A method, device and medium for regional overcrowding control based on multi-source video fusion
Through multi-source video fusion technology, electronic fences are set up and video data are spliced, which solves the false alarm problem caused by overlapping multiple camera areas, and realizes accurate monitoring and security management of large-scale areas.
Patent Information
- Application Number
- CN202311314691.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-10-11
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2043-10-11
AI Technical Summary
In a large-scale dangerous area, a single camera cannot be covered. If multiple cameras are used directly, the number of people will be repeated due to the overlapping of multiple cameras, resulting in false alarms.
Adopt the regional over-capacity control method based on multi-source video fusion, by obtaining multiple regional video data and video acquisition parameters, setting up electronic fences, splicing monitoring area video data, generating complete regional monitoring video data, performing target monitoring and over-capacity judgment, and avoiding Repeated monitoring and false alarms.
It improves the accuracy and efficiency of personnel monitoring, reduces repeated monitoring, and realizes all-round real-time monitoring of large areas, ensures on-site safety, and improves the safety and operational efficiency of enterprises.
Smart Images

Figure CN117395369B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of machine vision technology, and in particular to a method, device and medium for regional overcrowding control based on multi-source video fusion. Background Art
[0002] There are many dangerous areas in industrial production, and these dangerous areas are often accompanied by restrictions on the number of people entering. If too many people enter the area, that is, overcrowding occurs, it may cause a safety accident. With the development of machine vision and artificial intelligence, machine vision algorithms for real-time video analysis have become very mature. By performing real-time analysis and reasoning on the on-site images transmitted by the camera, an alarm can be issued in time for overcrowding on the scene.
[0003] When faced with a small dangerous area, a single camera can achieve full coverage of the dangerous area, conduct real-time image analysis, and trigger an alarm when overcrowding occurs. However, when faced with a large dangerous area, a single camera often cannot cover the entire area. In this case, full coverage of the dangerous area can be achieved by increasing the number of cameras and using multiple cameras to monitor simultaneously. Although the monitoring coverage can be increased by directly increasing the number of cameras, it also greatly increases the computing resource usage of the device. On the other hand, multiple cameras often have overlapping areas, which may lead to repeated counting of the number of people on site, resulting in false alarms.
[0004] Therefore, a single camera cannot cover a larger dangerous area. If multiple cameras are used directly, the overlapping areas of the multiple cameras will cause repeated headcounts and lead to false alarms. Summary of the invention
[0005] One or more embodiments of the present specification provide a method, device and medium for regional overcrowding control based on multi-source video fusion, which is used to solve the following technical problems: for dangerous areas with a larger range, a single camera cannot provide coverage. If multiple cameras are used directly, the overlapping areas of the multiple cameras will cause repeated headcount counting and lead to false alarms.
[0006] One or more embodiments of this specification adopt the following technical solutions:
[0007] One or more embodiments of the present specification provide a method for regional overcrowding control based on multi-source video fusion, the method comprising: obtaining multiple regional video data corresponding to a to-be-controlled area, video acquisition parameters corresponding to each of the regional video data, and regional distribution data of the to-be-controlled area; setting a corresponding electronic fence for each of the regional video data, so as to determine the corresponding to-be-monitored regional video data in each of the regional video data through the electronic fence; splicing the video images of multiple to-be-monitored regional video data based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the to-be-controlled area, so as to generate complete regional monitoring video data corresponding to the to-be-controlled area; performing target monitoring on the complete regional monitoring video data, determining target personnel information in the to-be-controlled area, so as to make an overcrowding judgment on the to-be-controlled area based on the target personnel information; when it is judged that there is overcrowding in the to-be-controlled area, generating overcrowding warning information based on the target personnel information, so as to perform overcrowding control through the overcrowding warning information.
[0008] Furthermore, a corresponding electronic fence is set for each of the regional video data, specifically comprising: under the trigger of manual point selection by the user, obtaining feature point information of a plurality of feature points of the region to be monitored of each of the regional video data; according to the plurality of feature point information of the region to be monitored of each of the regional video data, sequentially connecting the plurality of feature points of the region to be monitored in each of the regional video data to generate an electronic fence of the monitoring area of each of the regional video data; determining a plurality of electronic fence monitoring area video data through an area corresponding to the electronic fence of the monitoring area in each of the regional video data; performing overlapping area analysis on the plurality of electronic fence monitoring area video data to determine a plurality of overlapping area feature points in each of the electronic fence area video data; screening the plurality of overlapping area feature points according to the position data of each of the overlapping area feature points to obtain a plurality of electronic fence feature points, and determining an initial electronic fence corresponding to each of the regional video data through the plurality of electronic fence feature points; and correcting the initial electronic fence to determine an electronic fence that meets the requirements.
[0009] Furthermore, the initial electronic fence is corrected to determine an electronic fence that meets the requirements, specifically including: according to a preset method, an operation test of overcrowding monitoring is performed on the video data of the area to be monitored corresponding to each of the initial electronic fences, and the overcrowding test results are collected, wherein the overcrowding test results include whether the same test user appears in multiple video data of the area to be monitored at the same test time; if the same test user appears in multiple video data of the area to be monitored at the same test time, the initial electronic fence corresponding to the multiple video data of the area to be monitored that appear at the same time is corrected; when the overcrowding test result is that the same test user appears in only one video data of the area to be monitored at the same test time, the corrected initial electronic fence is determined as the electronic fence that meets the requirements.
[0010] Further, based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, multiple video data of the area to be monitored are spliced to generate complete regional monitoring video data corresponding to the area to be controlled, specifically including: based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, determining the relative position relationship between the acquisition device corresponding to each of the regional video data and the area to be controlled; determining the current acquisition area of each of the acquisition devices through the relative position relationship between each of the acquisition devices and the area to be controlled and the acquisition angle information of the acquisition device in each of the video acquisition parameters; in the pre-constructed regional three-dimensional model corresponding to the area to be controlled, regional positioning is performed on each of the current acquisition areas to determine the regional splicing information of multiple current acquisition areas; the designated video parameters of each of the regional video data are adjusted to obtain multiple designated regional video data with consistent designated parameters, wherein the designated video parameters include video image size and video image resolution; based on the regional splicing information of multiple current acquisition areas, the designated regional video data corresponding to each of the current acquisition areas are spliced to generate complete regional monitoring video data corresponding to the area to be controlled.
[0011] Furthermore, based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, the relative position relationship between the acquisition device corresponding to each of the regional video data and the area to be controlled is determined, specifically including: constructing a regional three-dimensional stereo model corresponding to the area to be controlled according to the regional distribution data of the area to be controlled; obtaining the video acquisition parameters corresponding to each of the regional video data, the video acquisition parameters including the device installation position information of the acquisition device corresponding to the regional video data and the acquisition angle information of the acquisition device; positioning the acquisition device in the regional three-dimensional stereo model through the device installation position information of each of the acquisition devices, so as to determine the relative position relationship between each of the acquisition devices and the area to be controlled.
[0012] Furthermore, the video data of the designated area corresponding to each of the current acquisition areas are stitched together, specifically including: extracting features from each of the video data of the designated area to determine at least one feature information in each of the video data of the designated area, wherein the feature information includes any one or more of edge feature points, corner feature points, and texture feature points; based on at least one feature information in each of the video data of the designated area, feature pixels corresponding to the same feature information in multiple video data of the designated area are merged to stitch together the video data of the designated area corresponding to each of the current acquisition areas.
[0013] Furthermore, target monitoring is performed on the complete area monitoring video data to determine the target personnel information in the area to be controlled, specifically including: performing target feature extraction on the complete area monitoring video data to obtain target personnel attribute information of each target person, wherein the target personnel attribute information includes the belonging area identifier of each target person in the complete area monitoring video data and multiple physical feature information of each target person; based on the target personnel attribute information, target personnel information of multiple target persons in the area to be controlled is determined, wherein the target personnel information includes the number of monitored target persons; judging whether there is repeated monitoring of personnel through the belonging area identifier in the target personnel attribute information of each target person and the multiple physical feature information of each target person; when there is repeated monitoring of personnel, updating the number of monitored target persons in the target personnel information in the area to be controlled, determining the current number of target persons in the area to be controlled, so as to determine the target personnel information in the area to be controlled.
[0014] Furthermore, based on the target personnel information, overcrowding warning information is generated, specifically including: obtaining the identity attributes of each of the current target personnel in the target personnel information, wherein the identity attributes include an on-site working identity and an outsider identity; based on the identity attributes of each of the current target personnel, determining the number of on-site working personnel and the number of outsiders among multiple current target personnel in the area to be controlled; when the difference between the number of current target personnel and the number of outsiders is not greater than a preset overcrowding threshold, treating the outsiders among the multiple current target personnel as personnel to be evacuated, and generating overcrowding warning information; when the difference between the number of current target personnel and the number of outsiders is greater than the preset overcrowding threshold, based on the The location information of each current target person in the target person information and the complete area monitoring video data are used to obtain the operation node information corresponding to each on-site operator among the current target person; based on the operation node information of each on-site operator, the number of node operators corresponding to each operation node is determined, and through the number of operators at each node, a designated operation node with the largest number of node operators is determined among multiple operation nodes; according to the number of current target personnel, the number of outsiders and the preset overcrowding threshold, the recommended number of evacuees for the designated operation node is determined; based on the recommended number of evacuees for the designated operation node and the number of outsiders, overcrowding warning information is generated.
[0015] One or more embodiments of this specification provide a regional overcrowding control device based on multi-source video fusion, including:
[0016] at least one processor; and,
[0017] a memory communicatively connected to the at least one processor; wherein,
[0018] The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to:
[0019] Acquire multiple regional video data corresponding to the area to be controlled, video acquisition parameters corresponding to each of the regional video data, and regional distribution data of the area to be controlled; set a corresponding electronic fence for each of the regional video data, so as to determine the corresponding regional video data to be monitored in each of the regional video data through the electronic fence; splice the video images of multiple regional video data to be monitored based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, and generate complete regional monitoring video data corresponding to the area to be controlled; perform target monitoring on the complete regional monitoring video data, determine the target personnel information in the area to be controlled, and make an overcrowding judgment on the area to be controlled based on the target personnel information; when it is judged that there is overcrowding in the area to be controlled, generate overcrowding warning information based on the target personnel information, so as to perform overcrowding control through the overcrowding warning information.
[0020] One or more embodiments of this specification provide a non-volatile computer storage medium storing computer executable instructions, wherein the computer executable instructions are configured to:
[0021] Acquire multiple regional video data corresponding to the area to be controlled, video acquisition parameters corresponding to each of the regional video data, and regional distribution data of the area to be controlled; set a corresponding electronic fence for each of the regional video data, so as to determine the corresponding regional video data to be monitored in each of the regional video data through the electronic fence; splice the video images of multiple regional video data to be monitored based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, and generate complete regional monitoring video data corresponding to the area to be controlled; perform target monitoring on the complete regional monitoring video data, determine the target personnel information in the area to be controlled, and make an overcrowding judgment on the area to be controlled based on the target personnel information; when it is judged that there is overcrowding in the area to be controlled, generate overcrowding warning information based on the target personnel information, so as to perform overcrowding control through the overcrowding warning information.
[0022] At least one of the above technical solutions adopted in the embodiments of this specification can achieve the following beneficial effects: through the above technical solution, a corresponding electronic fence is set for each area video data, so that the corresponding video data of the area to be monitored can be determined in each area video data through the electronic fence, thereby ensuring the monitoring uniformity of the area video data under multiple acquisition devices, further improving the accuracy of personnel monitoring, and avoiding repeated monitoring; in addition, after the electronic fence is set, video analysis is only performed on the area formed by the electronic fence, reducing the amount of analysis data and improving monitoring efficiency; the video screens of multiple area video data to be monitored are spliced to generate complete area monitoring video data corresponding to the area to be controlled, and in the complete area monitoring video Real-time video data from all directions can be viewed in the video data. The large area can be segmented through the complete area monitoring video data, and the controlled area can be monitored in real time in all directions, which can be displayed more intuitively and is more suitable for overcrowding control in larger areas. When identifying the target personnel, repeated identification and judgment are performed based on the characteristics of the target personnel, and repeated identification and judgment are performed twice in the identification step to further ensure the accuracy of the number of monitored personnel. When overcrowding occurs on site, the regional overcrowding warning information is broadcast on site to remind on-site personnel to respond quickly, avoid danger, ensure the safety of personnel and enterprises, improve the safety of enterprises, and help improve the overall operational efficiency and stability of enterprises. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] In order to more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art description. Obviously, the drawings described below are only some embodiments recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. In the drawings:
[0024] Figure 1 A flowchart of a method for regional overcrowding control based on multi-source video fusion provided in an embodiment of this specification;
[0025] Figure 2 A schematic diagram of the structure of a regional overcrowding control device based on multi-source video fusion provided in an embodiment of this specification. DETAILED DESCRIPTION
[0026] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.
[0027] There are many dangerous areas in industrial production, and these dangerous areas are often accompanied by restrictions on the number of people entering. If too many people enter the area, that is, overcrowding occurs, it may cause a safety accident. With the development of machine vision and artificial intelligence, machine vision algorithms for real-time video analysis have become very mature. By performing real-time analysis and reasoning on the on-site images transmitted by the camera, an alarm can be issued in time for overcrowding on the scene.
[0028] When faced with a small dangerous area, a single camera can achieve full coverage of the dangerous area, conduct real-time image analysis, and trigger an alarm when overcrowding occurs. However, when faced with a large dangerous area, a single camera often cannot cover the entire area. In this case, full coverage of the dangerous area can be achieved by increasing the number of cameras and using multiple cameras to monitor simultaneously. Although the monitoring coverage can be increased by directly increasing the number of cameras, it also greatly increases the computing resource usage of the device. On the other hand, multiple cameras often have overlapping areas, which may lead to repeated counting of the number of people on site, resulting in false alarms.
[0029] Therefore, a single camera cannot cover a larger dangerous area. If multiple cameras are used directly, the overlapping areas of the multiple cameras will cause repeated headcounts and lead to false alarms.
[0030] The embodiment of this specification provides a method for regional overcrowding control based on multi-source video fusion. It should be noted that the execution subject in the embodiment of this specification can be a server or any device with data processing capabilities. Figure 1 A flowchart of a method for regional overcrowding control based on multi-source video fusion is provided in an embodiment of this specification, such as Figure 1 As shown, it mainly includes the following steps:
[0031] Step S101, obtaining a plurality of regional video data corresponding to the area to be controlled, a video acquisition parameter corresponding to each regional video data, and regional distribution data of the area to be controlled.
[0032] In one embodiment of the present specification, a plurality of acquisition devices are set in the area to be controlled, and the acquisition devices are used to collect video data of multiple areas in the area to be controlled. It should be noted that the multi-source video in the embodiment of the present specification refers to video data from different acquisition devices. In addition, when the acquisition device is set in the area to be controlled, the device installation position information of the acquisition device and the acquisition angle information of the acquisition device are recorded, so as to determine the video acquisition parameters corresponding to each video data when acquiring the regional video data, that is, the device installation position information and the acquisition angle information. In addition, it is also necessary to obtain the regional distribution data of the area to be controlled, which can be a spatial distribution map of the area to be controlled.
[0033] Step S102: setting a corresponding electronic fence for each area video data, so as to determine the corresponding area video data to be monitored in each area video data through the electronic fence.
[0034] The method of setting a corresponding electronic fence for each video data of the area specifically comprises: obtaining feature point information of a plurality of feature points of the area to be monitored of each video data of the area under the trigger of manual point selection by a user; sequentially connecting the plurality of feature points of the area to be monitored in each video data of the area according to the plurality of feature point information of the area to be monitored to generate an electronic fence of the monitoring area for each video data of the area; determining a plurality of video data of the monitoring area of the electronic fence through an area corresponding to the electronic fence of the monitoring area in each video data of the area; performing overlapping area analysis on the plurality of video data of the monitoring area of the electronic fence to determine a plurality of overlapping area feature points in each video data of the electronic fence area; screening the plurality of overlapping area feature points according to the position data of each overlapping area feature point to obtain a plurality of electronic fence feature points, and determining an initial electronic fence corresponding to each video data of the area through the plurality of electronic fence feature points; and correcting the initial electronic fence to determine an electronic fence that meets the requirements.
[0035] In actual application scenarios, after setting up multiple acquisition devices, there may be overlapping areas and non-monitoring areas under different perspectives between the video data of multiple areas collected by the acquisition devices. The existence of overlapping areas may easily lead to the problem of repeated monitoring by monitoring personnel, resulting in inaccurate overcrowding monitoring. If there are non-monitoring areas, if personnel detection is performed on the non-monitoring areas at the same time, the operating personnel in the non-monitoring areas will also be regarded as the target personnel in the area, resulting in inaccurate overcrowding warnings.
[0036] In one embodiment of the present specification, the user determines the operating area to be monitored by manually selecting points, and selects any frame as the regional video data in the current operating area monitoring screen corresponding to each acquisition device. The user manually selects multiple feature points of the area to be monitored in each regional video data, and under the triggering of the user's manual point selection, the feature point information of the area to be monitored of the multiple feature points of the area to be monitored in each regional video data is obtained. The feature point information of the area to be monitored here can be the coordinate information of the feature points of the area to be monitored in the regional video data. According to the coordinate information of the multiple feature points of the area to be monitored in the regional video data of each regional video data, the multiple feature points of the area to be monitored are sequentially connected to generate the monitoring area electronic fence of each regional video data. Through the area within the monitoring area electronic fence in each of the regional video data, multiple electronic fence monitoring area video data are determined. It should be noted that the electronic fence monitoring area video data here is the regional video data with the electronic fence set. In the subsequent overlapping area analysis, only the video data within the monitoring area electronic fence is analyzed. Through the above technical solution, under the triggering of the user's manual point selection, the area to be monitored in the operating area is determined, avoiding the problem of inaccurate overcrowding warning when there is a non-monitoring area.
[0037] An overlapping area analysis is performed on multiple electronic fence monitoring area video data, multiple overlapping area feature points are determined in each electronic fence area video data, and the position data of each overlapping area feature point is obtained. According to the position data of each overlapping area feature point, multiple overlapping area feature points are screened to obtain multiple electronic fence feature points. It should be noted that the area composed of the obtained overlapping area feature points is the part that overlaps with other electronic fence monitoring area video data, and the overlapping part needs to be retained only in one of the electronic fence monitoring area video data. Therefore, when screening the overlapping area feature points, the monitoring area electronic fence in one of the electronic fence monitoring area video data can be directly used as the initial electronic fence of the electronic fence monitoring area video data, and when screening the overlapping area feature points in other electronic fence monitoring area video data, the regional feature points close to the middle part of other electronic fence monitoring area video data are used as electronic fence feature points, so as to determine the initial electronic fence corresponding to the regional video data through the electronic fence feature points and the monitoring area electronic fence in other electronic fence monitoring area video data. It should be noted that, in general, the overlapping area in the monitoring area is a small area. Through the electronic fence feature points, the electronic fence of the monitoring area in the video data of other electronic fence monitoring areas is divided into two areas, and the area boundary corresponding to the larger area is selected as the initial electronic fence. In order to avoid errors in repeated area analysis, after the initial electronic fence is generated, the initial electronic fence is corrected to determine the electronic fence that meets the requirements.
[0038] In one embodiment of the present specification, the initial electronic fence is corrected to determine an electronic fence that meets the requirements, specifically including: according to a preset method, performing an operation test of overcrowding monitoring on the video data of the area to be monitored corresponding to each of the initial electronic fences, and collecting overcrowding test results, wherein the overcrowding test results include whether the same test user appears in multiple video data of the area to be monitored at the same test time; if the same test user appears in multiple video data of the area to be monitored at the same test time, then correcting the initial electronic fence corresponding to the multiple video data of the area to be monitored that appear at the same time; when the overcrowding test result is that the same test user appears in only one video data of the area to be monitored at the same test time, the corrected initial electronic fence is determined to be an electronic fence that meets the requirements.
[0039] In one embodiment of the present specification, after setting a corresponding electronic fence for each area video data, the corresponding video data of the area to be monitored is determined in each area video data through the electronic fence. It should be noted that the video data of the area to be monitored is the area video data where the electronic fence is set, and subsequently only the area formed by the electronic fence is subjected to video analysis.
[0040] Through the above technical solution, a corresponding electronic fence is set for each area video data, so that the corresponding video data of the area to be monitored can be determined in each area video data through the electronic fence, thereby ensuring the monitoring uniformity of the area video data under multiple acquisition devices, further improving the accuracy of personnel monitoring, and avoiding repeated monitoring; in addition, after setting up the electronic fence, video analysis is only performed on the area formed by the electronic fence, reducing the amount of analysis data and improving monitoring efficiency.
[0041] Step S103, based on the video acquisition parameters corresponding to each area video data and the area distribution data of the area to be controlled, the video images of multiple areas to be monitored are spliced to generate complete area monitoring video data corresponding to the area to be controlled.
[0042] Based on the video acquisition parameters corresponding to each video data of the area and the regional distribution data of the area to be controlled, multiple video data of the area to be monitored are spliced to generate complete regional monitoring video data corresponding to the area to be controlled, specifically including: based on the video acquisition parameters corresponding to each video data of the area and the regional distribution data of the area to be controlled, determining the relative position relationship between the acquisition device corresponding to each video data of the area and the area to be controlled; determining the current acquisition area of each acquisition device through the relative position relationship between each acquisition device and the area to be controlled and the acquisition angle information of the acquisition device in each video acquisition parameter; in the pre-constructed regional three-dimensional model corresponding to the area to be controlled, each current acquisition area is regionally located to determine the regional splicing information of multiple current acquisition areas; the designated video parameters of each video data of the area are adjusted to obtain multiple designated regional video data with consistent designated parameters, wherein the designated video parameters include video image size and video image resolution; based on the regional splicing information of multiple current acquisition areas, the video images of the designated area corresponding to each current acquisition area are spliced to generate complete regional monitoring video data corresponding to the area to be controlled.
[0043] In one embodiment of the present specification, the relative position relationship between the acquisition device corresponding to each regional video data and the area to be controlled is determined by the video acquisition parameters corresponding to the regional video data and the regional distribution data of the area to be controlled. First, according to the regional distribution data of the area to be controlled, a regional three-dimensional stereo model corresponding to the area to be controlled is constructed, and the video acquisition parameters corresponding to each of the regional video data are obtained. The video acquisition parameters include the device installation position information of the acquisition device corresponding to the regional video data. Through the device installation position information of each of the acquisition devices, the acquisition device is positioned in the regional three-dimensional model to determine the relative position relationship between each acquisition device and the area to be controlled. For example, the relative position relationship here can be a position relationship such as the acquisition device is located on the north wall and the acquisition device is located on the east wall.
[0044] The current acquisition area of each acquisition device is determined by the relative positional relationship between each acquisition device and the area to be controlled and the acquisition angle information of the acquisition device in each video acquisition parameter. The current acquisition area refers to the acquisition area in the area to be controlled corresponding to the acquisition device. In the pre-constructed three-dimensional model of the area corresponding to the area to be controlled, each current acquisition area is regionally positioned to determine the regional splicing information of multiple current acquisition areas, for example, the current acquisition area 1 is spliced with the current acquisition area 2 left and right, and the current acquisition area 1 is spliced with the current acquisition area 3 up and down. The size and resolution of each regional video data are adjusted to obtain multiple designated regional video data with consistent parameters. Based on the regional splicing information of multiple current acquisition areas, the video data of the designated area corresponding to each current acquisition area is spliced to generate complete regional monitoring video data corresponding to the area to be controlled. Taking the example of setting up a collection device on the four walls of the east, west, south and north of the area to be controlled, the image of the first camera is placed in the upper left corner of the stitched screen, the image of the second camera is placed in the upper right corner of the stitched screen, the image of the third camera is placed in the lower left corner of the stitched screen, and the image of the fourth camera is placed in the lower right corner of the stitched screen, so as to complete the preliminary stitching. The area is regarded as a larger cubic area, and the four cameras monitor the entire area in real time from the front, back, left and right directions respectively.
[0045] The video data of the designated area corresponding to each current acquisition area are stitched together, specifically comprising: extracting features from each of the video data of the designated area, and determining at least one feature information in each of the video data of the designated area, wherein the feature information includes any one or more of edge feature points, corner feature points, and texture feature points; based on the at least one feature information in each of the video data of the designated area, feature pixels corresponding to the same feature information in multiple video data of the designated area are merged, so as to stitch together the video data of the designated area corresponding to each of the current acquisition area.
[0046] In actual application scenarios, in order to ensure the viewing experience of maintenance personnel, the image feature stitching method is used to stitch the feature information in the image, such as through edges, corners, textures, etc. First, the image features are extracted, then the images are aligned according to the feature information, and finally the pixels between the features are fused. The feature stitching method can ensure that the transition between images is natural and smooth, without obvious stitching marks. Finally, the stitched images can be further processed, such as adjusting the brightness and contrast, increasing the color balance, etc., to make the image more natural and beautiful. It should be noted that when stitching the images, factors such as the shooting angle and light between the cameras need to be considered to ensure the best stitching effect of the images. In addition, the accuracy and computational efficiency of the stitching algorithm will also affect the quality and speed of the stitched images. Therefore, it is necessary to select according to the computing power conditions of the on-site equipment.
[0047] Through the above technical solution, the video data of multiple monitored areas are spliced to generate complete area monitoring video data corresponding to the area to be controlled. Real-time video data from all directions can be viewed in the complete area monitoring video data. Large areas can be divided into regions through the complete area monitoring video data, and all-round real-time monitoring of the controlled area can be carried out. A more intuitive display can be made, which is more suitable for overcrowding control in larger areas.
[0048] Step S104, target monitoring is performed on the complete area monitoring video data to determine the target personnel information in the area to be controlled, and based on the target personnel information, overcrowding judgment is made in the area to be controlled.
[0049] The target monitoring is performed on the complete area monitoring video data to determine the target personnel information in the area to be controlled, specifically including: performing target feature extraction on the complete area monitoring video data to obtain the target personnel attribute information of each target person, wherein the target personnel attribute information includes the belonging area identifier of each target person in the complete area monitoring video data and multiple physical feature information of each target person; based on the target personnel attribute information, the target personnel information of multiple target persons in the area to be controlled is determined, wherein the target personnel information includes the number of monitored target persons; judging whether there is repeated monitoring of personnel through the belonging area identifier in the target personnel attribute information of each target person and the multiple physical feature information of each target person; when there is repeated monitoring of personnel, updating the number of monitored target persons in the target personnel information in the area to be controlled, determining the current number of target persons in the area to be controlled, so as to determine the target personnel information in the area to be controlled.
[0050] In one embodiment of the present specification, target feature extraction is performed on the complete area monitoring video data to obtain target person attribute information of each target person, wherein the target person attribute information here includes the belonging area identification of each target person in the complete area monitoring video data and multiple physical feature information of each target person. Since the complete area monitoring video data is obtained by splicing multiple video data of the area to be monitored, the relative position relationship of each monitoring area is considered during the splicing, and the belonging area identification here can be set according to the video data of the area to be monitored, or according to the monitoring area corresponding to the video data of the area to be monitored. The purpose is to quickly locate the target person through the belonging area identification, which is suitable for overcrowding control in a larger range, and reduces the larger control area to a small monitoring area corresponding to the video acquisition device. The physical feature information can be multiple body feature points and clothing feature points of the target person. The body feature points include key part feature points such as the top of the head feature points and the foot feature points. The purpose is to compare the features of multiple target persons to determine whether repeated monitoring is required and to count the number of target persons. The identity attributes of the target user are confirmed through the clothing feature points. Generally, on-site workers usually wear specific work clothes, and the clothing features of outsiders are different from those of work clothes.
[0051] Based on the target person attribute information, the target person information of multiple target persons in the area to be controlled is determined, and the target person information includes the number of monitored target persons. Through the belonging area identifier in the target person attribute information of each target person and the multiple physical characteristics information of each target person, it is judged whether there is repeated monitoring of personnel. When there is repeated monitoring of personnel, the number of repeated monitoring is determined, the number of monitored target persons in the target person information in the area to be controlled is updated, and the current number of target persons in the area to be controlled is determined, and the current number of target persons is ensured to be the real-time number of persons, so as to determine the target person information in the area to be controlled.
[0052] In one embodiment of the present specification, the current target number of the target personnel information is compared with a preset overcrowding threshold. When the current target number of personnel is not greater than the preset overcrowding threshold, it is determined that there is no overcrowding. On the contrary, when the current target number of personnel is greater than the preset overcrowding threshold, it is determined that there is overcrowding.
[0053] Through the above technical solution, when identifying the target person, repeated identification and judgment are performed according to the characteristics of the target person, and a second repeated identification and judgment is performed in the identification step, which further ensures the accuracy of the number of monitored personnel.
[0054] Step S105, when it is determined that there is overcrowding in the area to be controlled, overcrowding warning information is generated based on the target personnel information, so as to perform overcrowding control through the overcrowding warning information.
[0055] Based on the target personnel information and the complete area monitoring video data, overcrowding warning information is generated, specifically including: obtaining the identity attributes of each current target personnel in the target personnel information, wherein the identity attributes include the on-site operation identity and the outsider identity; based on the identity attributes of each current target personnel, determining the number of on-site operation personnel and the number of outsiders among the multiple current target personnel in the area to be controlled; when the difference between the number of the current target personnel and the number of outsiders is not greater than the preset overcrowding threshold, treating the outsiders among the multiple current target personnel as personnel to be evacuated, and generating overcrowding warning information; when the difference between the number of the current target personnel and the number of outsiders is greater than When the overcrowding threshold is preset, based on the position information of each current target person in the target person information, the operation node information corresponding to each on-site operator in the current target person is obtained; based on the operation node information of each on-site operator, the number of node operators corresponding to each operation node is determined, and through the number of operators at each node, the designated operation node with the largest number of node operators is determined among multiple operation nodes; based on the number of current target personnel, the number of outsiders and the preset overcrowding threshold, the recommended number of people to be evacuated from the designated operation node is determined; based on the recommended number of people to be evacuated from the designated operation node and the number of outsiders, overcrowding warning information is generated.
[0056] In one embodiment of the present specification, the number of on-site operators and the number of outsiders among the multiple current target personnel in the area to be controlled are determined by the identity attribute of each current target person; when the difference between the number of the current target personnel and the number of outsiders is not greater than the preset overcrowding threshold, the outsiders among the multiple current target personnel are regarded as personnel to be evacuated, overcrowding warning information is generated, and the overcrowding warning information is broadcast, and the overcrowding warning information includes information such as prompting overcrowding and warning outsiders to evacuate quickly. When the difference between the number of the current target personnel and the number of outsiders is greater than the preset overcrowding threshold, based on the position information of each current target person in the target personnel information, the operation node information corresponding to each on-site operator among the current target personnel is obtained. It should be noted that the operation node information can be determined by the position information of each current target person. If the positions of multiple current target persons are close, it is considered that the multiple current target persons belong to the same operation node. The position information of each current target person can also be compared with the position information of the operation equipment in the area to be controlled, and the operation node information of each current target person can be determined by the operation equipment close to the current target person.
[0057] Based on the operation node information of each on-site operator, the number of node operators corresponding to each operation node is determined. Through the number of operators at each node, the designated operation node with the largest number of node operators is determined among multiple operation nodes. According to the current number of target personnel, the number of outsiders and the preset overcrowding threshold, the recommended number of evacuees for the designated operation node is determined. For example, the difference between the current number of target personnel and the preset overcrowding threshold is calculated to determine the total number of overcrowding. The recommended number of operators to be evacuated is determined by the difference between the total number of overcrowding and the number of outsiders. Based on the recommended number of evacuees for the designated operation node and the number of outsiders, overcrowding warning information is generated. The generated overcrowding warning information includes prompts for overcrowding, warnings for outsiders to evacuate quickly, and recommended designated operation nodes for evacuation, as well as the recommended number of evacuations.
[0058] In one embodiment of this specification, once an overcrowding situation occurs on site, the system will immediately detect and trigger an alarm signal, which will be broadcast on site through an alarm to warn of overcrowding in the area, reminding on-site personnel to respond quickly, avoid danger, and ensure the safety of personnel and enterprises. In addition, when overcrowding occurs, real-time snapshots will be taken to provide clear picture evidence for enterprise managers, which will help them better understand the details of the on-site overcrowding situation. Detailed alarm records will also be generated to provide enterprises with a basis for comprehensive tracing and investigation of overcrowding situations, which can help managers evaluate the effectiveness of on-site safety measures, discover potential problems in the operation process, and take targeted improvement measures, which not only improves the safety of the enterprise, but also helps to improve the overall operational efficiency and stability of the enterprise.
[0059] Through the above technical scheme, a corresponding electronic fence is set for each regional video data, so that the corresponding video data of the area to be monitored can be determined in each regional video data through the electronic fence, which ensures the monitoring singleness of the regional video data under multiple acquisition devices, further improves the accuracy of personnel monitoring, and avoids repeated monitoring; in addition, after the electronic fence is set, only the area composed of the electronic fence is analyzed, which reduces the amount of analysis data and improves the monitoring efficiency; the video screens of multiple video data of the monitored areas are spliced to generate complete regional monitoring video data corresponding to the area to be controlled, and real-time video data of various directions can be viewed in the complete regional monitoring video data. The large area can be divided into regions through the complete regional monitoring video data, and the area to be controlled can be monitored in all directions in real time, which can be displayed more intuitively and is more suitable for overcrowding control in larger areas; when identifying the target personnel, repeated identification and judgment are performed according to the characteristics of the target personnel, and repeated identification and judgment are performed twice in the identification step, which further ensures the accuracy of the number of personnel monitored; when overcrowding occurs on site, the regional overcrowding warning information is broadcast on site to remind the on-site personnel to respond quickly, avoid danger, ensure the safety of personnel and enterprises, improve the safety of enterprises, and help improve the overall operational efficiency and stability of enterprises.
[0060] The embodiment of this specification also provides a regional overcrowding control device based on multi-source video fusion, such as Figure 2 As shown, the device includes: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to:
[0061] Acquire multiple regional video data corresponding to the area to be controlled, video acquisition parameters corresponding to each of the regional video data, and regional distribution data of the area to be controlled; set a corresponding electronic fence for each of the regional video data, so as to determine the corresponding video data of the area to be monitored in each of the regional video data through the electronic fence; splice the video images of multiple video data of the area to be monitored based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, and generate complete regional monitoring video data corresponding to the area to be controlled; perform target monitoring on the complete regional monitoring video data, determine the target personnel information in the area to be controlled, and make an overcrowding judgment on the area to be controlled based on the target personnel information; when it is judged that there is overcrowding in the area to be controlled, generate overcrowding warning information based on the target personnel information, so as to perform overcrowding control through the overcrowding warning information.
[0062] The embodiment of the present specification also provides a non-volatile computer storage medium storing computer executable instructions, wherein the computer executable instructions are configured as follows:
[0063] Acquire multiple regional video data corresponding to the area to be controlled, video acquisition parameters corresponding to each of the regional video data, and regional distribution data of the area to be controlled; set a corresponding electronic fence for each of the regional video data, so as to determine the corresponding video data of the area to be monitored in each of the regional video data through the electronic fence; splice the video images of multiple video data of the area to be monitored based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, and generate complete regional monitoring video data corresponding to the area to be controlled; perform target monitoring on the complete regional monitoring video data, determine the target personnel information in the area to be controlled, and make an overcrowding judgment on the area to be controlled based on the target personnel information; when it is judged that there is overcrowding in the area to be controlled, generate overcrowding warning information based on the target personnel information, so as to perform overcrowding control through the overcrowding warning information.
[0064] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.
[0065] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0066] The devices and media provided in the embodiments of this specification correspond one-to-one to the methods. Therefore, the devices and media also have similar beneficial technical effects as the corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be repeated here.
[0067] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, this specification may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.
[0068] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0069] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0070] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0071] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0072] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.
[0073] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.
[0074] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0075] The above description is only one or more embodiments of this specification and is not intended to limit this specification. For those skilled in the art, one or more embodiments of this specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of one or more embodiments of this specification shall be included in the scope of the claims of this specification.
Claims
1. A regional overcrowding control method based on multi-source video fusion, characterized in that: The method comprises: Acquire multiple regional video data corresponding to the area to be controlled, video acquisition parameters corresponding to each of the regional video data, and regional distribution data of the area to be controlled; Setting a corresponding electronic fence for each of the regional video data, so as to determine the corresponding regional video data to be monitored in each of the regional video data through the electronic fence; Based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, the video images of the plurality of the video data of the area to be monitored are spliced to generate complete regional monitoring video data corresponding to the area to be controlled; Performing target monitoring on the complete area monitoring video data to determine target personnel information in the area to be controlled, and making an overcrowding judgment on the area to be controlled based on the target personnel information; When it is determined that there is overcrowding in the area to be controlled, overcrowding warning information is generated based on the target personnel information, so as to perform overcrowding control through the overcrowding warning information; Setting a corresponding electronic fence for each of the area video data specifically includes: Under the trigger of the user's manual point selection, the feature point information of the area to be monitored of the plurality of feature points of the area to be monitored of each of the video data of the area is obtained; According to the information of the plurality of feature points of the area to be monitored in each of the area video data, in each of the area video data, the plurality of feature points of the area to be monitored are sequentially connected to generate an electronic fence of the monitoring area of each of the area video data; Determine a plurality of video data of the electronic fence monitoring area through the area corresponding to the electronic fence of the monitoring area in each of the video data of the area; Performing overlapping area analysis on the video data of the plurality of electronic fence monitoring areas, and determining a plurality of overlapping area feature points in the video data of each electronic fence area; According to the position data of each overlapping area feature point, the plurality of overlapping area feature points are screened to obtain a plurality of electronic fence feature points, and the initial electronic fence corresponding to each area video data is determined through the plurality of electronic fence feature points; Modifying the initial electronic fence to determine an electronic fence that meets the requirements; The initial electronic fence is modified to determine an electronic fence that meets the requirements, specifically including: According to a preset method, an operation test of overcrowding monitoring is performed on the video data of the monitored area corresponding to each of the initial electronic fences, and an overcrowding test result is collected, wherein the overcrowding test result includes whether the same test user appears in the video data of multiple monitored areas at the same test time; If the same test user appears in multiple video data of the to-be-monitored area at the same test time, the initial electronic fence corresponding to the multiple video data of the to-be-monitored area appearing at the same time is corrected; when the overcrowding test result is that the same test user only appears in one video data of the to-be-monitored area at the same test time, the corrected initial electronic fence is determined as the electronic fence that meets the requirements; Based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, the video images of the plurality of the regional video data to be monitored are spliced to generate complete regional monitoring video data corresponding to the area to be controlled, specifically including: Based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, determining the relative position relationship between the acquisition device corresponding to each of the regional video data and the area to be controlled; Determine the current acquisition area of each acquisition device through the relative position relationship between each acquisition device and the area to be controlled and the acquisition angle information of the acquisition device in each video acquisition parameter; In the pre-constructed regional three-dimensional model corresponding to the area to be controlled, regional positioning is performed on each of the current acquisition areas to determine regional splicing information of multiple current acquisition areas; Adjusting the designated video parameters of each of the regional video data to obtain a plurality of designated regional video data with consistent designated parameters, wherein the designated video parameters include video image size and video image resolution; Based on the regional splicing information of the multiple current acquisition areas, the video data of the designated area corresponding to each current acquisition area is spliced to generate complete regional monitoring video data corresponding to the area to be controlled; Based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, determining the relative position relationship between the acquisition device corresponding to each of the regional video data and the area to be controlled specifically includes: Constructing a three-dimensional regional model corresponding to the area to be controlled based on the regional distribution data of the area to be controlled; Acquire video acquisition parameters corresponding to each of the regional video data, wherein the video acquisition parameters include device installation position information of the acquisition device corresponding to the regional video data and acquisition angle information of the acquisition device; The acquisition device is positioned in the three-dimensional model of the area by using the device installation position information of each acquisition device to determine the relative position relationship between each acquisition device and the area to be controlled; The video data of the designated area corresponding to each of the current acquisition areas is spliced together, specifically comprising: Performing feature extraction on each of the designated area video data to determine at least one feature information in each of the designated area video data, wherein the feature information includes any one or more of edge feature points, corner feature points, and texture feature points; Based on at least one feature information in each of the designated area video data, feature pixels corresponding to the same feature information in multiple designated area video data are fused to splice video images of the designated area video data corresponding to each of the current acquisition areas.
2. According to the method of claim 1, the method is characterized in that: Performing target monitoring on the complete area monitoring video data to determine the target personnel information in the area to be controlled, specifically including: Extracting target features from the complete area monitoring video data to obtain target person attribute information of each target person, wherein the target person attribute information includes a belonging area identifier of each target person in the complete area monitoring video data and multiple body feature information of each target person; Based on the target personnel attribute information, determine the target personnel information of multiple target personnel in the area to be controlled, wherein the target personnel information includes the number of monitored target personnel; Determining whether there is repeated monitoring of a person through the belonging area identifier in the target person attribute information of each target person and multiple physical feature information of each target person; When repeated monitoring of personnel exists, the number of monitored target personnel in the target personnel information in the area to be controlled is updated, and the current number of target personnel in the area to be controlled is determined to determine the target personnel information in the area to be controlled.
3. The method for regional overcrowding control based on multi-source video fusion according to claim 2 is characterized in that: Based on the target personnel information, overcrowding warning information is generated, specifically including: Acquire the identity attributes of each current target person in the target person information, wherein the identity attributes include on-site operation identity and external identity; Based on the identity attribute of each of the current target persons, determine the number of on-site workers and the number of outsiders among the multiple current target persons in the area to be controlled; When the difference between the number of current target personnel and the number of outsiders is not greater than a preset overcrowding threshold, outsiders among the multiple current target personnel are regarded as personnel to be evacuated, and overcrowding warning information is generated; When the difference between the number of current target personnel and the number of outsiders is greater than a preset overcrowding threshold, based on the location information of each current target person in the target personnel information and the complete area monitoring video data, the operation node information corresponding to each on-site operator in the current target personnel is obtained; Based on the operation node information of each on-site operator, determine the number of node operators corresponding to each operation node, and determine the designated operation node with the largest number of node operators among multiple operation nodes through the number of each node operator; Determine the recommended number of evacuees for the designated operation node according to the current number of target personnel, the number of outsiders, and the preset overcrowding threshold; Based on the recommended number of evacuees for the designated operation node and the number of outsiders, overcrowding warning information is generated.
4. A regional overcrowding control device based on multi-source video fusion, characterized in that: The device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to: Acquire multiple regional video data corresponding to the area to be controlled, video acquisition parameters corresponding to each of the regional video data, and regional distribution data of the area to be controlled; Setting a corresponding electronic fence for each of the regional video data, so as to determine the corresponding regional video data to be monitored in each of the regional video data through the electronic fence; Based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, the video images of the plurality of the video data of the area to be monitored are spliced to generate complete regional monitoring video data corresponding to the area to be controlled; Performing target monitoring on the complete area monitoring video data to determine target personnel information in the area to be controlled, and making an overcrowding judgment on the area to be controlled based on the target personnel information; When it is determined that there is overcrowding in the area to be controlled, overcrowding warning information is generated based on the target personnel information, so as to perform overcrowding control through the overcrowding warning information; Setting a corresponding electronic fence for each of the area video data specifically includes: Under the trigger of the user's manual point selection, the feature point information of the area to be monitored of the plurality of feature points of the area to be monitored of each of the video data of the area is obtained; According to the information of the plurality of feature points of the area to be monitored in each of the area video data, in each of the area video data, the plurality of feature points of the area to be monitored are sequentially connected to generate an electronic fence of the monitoring area of each of the area video data; Determine a plurality of video data of the electronic fence monitoring area through the area corresponding to the electronic fence of the monitoring area in each of the video data of the area; Performing overlapping area analysis on the video data of the plurality of electronic fence monitoring areas, and determining a plurality of overlapping area feature points in the video data of each electronic fence area; According to the position data of each overlapping area feature point, the plurality of overlapping area feature points are screened to obtain a plurality of electronic fence feature points, and the initial electronic fence corresponding to each area video data is determined through the plurality of electronic fence feature points; Modifying the initial electronic fence to determine an electronic fence that meets the requirements; The initial electronic fence is modified to determine an electronic fence that meets the requirements, specifically including: According to a preset method, an operation test of overcrowding monitoring is performed on the video data of the monitored area corresponding to each of the initial electronic fences, and an overcrowding test result is collected, wherein the overcrowding test result includes whether the same test user appears in the video data of multiple monitored areas at the same test time; If the same test user appears in multiple video data of the to-be-monitored area at the same test time, the initial electronic fence corresponding to the multiple video data of the to-be-monitored area appearing at the same time is corrected; when the overcrowding test result is that the same test user only appears in one video data of the to-be-monitored area at the same test time, the corrected initial electronic fence is determined as the electronic fence that meets the requirements; Based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, the video images of the plurality of the regional video data to be monitored are spliced to generate complete regional monitoring video data corresponding to the area to be controlled, specifically including: Based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, determining the relative position relationship between the acquisition device corresponding to each of the regional video data and the area to be controlled; Determine the current acquisition area of each acquisition device through the relative position relationship between each acquisition device and the area to be controlled and the acquisition angle information of the acquisition device in each video acquisition parameter; In the pre-constructed regional three-dimensional model corresponding to the area to be controlled, regional positioning is performed on each of the current acquisition areas to determine regional splicing information of multiple current acquisition areas; Adjusting the designated video parameters of each of the regional video data to obtain a plurality of designated regional video data with consistent designated parameters, wherein the designated video parameters include video image size and video image resolution; Based on the regional splicing information of the multiple current acquisition areas, the video data of the designated area corresponding to each current acquisition area is spliced to generate complete regional monitoring video data corresponding to the area to be controlled; Based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, determining the relative position relationship between the acquisition device corresponding to each of the regional video data and the area to be controlled specifically includes: Constructing a three-dimensional regional model corresponding to the area to be controlled based on the regional distribution data of the area to be controlled; Acquire video acquisition parameters corresponding to each of the regional video data, wherein the video acquisition parameters include device installation position information of the acquisition device corresponding to the regional video data and acquisition angle information of the acquisition device; The acquisition device is positioned in the three-dimensional model of the area by using the device installation position information of each acquisition device to determine the relative position relationship between each acquisition device and the area to be controlled; The video data of the designated area corresponding to each of the current acquisition areas is spliced together, specifically comprising: Performing feature extraction on each of the designated area video data to determine at least one feature information in each of the designated area video data, wherein the feature information includes any one or more of edge feature points, corner feature points, and texture feature points; Based on at least one feature information in each of the designated area video data, feature pixels corresponding to the same feature information in multiple designated area video data are fused to splice video images of the designated area video data corresponding to each of the current acquisition areas.
5. A non-volatile computer storage medium storing computer executable instructions, characterized in that: The computer executable instructions are configured to: Acquire multiple regional video data corresponding to the area to be controlled, video acquisition parameters corresponding to each of the regional video data, and regional distribution data of the area to be controlled; Setting a corresponding electronic fence for each of the regional video data, so as to determine the corresponding regional video data to be monitored in each of the regional video data through the electronic fence; Based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, the video images of the plurality of the video data of the area to be monitored are spliced to generate complete regional monitoring video data corresponding to the area to be controlled; Performing target monitoring on the complete area monitoring video data to determine target personnel information in the area to be controlled, and making an overcrowding judgment on the area to be controlled based on the target personnel information; When it is determined that there is overcrowding in the area to be controlled, overcrowding warning information is generated based on the target personnel information, so as to perform overcrowding control through the overcrowding warning information; Setting a corresponding electronic fence for each of the area video data specifically includes: Under the trigger of the user's manual point selection, the feature point information of the area to be monitored of the plurality of feature points of the area to be monitored of each of the video data of the area is obtained; According to the information of the plurality of feature points of the area to be monitored in each of the area video data, in each of the area video data, the plurality of feature points of the area to be monitored are sequentially connected to generate an electronic fence of the monitoring area of each of the area video data; Determine a plurality of video data of the electronic fence monitoring area through the area corresponding to the electronic fence of the monitoring area in each of the video data of the area; Performing overlapping area analysis on the video data of the plurality of electronic fence monitoring areas, and determining a plurality of overlapping area feature points in the video data of each electronic fence area; According to the position data of each overlapping area feature point, the plurality of overlapping area feature points are screened to obtain a plurality of electronic fence feature points, and the initial electronic fence corresponding to each area video data is determined through the plurality of electronic fence feature points; Modifying the initial electronic fence to determine an electronic fence that meets the requirements; The initial electronic fence is modified to determine an electronic fence that meets the requirements, specifically including: According to a preset method, an operation test of overcrowding monitoring is performed on the video data of the monitored area corresponding to each of the initial electronic fences, and an overcrowding test result is collected, wherein the overcrowding test result includes whether the same test user appears in the video data of multiple monitored areas at the same test time; If the same test user appears in multiple video data of the to-be-monitored area at the same test time, the initial electronic fence corresponding to the multiple video data of the to-be-monitored area appearing at the same time is corrected; when the overcrowding test result is that the same test user only appears in one video data of the to-be-monitored area at the same test time, the corrected initial electronic fence is determined as the electronic fence that meets the requirements; Based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, the video images of the plurality of the video data of the area to be monitored are spliced to generate the complete regional monitoring video data corresponding to the area to be controlled, specifically including: Based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, determining the relative position relationship between the acquisition device corresponding to each of the regional video data and the area to be controlled; Determine the current acquisition area of each acquisition device through the relative position relationship between each acquisition device and the area to be controlled and the acquisition angle information of the acquisition device in each video acquisition parameter; In the pre-constructed regional three-dimensional model corresponding to the area to be controlled, regional positioning is performed on each of the current acquisition areas to determine regional splicing information of multiple current acquisition areas; Adjusting the designated video parameters of each of the regional video data to obtain a plurality of designated regional video data with consistent designated parameters, wherein the designated video parameters include video image size and video image resolution; Based on the regional splicing information of the multiple current acquisition areas, the video data of the designated area corresponding to each current acquisition area is spliced to generate complete regional monitoring video data corresponding to the area to be controlled; Based on the video acquisition parameters corresponding to each of the regional video data and the regional distribution data of the area to be controlled, determining the relative position relationship between the acquisition device corresponding to each of the regional video data and the area to be controlled specifically includes: Constructing a three-dimensional regional model corresponding to the area to be controlled based on the regional distribution data of the area to be controlled; Acquire video acquisition parameters corresponding to each of the regional video data, wherein the video acquisition parameters include device installation position information of the acquisition device corresponding to the regional video data and acquisition angle information of the acquisition device; The acquisition device is positioned in the three-dimensional model of the area by using the device installation position information of each acquisition device to determine the relative position relationship between each acquisition device and the area to be controlled; The video data of the designated area corresponding to each of the current acquisition areas is spliced together, specifically comprising: Performing feature extraction on each of the designated area video data to determine at least one feature information in each of the designated area video data, wherein the feature information includes any one or more of edge feature points, corner feature points, and texture feature points; Based on at least one feature information in each of the designated area video data, feature pixels corresponding to the same feature information in multiple designated area video data are fused to splice video images of the designated area video data corresponding to each of the current acquisition areas.
