Video detection system based on duty command platform

By using image feature recognition module and edge server support module in the video detection system, combining high-definition thermal imaging monitoring and SIFT operators, the problems of complex detection and difficulty in detecting behaviors of human bodies and objects in the prior art are solved, and high accuracy is achieved and low-cost multi-regional testing have improved the capture rate of target personnel and dangerous items, ensuring the safety of life and property.

CN119091342APending Publication Date: 2024-12-06SHENZHEN BITNET INFORMATION TECH CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410258665.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-03-07
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The existing video detection system is not very intelligent in the behavioral detection of human bodies and objects, and cannot effectively screen and judge human bodies and objects, and cannot provide auxiliary command based on the risk level and position level, resulting in unreasonable human resources dispatch.

Method used

Design a video detection system based on the job command platform, adopts image feature recognition module and edge server support module, and obtains infrared characteristics of human bodies and objects through high-definition thermal imaging monitoring, and pre-stores the contours and infrared of target personnel and target hazardous products of different hazard levels. Features: Calculate the similarity ratio value to judge the behavioral situation, and calculate infrared feature information through the SIFT operator to realize the systematic management of behavior detection of human bodies and objects in multiple regions.

Benefits of technology

It improves the accuracy and capture rate of human body and objects behavior detection, reduces labor costs, simplifies the operation and analysis process, and ensures the accuracy and safety of multi-regional inspections.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119091342A_ABST
    Figure CN119091342A_ABST
Patent Text Reader

Abstract

The invention discloses a video detection system based on a duty command platform. The video detection system comprises an image feature recognition module and an edge server support module, the image feature recognition module is provided with human image and object image detection groups, prestores and sets contours and infrared features of target persons and target hazard products of different hazard levels, and obtains the contours and infrared features of current human bodies and objects through high-definition thermal imaging monitoring. Comparing the current human body and article contours with the infrared features and comparing the contours of the target personnel and the target hazardous articles with the infrared features to obtain a similarity proportion value, and comparing a preset similarity threshold value with the similarity proportion value to obtain a judgment result whether the working personnel take actions or not. Early warning prompting is carried out on the human body and article contours and infrared feature information; according to the invention, the problems of complex detection and difficult identification of the current human body and article behavior detection method are solved, and the police efficiency is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of video detection, and in particular to a video detection system based on a job command platform. Background Art

[0002] Smart positions, the position system of relevant departments "is based on the Internet, Internet of Things, computing, intelligent engines, video technology, data mining, knowledge management and other technical support. It is a development strategy to strengthen the management of relevant departments based on science and technology, making full use of cutting-edge technologies such as cloud technology, video Internet of Things, big data and video intelligent analysis technology, GIS, GPS, 3 / 4G, mobile position intelligence system, digital cluster, etc., through the project-level design for actual combat and serving the people, as well as the construction of a high sharing mechanism for position information, a strong integration mechanism for the system, and a deep analysis mechanism for applications. With the informatization of relevant departments as the core, through the interconnection, Internet of Things, and intelligent methods, it promotes the high integration and coordinated operation of various functional modules of the relevant department system, and realizes the new concept and new model of position development with the goal of "strong integration, high sharing, and deep application" of position information.

[0003] The job command platform (which can be called a command center if the management department has a larger jurisdiction) is a command center that gathers data and information related to the jobs in the jurisdiction, responds to various situations, and coordinates human resources.

[0004] A system has emerged in the prior art, which is implemented by setting up a network communication module, a server, a data interaction module, a support module and a user service module. However, the system with the above structure only manages the patrol of staff accordingly, and the degree of intelligence is not high. It is unable to screen and judge human bodies and objects under the video screen, and the classification of job levels, so it is unable to provide assistance and reference for command according to the danger level and job level, and it is unable to reasonably dispatch human resources.

[0005] Based on this, it is necessary to propose a video detection system based on a job command platform to solve the above problems. Summary of the invention

[0006] The present invention provides a video detection system based on a job command platform, aiming to solve the problems of complex detection, difficult identification, and inability to systematically manage existing in current human and object behavior detection methods.

[0007] To achieve the above-mentioned purpose, the present invention provides a video detection system based on a job command platform, including an image feature recognition module and an edge server support module; the image feature recognition module is provided with a portrait and an object image detection group, the image feature recognition module pre-stores the contours and infrared features of target personnel and target hazardous articles of different hazard levels, obtains the current human body and object contours and infrared features through high-definition thermal imaging monitoring, compares the current human body and object contours and infrared features with the contours and infrared features of target personnel and target hazardous articles of different hazard levels to obtain a similarity ratio value, obtains a judgment result on whether relevant departments take action by comparing a preset similarity threshold and a similarity ratio value, and issues an early warning prompt for the human body and object contour and infrared feature information, the human body and object contour and infrared feature information including the judgment result on whether relevant departments take action, the contours and infrared features of target personnel and target hazardous articles of different hazard levels and similarity ratio values; the edge server support module obtains and calculates the human body and object contour and infrared feature information by using a SIFT operator.

[0008] Compared with the prior art, the present invention discloses a video detection system based on a job command platform, which realizes system management of human and object behavior detection in multiple regions by setting an image feature recognition module and an edge server support module. The image feature recognition module obtains the current human and object contours and infrared features through high-definition thermal imaging monitoring, compares the current human and object contours and infrared features with the contours and infrared features of target personnel and target hazardous objects of different hazard levels, and compares them with the preset similarity threshold to obtain the judgment result of whether the relevant departments should take action. Since the thermal imaging signal is not amplified during the evaluation process, the thermal imaging signal is relatively weak and easily affected by danger signals, so danger signals are prone to lead to inaccurate recognition when using thermal imaging recognition. The present invention sets an image feature recognition module in a portrait and object detection group, and pre-stores the contours and infrared features of target persons and target hazardous articles of different hazard levels in the image feature recognition module, and obtains the current human body and object through high-definition thermal imaging monitoring. Since high-definition thermal imaging monitoring uses thermal imaging signals for evaluation, it is easily affected by dangerous signals. Therefore, if there is human body and object behavior detection in the portrait and object detection group, the current human body and object contours and infrared features obtained through high-definition thermal imaging monitoring will have a large deviation from the pre-stored contours and infrared features of target persons and target hazardous articles of different hazard levels. Therefore, the present invention can determine the current human body and object behavior detection situation of the portrait and object detection group by judging the similarity ratio between the current human body and object contours and infrared features and the contours and infrared features of target persons and target hazardous articles of different hazard levels. In the present invention, human bodies and objects in the portrait and object detection group are used as test signals, and different hazard levels are detected. The contours and infrared features of the target personnel and target hazardous objects are equivalent to the original test signals, the current contours and infrared features of the human body and the object obtained through the thermal imaging signal are equivalent to the received test signals, and the comparative analysis of the test signals is achieved by comparing the current contours and infrared features of the human body and the object with the pre-stored contours and infrared features of the target personnel and target hazardous objects of different hazard levels. The present invention does not need to increase the hardware cost, but uses the existing hardware facilities. The current contours and infrared features of the human body and the object are directly obtained by the image feature recognition module through high-definition thermal imaging monitoring, which reduces the labor cost and is simple to operate. The human body and object behavior detection situation can be judged by calculating the distance between the two contours and infrared features, and the analysis is simple. Moreover, since the present invention judges the human body and object behavior detection situation through the contours and infrared features of the human body and the object, the human body and object information is obtained during the human body and object behavior detection, and there is no problem of difficult identification of human body and object behavior detection.In summary, the video detection system based on the job command platform provided by the present invention solves the problems of complex detection, difficult identification, and inability to system management in current human and object behavior detection methods, and provides a human and object behavior detection system with simple structure, simple operation, simple analysis, high detection accuracy, and multi-regional detection system analysis, which improves the capture rate of target personnel and dangerous goods and ensures the safety of people’s lives and property.

[0009] The present invention also provides a video detection system based on a job command platform, comprising the following steps:

[0010] S1, setting an image feature recognition module in all human image and object image detection groups, and pre-storing the contours and infrared features of target personnel and target hazardous objects of different hazard levels in the image feature recognition module;

[0011] S2, the image feature recognition module obtains the current human body and object contours and infrared features through high-definition thermal imaging monitoring, and compares the current human body and object contours and infrared features with the contours and infrared features of target personnel and target hazardous objects of different hazard levels to obtain a similarity ratio value;

[0012] S3, the image feature recognition module compares the preset similarity threshold and the similarity ratio to obtain the result of whether the relevant departments should take action, and issues an early warning prompt for the contours and infrared feature information of the human body and the object, wherein the contours and infrared feature information of the human body and the object include the result of whether the relevant departments should take action, the contours and infrared features of target persons and target hazardous objects of different hazard levels, and the similarity ratio values;

[0013] S4, the edge server support module obtains and uses the SIFT operator to calculate the contours and infrared feature information of the human body and objects.

[0014] Compared with the prior art, the method for detecting human and object behaviors based on high-definition thermal imaging monitoring disclosed in the present invention realizes system management of human and object behavior detection situations in multiple regions through an image feature recognition module and an edge server support module. The image feature recognition module obtains the current human and object contours and infrared features through high-definition thermal imaging monitoring, compares the current human and object contours and infrared features with the contours and infrared features of target personnel and target hazardous objects of different hazard levels, and compares them with a preset similarity threshold to obtain a judgment result on whether the relevant departments should take action. Since the thermal imaging signal is not amplified during the evaluation process, the thermal imaging signal is relatively weak and easily affected by danger signals, so danger signals are prone to lead to inaccurate recognition when using thermal imaging recognition. The present invention sets an image feature recognition module in a portrait and object detection group, and pre-stores the contours and infrared features of target persons and target hazardous articles of different hazard levels in the image feature recognition module, and obtains the current human body and object through high-definition thermal imaging monitoring. Since high-definition thermal imaging monitoring uses thermal imaging signals for evaluation, it is easily affected by dangerous signals. Therefore, if there are abnormalities in the human body and object behavior detection in the portrait and object detection group, the current human body and object contours and infrared features obtained through high-definition thermal imaging monitoring will have a large deviation from the pre-stored contours and infrared features of target persons and target hazardous articles of different hazard levels. Therefore, the present invention can determine the current human body and object behavior detection situation of the portrait and object detection group by judging the similarity ratio between the current human body and object contours and infrared features and the contours and infrared features of target persons and target hazardous articles of different hazard levels. In the present invention, human bodies and objects in the portrait and object detection group are used as test signals, and different hazard The contours and infrared features of target personnel and target hazardous objects of different hazard levels are equivalent to the original test signals, and the current contours and infrared features of the human body and objects obtained through the thermal imaging signals are equivalent to the received test signals. By comparing the current contours and infrared features of the human body and objects with the pre-stored contours and infrared features of target personnel and target hazardous objects of different hazard levels, the comparative analysis of the test signals is realized. The present invention does not need to increase the hardware cost, but uses the existing hardware facilities. The current contours and infrared features of the human body and objects are directly obtained by the image feature recognition module through high-definition thermal imaging monitoring, which reduces the labor cost and is simple to operate. The human body and object behavior detection situation can be judged by calculating the distance between the two contours and infrared features, and the analysis is simple. Moreover, since the present invention judges the human body and object behavior detection situation through the contours and infrared features of the human body and objects, the human body and object information is obtained during the human body and object behavior detection, and there is no problem of difficult identification of human body and object behavior detection.In summary, the human body and object behavior detection method based on high-definition thermal imaging monitoring provided by the present invention solves the problems of complex detection, difficult identification, and inability to system management in current human body and object behavior detection methods, and provides a human body and object behavior detection method with simple structure, simple operation, simple analysis, high detection accuracy, and capable of multi-regional detection system analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a system composition diagram of the present invention;

[0016] Figure 2 is a first flow chart of the operation of the system of the present invention;

[0017] Figure 3 It is the second flow chart of the operation of the system of the present invention. DETAILED DESCRIPTION

[0018] Example 1

[0019] like Figure 1 As shown, a video detection system based on a job command platform described in the present invention includes an image feature recognition module and an edge server support module; the image feature recognition module is provided with a portrait and an object image detection group, the image feature recognition module pre-stores the contours and infrared features of target personnel and target hazardous articles of different hazard levels, obtains the current contours and infrared features of a human body and an object through high-definition thermal imaging monitoring, compares the current contours and infrared features of a human body and an object with the contours and infrared features of target personnel and target hazardous articles of different hazard levels to obtain a similarity ratio value, obtains a judgment result on whether relevant departments take action by comparing a preset similarity threshold and a similarity ratio value, and issues an early warning prompt for the contours and infrared feature information of a human body and an object, the contours and infrared feature information of a human body and an object includes a judgment result on whether the staff takes action, the contours and infrared features of target personnel and target hazardous articles of different hazard levels, and the similarity ratio values; the edge server support module obtains and calculates the contours and infrared feature information of a human body and an object by using a SIFT operator.

[0020] Compared with the prior art, the present invention discloses a video detection system based on a job command platform, which realizes system management of human and object behavior detection in multiple regions by setting an image feature recognition module and an edge server support module. The image feature recognition module obtains the current human and object contours and infrared features through high-definition thermal imaging monitoring, compares the current human and object contours and infrared features with the contours and infrared features of target personnel and target hazardous objects of different hazard levels, and compares them with the preset similarity threshold to obtain the judgment result of whether the relevant departments should take action. Since the thermal imaging signal is not amplified during the evaluation process, the thermal imaging signal is relatively weak and easily affected by danger signals, so danger signals are prone to lead to inaccurate recognition when using thermal imaging recognition. The present invention sets an image feature recognition module in a portrait and object detection group, and pre-stores the contours and infrared features of target persons and target hazardous articles of different hazard levels in the image feature recognition module, and obtains the current human body and object through high-definition thermal imaging monitoring. Since high-definition thermal imaging monitoring uses thermal imaging signals for evaluation, it is easily affected by dangerous signals. Therefore, if there is human body and object behavior detection in the portrait and object detection group, the current human body and object contours and infrared features obtained through high-definition thermal imaging monitoring will have a large deviation from the pre-stored contours and infrared features of target persons and target hazardous articles of different hazard levels. Therefore, the present invention can determine the current human body and object behavior detection situation of the portrait and object detection group by judging the similarity ratio between the current human body and object contours and infrared features and the contours and infrared features of target persons and target hazardous articles of different hazard levels. In the present invention, human bodies and objects in the portrait and object detection group are used as test signals, and different hazard levels are detected. The contours and infrared features of the target personnel and target hazardous objects are equivalent to the original test signals, the current contours and infrared features of the human body and the object obtained through the thermal imaging signal are equivalent to the received test signals, and the comparative analysis of the test signals is achieved by comparing the current contours and infrared features of the human body and the object with the pre-stored contours and infrared features of the target personnel and target hazardous objects of different hazard levels. The present invention does not need to increase the hardware cost, but uses the existing hardware facilities. The current contours and infrared features of the human body and the object are directly obtained by the image feature recognition module through high-definition thermal imaging monitoring, which reduces the labor cost and is simple to operate. The human body and object behavior detection situation can be judged by calculating the distance between the two contours and infrared features, and the analysis is simple. Moreover, since the present invention judges the human body and object behavior detection situation through the contours and infrared features of the human body and the object, the human body and object information is obtained during the human body and object behavior detection, and there is no problem of difficult identification of human body and object behavior detection.In summary, the video detection system based on the job command platform provided by the present invention solves the problems of complex detection, difficult identification, and inability to system management in current human and object behavior detection methods, and provides a human and object behavior detection system with simple structure, simple operation, simple analysis, high detection accuracy, and multi-regional detection system analysis, which improves the capture rate of target personnel and dangerous goods and ensures the safety of people’s lives and property.

[0021] like Figure 1 As shown, the image feature recognition module includes a high-definition thermal imaging monitoring component, an image denoising and comparison component and a danger signal warning component. The high-definition thermal imaging monitoring component obtains the current human body and object contours and infrared features through the thermal imaging system, and sends them to the image denoising and comparison component. The image denoising and comparison component receives the current human body and object contours and infrared features and compares them with the pre-stored contours and infrared features of target personnel and target hazardous objects of different hazard levels, calculates the similarity ratio value between the two, and compares the similarity ratio value with a preset similarity threshold. When the similarity ratio value is greater than the preset similarity threshold, the result of the judgment of whether the relevant department takes action is the current human body and object danger signal. The danger signal warning component receives the similarity ratio value sent by the main control component and the judgment result of whether the staff takes action, and will issue a warning prompt for the received human body and object contour and infrared feature information. The image denoising and comparison component stores standard contours and infrared features of human bodies and objects, and the high-definition thermal imaging monitoring component obtains the current contours and infrared features of human bodies and objects. The preset similarity threshold is the similarity ratio value between the current contours and infrared features of human bodies and objects to be detected without human body and object behavior detection and the contours and infrared features of target persons and target hazardous objects of different hazard levels. The image feature recognition module can determine the current human body and object behavior detection situation of the human image and object image detection group by judging the similarity ratio value between the current human body and object contours and infrared features and the contours and infrared features of target personnel and target hazardous objects of different hazard levels. There is no need to set up a test signal transceiver. The existing hardware facilities are used to directly obtain the current human body and object contours and infrared features through high-definition thermal imaging monitoring by the image feature recognition module, thereby reducing labor costs and being simple to operate. The human body and object behavior detection situation can be determined by calculating the distance between the two contours and infrared features, and the analysis is simple. Moreover, since the present invention determines the human body and object behavior detection situation through the human body and object contours and infrared features, the human body and object information is obtained during the human body and object behavior detection, and there is no problem of difficulty in identifying the human body and object behavior detection. The human body and object contours and infrared feature information are sent through the danger signal warning component, so that the edge server support module can obtain the human body and object contours and infrared feature information, thereby realizing multi-region detection system analysis and solving the problems existing in the current human body and object behavior detection detection method.

[0022] like Figure 1 As shown, the high-definition thermal imaging monitoring component includes a thermal imaging interference recognition module, which works under the control of the image denoising and contrast component. The thermal imaging interference recognition module reduces the impact of intentional occlusion of human bodies and objects on high-definition thermal imaging monitoring. The thermal imaging interference recognition module can recognize the human body and object behaviors of the high-definition thermal imaging monitoring component. Since the thermal imaging interference recognition module recognizes the human body and object behaviors of the recognition component, the recognition effect of the high-definition thermal imaging monitoring component can be improved under the action of human body and object behavior recognition. Therefore, when the recognition effect of the thermal imaging interference recognition module is increased, the recognition effect of the high-definition thermal imaging monitoring component will not be greatly affected under the detection of larger human body and object behaviors. Therefore, the number of human bodies and objects detected by the image feature recognition module for human body and object behaviors is reduced, which can be applied to the situation with a certain tolerance for human body and object behavior detection. Conversely, when the recognition effect of the thermal imaging interference recognition module is reduced, the recognition effect of the high-definition thermal imaging monitoring component will also be greatly affected under the detection of smaller human body and object behaviors. Therefore, the number of human bodies and objects detected by the image feature recognition module for human body and object behaviors is increased, which can be applied to the situation with a certain tolerance for human body and object behavior detection. Conversely, when the recognition effect of the thermal imaging interference recognition module is reduced, the recognition effect of the high-definition thermal imaging monitoring component will also be greatly affected under the detection of smaller human body and object behaviors. Therefore, the number of human bodies and objects detected by the image feature recognition module for human body and object behaviors is increased, which can be applied to the situation with a high frequency of human body and object behavior detection. In the present invention, the edge server support module can control the recognition effect of the thermal imaging interference recognition module by issuing commands, thereby controlling the number of human beings and objects detected by the image feature recognition module for human and object behaviors, so that the system is suitable for various human and object detection environments.

[0023] like Figure 1 As shown, the danger signal warning component includes a thermal imaging image data evaluation component, and the thermal imaging image data evaluation component sends the judgment result of whether the staff has taken action, the similarity ratio value, and the contours and infrared characteristics of target personnel and target hazardous items of different hazard levels to the thermal imaging system, and the edge server support module receives relevant information through the thermal imaging system. The thermal imaging system independently developed by my country has broken through the technical limitation that the ground service station can only receive thermal imaging signals but cannot realize the two-way information transmission of thermal imaging in traditional thermal imaging data evaluation, and realizes the two-way data evaluation of thermal imaging and supports the sending of images. The present invention utilizes the image sending function of the thermal imaging system, and by setting the thermal imaging image data evaluation component, uses the thermal imaging system to realize the two-way data evaluation between the image feature recognition module and the edge server support module, without setting other data evaluation methods, making the data evaluation between the systems simpler and more convenient.

[0024] like Figure 1As shown, the video detection system based on the job command platform described in the present invention also includes an artificial intelligence service platform, the edge server support module analyzes the obtained judgment result of whether the staff takes action, the similarity ratio value and the contours and infrared characteristics of the target personnel and target hazardous objects of different hazard levels, generates a human and object behavior detection danger curve and sends it to the designated artificial intelligence service platform; the artificial intelligence service platform sends a request to the edge server support module to obtain the human and object contours and infrared characteristic information of the human image and object image detection group. The present invention generates a human and object behavior detection danger curve through the edge server support module, and sends the human and object behavior detection danger curve to the designated artificial intelligence service platform. The designated artificial intelligence service platform described in the present invention refers to the artificial intelligence service platform with management obligations for the human and object behavior detection area, such as the danger signal of the human image and object image detection group. The edge server support module sends the human and object behavior detection danger curve to the artificial intelligence service platform, and the relevant job personnel can receive the human and object behavior detection danger curve in time to confirm the current status of the human image and object image detection group and make corresponding feedback. At the same time, the artificial intelligence service platform can actively access the edge server support module to obtain the human and object behavior detection information of all or specified human and object detection groups that they want to know, so that staff can monitor relevant areas more proactively.

[0025] like Figure 1 As shown, the video detection system based on the job command platform described in the present invention also includes a human and object behavior recognition module, which receives commands sent by the server to recognize the human and object behaviors in the current monitoring area. The present invention improves the human and object behavior detection status in the area by setting up a human and object behavior recognition module, can respond quickly under the control of the edge server support module, improve regional security in time, avoid danger caused by human and object behavior detection, and can also improve the efficiency of staff actions when serious dangers occur in human and object behavior detection.

[0026] Example 2

[0027] like Figure 2As shown, the video detection system based on the job command platform described in the present invention includes the following steps: S1, setting an image feature recognition module in all human and object image detection groups, and pre-storing the contours and infrared features of the current target personnel and target hazardous objects of different hazard levels in the image feature recognition module; S2, the image feature recognition module obtains the current human body and object contours and infrared features through high-definition thermal imaging monitoring, and compares the current human body and object contours and infrared features with the contours and infrared features of target personnel and target hazardous objects of different hazard levels to obtain a similarity ratio value; S3, the image feature recognition module compares the preset similarity threshold and the similarity ratio value to obtain a judgment result on whether the staff takes action, and issues an early warning prompt for the human body and object contour and infrared feature information, wherein the human body and object contour and infrared feature information includes the judgment result on whether the staff takes action, the contours and infrared features of target personnel and target hazardous objects of different hazard levels, and the similarity ratio value; S4, the edge server support module obtains and calculates the human body and object contour and infrared feature information by SIFT operator.

[0028] Compared with the prior art, the method for detecting human and object behaviors based on high-definition thermal imaging monitoring disclosed in the present invention realizes system management of human and object behavior detection situations in multiple regions through an image feature recognition module and an edge server support module. The image feature recognition module obtains the current human and object contours and infrared features through high-definition thermal imaging monitoring, compares the current human and object contours and infrared features with the contours and infrared features of target personnel and target hazardous objects of different hazard levels, and compares them with a preset similarity threshold to obtain a judgment result on whether the staff has taken action. Since the thermal imaging signal is not amplified during the evaluation process, the thermal imaging signal is relatively weak and easily affected by danger signals, so danger signals are prone to lead to inaccurate recognition when using thermal imaging recognition. The present invention sets an image feature recognition module in a portrait and object detection group, and pre-stores the contours and infrared features of target persons and target hazardous articles of different hazard levels in the image feature recognition module, and obtains the current human body and object through high-definition thermal imaging monitoring. Since high-definition thermal imaging monitoring uses thermal imaging signals for evaluation, it is easily affected by dangerous signals. Therefore, if there is human body and object behavior detection in the portrait and object detection group, the current human body and object contours and infrared features obtained through high-definition thermal imaging monitoring will have a large deviation from the pre-stored contours and infrared features of target persons and target hazardous articles of different hazard levels. Therefore, the present invention can determine the current human body and object behavior detection situation of the portrait and object detection group by judging the similarity ratio between the current human body and object contours and infrared features and the contours and infrared features of target persons and target hazardous articles of different hazard levels. In the present invention, human bodies and objects in the portrait and object detection group are used as test signals, and different hazard levels are detected. The contours and infrared features of the target personnel and target hazardous objects are equivalent to the original test signals, the current contours and infrared features of the human body and the object obtained through the thermal imaging signal are equivalent to the received test signals, and the comparative analysis of the test signals is achieved by comparing the current contours and infrared features of the human body and the object with the pre-stored contours and infrared features of the target personnel and target hazardous objects of different hazard levels. The present invention does not need to increase the hardware cost, but uses the existing hardware facilities. The current contours and infrared features of the human body and the object are directly obtained by the image feature recognition module through high-definition thermal imaging monitoring, which reduces the labor cost and is simple to operate. The human body and object behavior detection situation can be judged by calculating the distance between the two contours and infrared features, and the analysis is simple. Moreover, since the present invention judges the human body and object behavior detection situation through the contours and infrared features of the human body and the object, the human body and object information is obtained during the human body and object behavior detection, and there is no problem of difficult identification of human body and object behavior detection.In summary, the human body and object behavior detection method based on high-definition thermal imaging monitoring provided by the present invention solves the problems of complex detection, difficult identification, and inability to system management in current human body and object behavior detection methods, and provides a human body and object behavior detection method with simple structure, simple operation, simple analysis, high detection accuracy, and capable of multi-regional detection system analysis.

[0029] The video detection system based on the job command platform described in the present invention also includes step S5, in which the edge server support module obtains the human body and object contour and infrared feature information, generates a human body and object behavior detection danger curve based on the human body and object contour and infrared feature information, and determines the human body and object in the human body and object behavior detection area through the contours and infrared features of target personnel and target hazardous objects of different hazard levels, and sends the human body and object behavior detection danger curve to the artificial intelligence service platform that manages the human body and object behavior detection area. The present invention generates the human body and object behavior detection danger curve through the edge server support module, and sends the human body and object behavior detection danger curve to the designated artificial intelligence service platform. The designated artificial intelligence service platform described in the present invention refers to an artificial intelligence service platform that has management obligations for the human body and object behavior detection area, such as a danger signal of a human image and object image detection group. The edge server support module sends the human body and object behavior detection danger curve to the artificial intelligence service platform, and the relevant job personnel can receive the human body and object behavior detection danger curve in time to confirm the current status of the human image and object image detection group and make a corresponding response. At the same time, the artificial intelligence service platform can actively access the edge server support module to obtain the human and object behavior detection information of all or specified human and object detection groups that they want to know, so that staff can monitor relevant areas more proactively.

[0030] Example 3

[0031] like Figure 3As shown, a method for detecting human and object behaviors described in the present invention also includes S6 human and object quantity verification step, wherein the human and object quantity verification step is that the edge server support module sends a thermal imaging interference recognition command to the image feature recognition module, and the image feature recognition module verifies the human and object behavior detection frequency of the thermal imaging interference recognition module. The present invention recognizes the human and object behaviors received by the high-definition thermal imaging monitoring component through the thermal imaging interference recognition module. Since the thermal imaging interference recognition module recognizes the human and object behaviors of the recognition component, the recognition effect of the high-definition thermal imaging monitoring component can be improved under the action of human and object behavior recognition. Therefore, when the recognition effect of the thermal imaging interference recognition module is increased, the recognition effect of the high-definition thermal imaging monitoring component will not be greatly affected under the larger human and object behavior detection. Therefore, the number of human and object detected by the image feature recognition module for human and object behaviors is reduced, which can be applied to the situation where there are certain abnormalities in the detection of human and object behaviors. On the contrary, when the recognition effect of the thermal imaging interference recognition module is reduced, the recognition effect of the high-definition thermal imaging monitoring component will also be greatly affected under smaller human and object behavior detection, so the number of human beings and objects detected by the image feature recognition module for human and object behavior increases, which can be applied to situations where the frequency of human and object behavior detection is high. In the present invention, the edge server support module can control the recognition effect of the thermal imaging interference recognition module by issuing commands, thereby controlling the number of human beings and objects detected by the image feature recognition module for human and object behavior, so that the system is applicable to various human and object detection environments.

[0032] A method for detecting human and object behaviors described in the present invention also includes an S7 human and object behavior recognition step, wherein the human and object behavior recognition step is that the edge server support module sends a human and object behavior recognition command to the human and object behavior recognition module, and the human and object behavior recognition module recognizes the human and object behaviors in the current monitoring area according to the received command. The present invention sets a human and object behavior recognition module to improve the human and object behavior detection status in the area, and can respond quickly under the control of the edge server support module, timely improve regional security, avoid risks caused by abnormal human and object behavior detection, and can also improve the efficiency of staff actions when serious human and object behavior detection is dangerous.

[0033] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A video detection system based on a job command platform, characterized in that: It includes an image feature recognition module and an edge server support module; the image feature recognition module is provided with a human image and an object image detection group, the image feature recognition module pre-stores the contours and infrared features of target personnel and target hazardous articles of different hazard levels, obtains the current contours and infrared features of a human body and an object through high-definition thermal imaging monitoring, compares the current contours and infrared features of a human body and an object with the contours and infrared features of target personnel and target hazardous articles of different hazard levels to obtain a similarity ratio value, obtains a judgment result on whether the staff has taken action by comparing a preset similarity threshold and a similarity ratio value, and issues an early warning prompt for the contours and infrared feature information of the human body and an object, the contours and infrared feature information of the human body and an object include the judgment result on whether the staff has taken action, the contours and infrared features of target personnel and target hazardous articles of different hazard levels, and the similarity ratio values; The edge server support module obtains and uses the SIFT operator to calculate the contour and infrared feature information of the human body and objects.

2. A video detection system based on a job command platform according to claim 1, characterized in that: The image feature recognition module includes a high-definition thermal imaging monitoring component, an image denoising and comparison component and a danger signal warning component. The high-definition thermal imaging monitoring component obtains the current human body and object contours and infrared features through the thermal imaging system, and sends them to the image denoising and comparison component. The image denoising and comparison component receives the current human body and object contours and infrared features and compares them with the pre-stored contours and infrared features of target personnel and target hazardous objects of different hazard levels, calculates the similarity ratio value between the two, and compares the similarity ratio value with a preset similarity threshold. When the similarity ratio value is greater than the preset similarity threshold, the judgment result of whether the staff takes action is obtained as the current human body and object danger signal. The danger signal warning component receives the similarity ratio value and the judgment result of whether the staff takes action sent by the main control component, and will issue a warning prompt for the received human body and object contour and infrared feature information.

3. A video detection system based on a job command platform according to claim 2, characterized in that: The high-definition thermal imaging monitoring component includes a thermal imaging interference recognition module, which works under the control of the image denoising and contrast component. The thermal imaging interference recognition module reduces the impact of intentional obstruction by human bodies and objects on high-definition thermal imaging monitoring.

4. The video detection system based on the job command platform according to claim 2 is characterized in that: The danger signal warning component includes a thermal imaging image data evaluation component, which sends to the thermal imaging system a judgment result on whether the staff has taken action, a similarity ratio value, and contours and infrared features of target personnel and target hazardous items of different hazard levels, and the edge server support module receives relevant information through the thermal imaging system.

5. A video detection system based on a job command platform according to any one of claims 1 to 4, characterized in that: It also includes an artificial intelligence service platform, wherein the edge server support module analyzes the obtained judgment results of whether the staff has taken action, the similarity ratio value, and the contours and infrared characteristics of target personnel and target hazardous objects of different hazard levels, generates a human and object behavior detection danger curve and sends it to the designated artificial intelligence service platform; The artificial intelligence service platform sends a request to the edge server support module to obtain the human body and object contour and infrared feature information of the portrait and object detection group.

6. A video detection system based on a job command platform according to any one of claims 1 to 4, characterized in that: It also includes a human body and object behavior recognition module, which receives commands sent by the server to recognize the human body and object behaviors in the current monitoring area.

7. A video detection system based on a job command platform according to any one of claims 1 to 6, characterized in that: The operation of the system includes the following steps: S1, setting an image feature recognition module in all human image and object image detection groups, and pre-storing the contours and infrared features of target personnel and target hazardous objects of different hazard levels in the image feature recognition module; S2, the image feature recognition module obtains the current human body and object contours and infrared features through high-definition thermal imaging monitoring, and compares the current human body and object contours and infrared features with the contours and infrared features of target personnel and target hazardous objects of different hazard levels to obtain a similarity ratio value; S3, the image feature recognition module compares the preset similarity threshold and the similarity ratio to obtain the result of whether the staff should take action, and issues an early warning prompt for the contours and infrared feature information of the human body and the object, wherein the contours and infrared feature information of the human body and the object include the result of whether the staff should take action, the contours and infrared features of target persons and target hazardous objects of different hazard levels, and the similarity ratio values; S4, the edge server supports the module to obtain and use the SIFT operator to calculate the contour and infrared feature information of the human body and objects; Step S5, the edge server support module obtains the human body and object contour and infrared feature information, generates a human body and object behavior detection danger curve based on the human body and object contour and infrared feature information, and determines the human bodies and objects in the human body and object behavior detection area through the contours and infrared features of target personnel and target hazardous objects with different hazard levels, and sends the human body and object behavior detection danger curve to the artificial intelligence service platform that manages the human body and object behavior detection area.

8. The video detection system based on the job command platform according to claim 7 is characterized in that: It also includes S6 human body and object quantity verification step, and the human body and object quantity verification step is that the edge server support module sends a thermal imaging interference recognition command to the image feature recognition module, and the image feature recognition module verifies the human body and object behavior detection frequency of the thermal imaging interference recognition module.

9. The video detection system based on the job command platform according to claim 7, characterized in that: It also includes S7 human body and object behavior recognition step, which is that the edge server support module sends a human body and object behavior recognition command to the human body and object behavior recognition module, and the human body and object behavior recognition module recognizes the human body and object behavior in the current monitoring area according to the received command.

Citation Information

Cited By

  • Landslide deep displacement field and stress field intelligent monitoring system

    CN119666064A