to inform the campus administrator of the existence of the campus abnormal behavior.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210832159.4A CN115100572A (en) | 2022-07-14 | 2022-07-14 | System and method for analyzing abnormal behaviors of campus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210832159.4A CN115100572A (en) | 2022-07-14 | 2022-07-14 | System and method for analyzing abnormal behaviors of campus |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115100572A true CN115100572A (en) | 2022-09-23 |
Family
ID=83299520
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210832159.4A Withdrawn CN115100572A (en) | 2022-07-14 | 2022-07-14 | System and method for analyzing abnormal behaviors of campus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115100572A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115884006A (en) * | 2023-02-23 | 2023-03-31 | 启实(烟台)数据技术有限公司 | Campus security prevention and control system and method based on AIoT |
| CN118368550A (en) * | 2024-06-19 | 2024-07-19 | 杭州奥克光电设备有限公司 | Operation and maintenance method and system based on intelligent optical transmission equipment of Internet of things |
-
2022
- 2022-07-14 CN CN202210832159.4A patent/CN115100572A/en not_active Withdrawn
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115884006A (en) * | 2023-02-23 | 2023-03-31 | 启实(烟台)数据技术有限公司 | Campus security prevention and control system and method based on AIoT |
| CN118368550A (en) * | 2024-06-19 | 2024-07-19 | 杭州奥克光电设备有限公司 | Operation and maintenance method and system based on intelligent optical transmission equipment of Internet of things |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110428522B (en) | Intelligent security system of wisdom new town | |
| CN115100572A (en) | System and method for analyzing abnormal behaviors of campus | |
| CN211293956U (en) | AI-based identification and alarm system for abnormal agent on construction site | |
| AU2009243916B2 (en) | A system and method for electronic surveillance | |
| CN104581081B (en) | Passenger flow analysing method based on video information | |
| CN107818312A (en) | A kind of embedded system based on abnormal behaviour identification | |
| JP2013131153A (en) | Autonomous crime prevention warning system and autonomous crime prevention warning method | |
| WO2019089441A1 (en) | Exclusion zone in video analytics | |
| CN113011833A (en) | Safety management method and device for construction site, computer equipment and storage medium | |
| CN111402532A (en) | Comprehensive security video management control system | |
| CN114418388A (en) | Personnel management method and device based on image recognition and storage medium | |
| CN110188617A (en) | Intelligent monitoring method and system for machine room | |
| KR102182660B1 (en) | System and method for detecting violence using spatiotemporal features and computer readable recording media storing program for executing method thereof | |
| CN116129490A (en) | Monitoring device and monitoring method for complex environment behavior recognition | |
| KR20230167549A (en) | The apparatus and method of monitoring cctv with control moudule | |
| KR20210043960A (en) | Behavior Recognition Based Safety Monitoring System and Method using Artificial Intelligence Technology and IoT | |
| CN117830053A (en) | Perimeter security alarm system and method | |
| CN111241918B (en) | Vehicle tracking prevention method and system based on face recognition | |
| KR20130047131A (en) | Method and system for surveilling contents of surveillance using mobile terminal | |
| Hubner et al. | Audio-video sensor fusion for the detection of security critical events in public spaces | |
| KR20100077662A (en) | Inteligent video surveillance system and method thereof | |
| KR101453386B1 (en) | Vehicle Intelligent Search System and Operating Method thereof | |
| CN113903003B (en) | Event occurrence probability determination method, storage medium, and electronic apparatus | |
| KR102313283B1 (en) | A artificial intelligence multipurpose emergency bell system with video analysis function | |
| CN114281656A (en) | Intelligent central control system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20220923 |
|
| WW01 | Invention patent application withdrawn after publication |