Industrial video monitoring system with automatic configuration function

Abstract

The invention relates to the technical field of video monitoring, in particular to an industrial video monitoring system with an automatic configuration function, and the system comprises a video collection and transmission module which is used for collecting a monitoring video in real time and transmitting the monitoring video to a monitoring center; the intelligent analysis and processing module is used for acquiring the weight of each adjacent image block of each image block in each frame of image, and acquiring the consistency coefficient of each image block by combining the similarity of the motion vectors between each image block and each adjacent image block; acquiring a characteristic value of each frame of image, and filtering each frame of image in combination with a filtering algorithm; the video storage and management module is used for storing the monitoring video after filtering processing and information and indexes of monitoring video data; and the monitoring picture display and operation module is used for displaying a monitoring picture and customizing a polling scheme. The invention aims to improve the quality of industrial video monitoring data by improving the filtering effect of the monitoring video, thereby improving the accuracy of industrial video monitoring.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of video monitoring, in particular to an industrial video monitoring system with automatic configuration function. BACKGROUND

[0002] With the development of science and technology, the application of industrial video monitoring system in coal mine dispatching and command is becoming more and more critical. High-definition cameras are installed in different monitoring areas of coal mine to take pictures of the status of each area and transmit the pictures to the dispatching and command center in real time, realizing omnidirectional monitoring and management. At present, in order to adapt to the complex and changeable industrial scene, the industrial monitoring system increases the automatic configuration function. When the industrial process is adjusted or the equipment layout is changed, the automatic configuration function can quickly adjust the camera angle and focal length parameters according to the new monitoring requirements, thereby greatly reducing manual intervention and effectively improving the reliability of the monitoring system.

[0003] However, in actual use, due to the interference of the construction site, the camera may shake or the light may change rapidly in a short time, resulting in poor quality of the collected industrial monitoring video, affecting the quality of real-time monitoring. The existing technology usually realizes the purpose of improving image quality by filtering each frame of the monitoring video, but when shaking and light change occur rapidly in a short time, the device parameters cannot be adjusted in time, resulting in significant increase in the frame interference difference of the monitoring video in different time periods, causing the processing precision of the industrial monitoring video to decrease, and reducing the accuracy of the industrial video monitoring. SUMMARY

[0004] In view of the above, it is necessary to provide an industrial video monitoring system with automatic configuration function. Compared with the traditional industrial video monitoring system with automatic configuration function, the quality of the industrial video monitoring data is improved by improving the filtering effect of the monitoring video, thereby improving the accuracy of the industrial video monitoring.

[0005] The industrial video monitoring system with automatic configuration function provided by the present application adopts the following technical scheme: One embodiment of the present application provides an industrial video monitoring system with automatic configuration function, wherein the system comprises: a video acquisition and transmission module, configured to acquire monitoring video in real time by using a camera and transmit the monitoring video to a monitoring center; an intelligent analysis and processing module, configured to acquire, in the monitoring center, motion vectors and matching errors of image blocks in each frame of the monitoring video by using a motion estimation search algorithm, and acquire weights of each neighboring image block of each image block according to the distribution of the matching errors of all the neighboring image blocks of each image block for any frame of image; The consistency coefficient of each image block is obtained by combining the similarity of motion vectors between each image block and its nearest neighbor image blocks with the weights. By analyzing the distribution characteristics of the matching errors of all image blocks within any given frame, and combining this with the consistency coefficient of the corresponding image blocks in the adjacent previous frame within any given frame, the feature value of any given frame is obtained. Based on the aforementioned feature values, and combined with a filtering algorithm, the image of any given frame is filtered. The video storage and management module is used to store the filtered surveillance video, as well as the information and index of the surveillance video data; The monitoring screen display and operation module is used to display monitoring screens and customize the patrol scheme.

[0006] In one embodiment, the system further includes an automatic device discovery and configuration module for deploying industrial cameras with automatic configuration functions at each preset monitoring point. The industrial cameras are connected to the monitoring system via an industrial Ethernet. The automatic configuration server in the monitoring system automatically assigns network parameters and video parameters to the industrial cameras according to preset configuration rules. The video parameters include at least video encoding format, resolution, and frame rate.

[0007] In one embodiment, the video acquisition and transmission module employs a redundant link design and compresses the monitoring video during transmission.

[0008] In one embodiment, a firewall and intrusion detection system are deployed at the network entry point of the monitoring center to perform security checks on the transmitted surveillance videos.

[0009] In one embodiment, the weights are calculated as follows: Calculate the normalized value of the matching error for each image block; calculate the difference between 1 and the normalized value; Calculate the sum of the differences between all neighboring image blocks of each image block; The weights of each neighboring image block of each image block are directly proportional to the difference between the neighboring image blocks and inversely proportional to the accumulated value, wherein the sum of the weights of all neighboring image blocks of each image block is 1.

[0010] In one embodiment, the weight is the ratio of the difference between each neighboring image block to the accumulated value.

[0011] In one embodiment, obtaining the consistency coefficient includes: The consistency coefficient is a weighted sum of the similarities between each image patch and all its nearest neighbor image patches, wherein the weight of the similarity between each image patch and its nearest neighbor image patches is the weight of each nearest neighbor image patch.

[0012] In one embodiment, the process of obtaining the feature value is as follows: Calculate the mean of the matching error between each image patch and all its nearest neighbor image patches; Calculate the product of the consistency coefficient of each image block in any frame and the corresponding image block in the adjacent previous frame, and the mean value; The feature value is the sum of the products of all image blocks within each frame.

[0013] In one embodiment, filtering the any frame image includes: A Gaussian filter is used to filter each frame of the image. The standard deviation of the Gaussian filter is the sum of the normalized value of the feature value of each frame and the preset standard deviation, which is used as the standard deviation when the Gaussian filter is used to filter each frame of the image.

[0014] In one embodiment, the video storage and management module uses redundant independent disk array technology to configure the disk array, stores the filtered surveillance video to the disk array, and uses a database management system to store the information and index of the surveillance video data.

[0015] This application has at least the following beneficial effects: This application focuses on the problem of significant inter-frame interference differences caused by short-term environmental interference changes in actual industrial monitoring. By combining motion estimation search algorithms to analyze the consistency characteristics of motion directions in local regions of industrial monitoring videos within a short period of time, as well as the interference difference characteristics in local regions of inter-frames within a short period of time, feature values ​​of each frame of the industrial monitoring video are obtained. This reflects the significant impact of short-term dynamic changes in the actual industrial environment on the monitoring video, improving the accuracy of inter-frame interference analysis. Furthermore, the feature values ​​are used to perform precise optimized filtering processing on the industrial monitoring video. Compared with traditional filtering methods, this application avoids the problem of poor filtering effects caused by large inter-frame differences due to short-term interference in industrial monitoring, thereby improving the quality and accuracy of industrial video monitoring data. Attached Figure Description

[0016] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A block diagram of an industrial video surveillance system with automatic configuration function is provided in this application; Figure 2 This is a schematic diagram of the filtering process; Figure 3 This is a flowchart of an industrial video surveillance system with automatic configuration capabilities. Detailed Implementation

[0018] In the description of the embodiments in this application, the words "exemplary," "or," and "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary," "or," and "for example" is intended to present the relevant concepts in a specific manner.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. It should be understood that, unless otherwise stated, " / " in this application means "or".

[0020] It should also be noted that the terms "first" and "second" in this application are used to distinguish similar objects, rather than to describe a specific order or sequence.

[0021] The following description, in conjunction with the accompanying drawings, details a specific solution for an industrial video surveillance system with automatic configuration capabilities provided in this application.

[0022] Please see Figure 1 The diagram illustrates a block diagram of an industrial video surveillance system with automatic configuration function according to an embodiment of the present invention. The system includes: an automatic device discovery and configuration module 101, a video acquisition and transmission module 102, an intelligent analysis and processing module 103, a video storage and management module 104, and a monitoring screen display and operation module 105.

[0023] The device auto-discovery and configuration module 101 is used to deploy industrial cameras with auto-configuration function at each preset monitoring point. The industrial cameras are connected to the monitoring system via industrial Ethernet. The auto-configuration server in the monitoring system automatically assigns network parameters and video parameters to the industrial cameras according to the preset configuration rules.

[0024] Industrial cameras with automatic configuration capabilities are deployed at various monitoring points both underground and above ground in the coal mine. These cameras use standard network interfaces and connect to the monitoring system via industrial Ethernet. Upon successful connection, the cameras automatically send discovery signals carrying their own device information according to a specific protocol. The configuration management software on the automatic configuration server in the monitoring system continuously monitors these discovery signals on the network. When it receives a discovery signal from a camera, the automatic configuration server automatically assigns an IP address to the camera from a specified IP address range based on pre-defined configuration rules in the database, and simultaneously sets network parameters such as subnet mask and gateway. Simultaneously, the server configures the cameras according to the needs of different monitoring points. For example, coal mining faces require high resolution and high frame rates to clearly capture operational details, while ventilation rooms have relatively lower frame rate requirements but require specific video encoding formats for storage. The server automatically configures the appropriate video encoding format, resolution, and frame rate for each camera. After all the above configurations are completed, the camera quickly initializes, connects to the monitoring network, and begins real-time monitoring.

[0025] The video acquisition and transmission module 102 is used to acquire surveillance video in real time using a camera and transmit it to the monitoring center.

[0026] Real-time video images are captured using configured cameras, covering all aspects of coal mine production, including at least the coal face, tunnels, conveyor belts, and ventilation rooms. The captured video is transmitted via the mine's existing industrial Ethernet network, which provides a stable data transmission channel. Industrial Ethernet switches are strategically distributed across production areas for convenient camera access. Redundant links are used to ensure data transmission stability and reliability; if one link fails, data automatically switches to a backup link. During transmission, the H.265 encoding algorithm is used to compress the video. Specifically, the camera's built-in encoding chip compresses the raw video data according to the H.265 standard after capture. The Transmission Control Protocol (TCP) is used to transmit the compressed video data in an orderly manner, ensuring its complete and error-free arrival at the monitoring center. A firewall and intrusion detection system are deployed at the network entrance of the monitoring center to perform security checks on the incoming video data, preventing unauthorized external intrusion and network attacks, ensuring the security of video transmission, and thus guaranteeing stable and real-time video transmission to the monitoring center.

[0027] The intelligent analysis and processing module 103 is used to analyze the transmitted surveillance video and obtain the feature values ​​of each frame of the surveillance video; based on the feature values, and combined with the filtering algorithm, the frame of the video is filtered.

[0028] At the monitoring center, intelligent video analytics servers process and analyze the transmitted video data. Since the received video images are real-time monitoring images from the coal mine site, and the actual working environment of a coal mine is complex, the real-time acquired monitoring video may be affected by complex interference factors. Traditional techniques, when using filtering methods to filter the monitoring video frame by frame, ignore the problem of significant inter-frame differences caused by camera shake or rapid changes in lighting due to interference at the construction site within a short period. Traditional filtering methods cannot cope with the significant differences in inter-frame interference caused by rapid changes within a short period, resulting in generally poor filtering effects on real-time acquired monitoring video, affecting the video monitoring results of industrial construction sites under automatic configuration.

[0029] To address the aforementioned issues, the intelligent video analysis server in this application optimizes and adjusts the filtering process based on the characteristics of the received surveillance video, thereby improving the quality of the surveillance video and achieving efficient video surveillance in industrial settings. The specific process is as follows: (1) The motion vector and matching error of each image block in each frame of the surveillance video are obtained by using a motion estimation search algorithm. For any frame, the weight of each image block is obtained by the distribution of the matching error of all neighboring image blocks of each image block. The consistency coefficient of each image block is obtained by combining the similarity of the motion vector between each image block and its neighboring image blocks with the weight.

[0030] Considering that changes in interference during actual monitoring typically occur within a short period, leading to significant differences in inter-frame interference in the monitoring video, this paper, based on these characteristics, uses the acquired monitoring video as input and employs a three-step search algorithm to output the motion vectors and matching errors of each image block within each frame of the monitoring video. The method of obtaining the motion vectors and matching errors of each image block within each frame using the three-step search algorithm is a well-known technique and will not be elaborated upon here. The three-step search method divides each frame of the monitoring video into image blocks and calculates the motion vectors of these blocks to estimate the inter-frame motion direction. Typically, in actual industrial monitoring scenarios, local areas undergo dynamic changes within a short period, and these changes often have a certain directionality, meaning the local area moves in a specific direction. However, if interference factors change rapidly within a short time, it can lead to significant differences in inter-frame interference, resulting in a substantial increase in motion differences between frames within a short period.

[0031] In actual monitoring, to accurately reflect the impact of interference factors within a short period of time, taking any frame as an example, each image block is considered as the central region, and the image blocks within the 8-neighborhood of each image block are considered as the nearest neighbor image blocks of the central image block. The cosine similarity of the motion vectors between each image block and its nearest neighbor image blocks is calculated. A higher cosine similarity indicates a smaller difference in the impact of local interference within a short period of time during actual monitoring, and a more significant consistency in the dynamic change characteristics of the local region within the monitoring range. Considering that differences exist in local regions during actual motion estimation, which may lead to inaccurate judgment of inter-frame dynamic change characteristics, and that a larger matching error has a greater impact on the estimation of local dynamic change characteristics, the weights of the nearest neighbor image blocks of each image block are obtained based on the distribution of matching errors of all nearest neighbor image blocks of each image block. The expression is as follows: In the formula, This represents the weight of the x-th nearest neighbor image block of the v-th image block; This represents the normalized value of the matching error of the x-th nearest neighbor image block of the v-th image block; represents the normalized value of the matching error of the i-th nearest neighbor image block of the v-th image block; n represents the number of nearest neighbor image blocks of the v-th image block.

[0032] In this embodiment, the Min-Max normalization method is used to obtain the normalized value of the matching error. The Min-Max normalization method is a well-known technique and will not be described in detail in this application.

[0033] Based on the above analysis, the consistency coefficient of each image patch is further obtained by using the similarity of motion vectors between each image patch and its nearest neighbor image patches, as well as the weights of each nearest neighbor image patch. The expression is as follows: In the formula, represents the consistency coefficient of the v-th image patch; n represents the number of neighboring image patches of the v-th image patch; This represents the similarity of motion vectors between the v-th image patch and its x-th nearest neighbor image patch; This represents the weight of the x-th nearest neighbor image block of the v-th image block.

[0034] In this embodiment, the similarity between motion vectors is cosine similarity, which is a well-known technique and will not be described further in this application.

[0035] It should be noted that the smaller the calculated consistency coefficient, the greater the impact of short-term interference on the local area, as can be seen from the analysis of motion differences in local areas between frames during the monitoring process.

[0036] (2) By combining the distribution characteristics of the matching error of all image blocks in any frame image with the consistency coefficient of each pixel block in any frame image, the feature value of any frame image is obtained.

[0037] Furthermore, to accurately highlight the interference impact characteristics of each frame image, starting from the first frame of the surveillance video, the dynamic change characteristics of local regions between frames are analyzed as described above. The consistency coefficient of each image block within each frame image is calculated. Since the inter-frame difference is the dynamic change difference that occurs when the next frame image is subject to significant interference, the mean of the matching error between each image block in the next frame image and all its nearest neighbor image blocks is calculated for the local interference impact characteristics of the adjacent next frame image of each frame image. The larger the mean value, the greater the difference in interference impact of the same region between frames due to interference in the next frame image. Based on the dynamic change characteristics between frames and the difference in interference impact of adjacent next frames of each frame image, the feature value of the next frame image is obtained. The larger the feature value, the more significant the interference impact of short-term dynamic changes in the actual industrial environment on the surveillance video, and the more severe the inter-frame interference impact.

[0038] Based on the consistency coefficient of each image patch in each frame within the corresponding image patch in the adjacent previous frame, and the distribution characteristics of the matching error of all image patches in each frame, the feature values ​​of each frame are obtained, expressed as follows: ; represents the feature value of the j-th frame image; m represents the number of image blocks within the j-th frame image; This represents the mean of the matching errors between the t-th image block in the j-th frame and all its nearest neighbor image blocks; It represents the consistency coefficient of the image block corresponding to the t-th image block in the j-1-th image.

[0039] Furthermore, the feature values ​​of all frame images are used as input, and a normalization algorithm is used to obtain the normalized value of the feature values ​​of each frame image. The normalized value of the feature values ​​of each frame image is then used as the adjustment coefficient of each frame image to adjust the smoothness of filtering each frame image.

[0040] In this embodiment, the Min-Max normalization method is used to obtain the normalized values ​​of the feature values ​​of each frame image. The Min-Max normalization method is a well-known technique and will not be described in detail in this application.

[0041] (3) Based on the feature value, combined with the filtering algorithm, filter any frame image.

[0042] Based on the adjustment coefficients of each frame image, the standard deviation in the Gaussian filter is adjusted. The specific adjustment formula is as follows: In the formula, This represents the standard deviation in the Gaussian filter when filtering the j-th frame of the image; This represents the preset standard deviation in the Gaussian filter; This represents the adjustment coefficient for the j-th frame of the image. Since in industrial filtering processes, the standard deviation of a Gaussian filter is typically required to be no less than 1, this embodiment presets the standard deviation to be 1.

[0043] After calculating the standard deviation of the Gaussian filter used for filtering each frame of the image, the Gaussian filter is applied to each frame. The filtered video is then transmitted to the monitoring center for storage and real-time on-site monitoring. A schematic diagram of the filtering process is shown below. Figure 2 As shown.

[0044] The video storage and management module 104 is used to store the filtered surveillance video, as well as information and indexes of the surveillance video data.

[0045] After the filtered surveillance video data is transmitted to the monitoring center, the monitoring center uses a disk array for storage. Specifically, the disk array must first be properly configured, using Redundant Independent Disk Array (RAID) technology. By distributing data and parity information across multiple disks, it not only ensures data read and write performance but also provides a certain degree of fault tolerance, preventing data loss due to single disk failure. After configuration, the disk array is connected to the monitoring center's server, and the server's storage management software writes the video data to the designated storage space of the disk array.

[0046] Simultaneously, when the monitoring center receives video data, it extracts time information from the video data's metadata, such as the video recording start time. Monitoring area information can be pre-defined during the camera configuration phase and associated with the video data. For example, each camera corresponds to a specific coal mine production area, such as camera position 1 in the coal mining face corresponding to a specific location within the coal mining face. Furthermore, a MySQL database management system is used to store various information and indexes related to the monitoring video data for classification, management, and indexing. This includes creating corresponding data tables in the database, with table structures containing a unique identifier for the video data, a time field, a monitoring area field, and a storage path field. Once new video data is stored, the MySQL database management system automatically inserts the relevant classification and index information into the data tables. For example, if a video data is stored under a specific partition path in a disk array, with a corresponding time of "April 14, 2025, 10:00-10:10" and a monitoring area of ​​"Tunneling Roadway A," the time, monitoring area, and the unique identifier of the video data are inserted into the data table to create a fast query index.

[0047] The monitoring screen display and operation module 105 is used to display the monitoring screen and the custom patrol scheme.

[0048] For industrial site monitoring, large-screen displays are used, connected via HDMI high-definition video interfaces. After connection, the corresponding graphics card driver and display management software are installed on the server side to set the resolution and display mode parameters of the large-screen display to adapt to the video signal output by the monitoring system, thus presenting a clear picture.

[0049] Specifically, monitoring personnel can select different multi-screen split modes in the software interface according to their actual monitoring needs. These modes include at least 4-screen, 9-screen, and 16-screen splits. Depending on the selected mode, real-time video signals from different cameras are integrated and processed, then displayed on the monitoring terminal screen according to a preset layout. For example, when the 9-screen split mode is selected, the video feeds from nine different cameras are evenly distributed on the screen in a 3x3 grid, ensuring that each feed clearly displays the monitored area.

[0050] Furthermore, monitoring personnel can access the monitoring center's rotation settings interface to customize the rotation plan. Specifically, they first select the camera groups participating in the rotation, for example, setting all cameras in the coal mining face as one group and cameras in the tunneling roadways as another group; then, they set the rotation time interval for each group of cameras, for example, setting each group's camera feed to automatically switch to the next group after 10 seconds of display on the screen. A flowchart of an industrial video monitoring system with automatic configuration function is shown below. Figure 3 As shown.

[0051] In summary, this application focuses on the problem of significant inter-frame interference differences caused by short-term environmental interference changes in actual industrial monitoring. By combining motion estimation search algorithms to analyze the consistency characteristics of motion directions in local regions of industrial monitoring videos within a short time and the interference difference characteristics in local regions of inter-frame videos within a short time, feature values ​​of each frame of the industrial monitoring video are obtained. This reflects the significant impact of short-term dynamic changes in the actual industrial environment on the monitoring video, improving the accuracy of inter-frame interference analysis. Furthermore, the feature values ​​are used to perform precise optimized filtering processing on the industrial monitoring video. Compared with traditional filtering methods, this application avoids the problem of poor filtering effects caused by large inter-frame differences due to short-term interference in industrial monitoring, thereby improving the quality and accuracy of industrial video monitoring data.

[0052] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0053] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that this application can be implemented in other specific forms without departing from its essential characteristics. Therefore, the embodiments described above should be considered exemplary and non-limiting in all respects.

Claims

1. An industrial video surveillance system with automatic configuration function, characterized in that, The system contains: The video acquisition and transmission module is used to acquire surveillance video in real time using cameras and transmit it to the monitoring center; The intelligent analysis and processing module is used in the monitoring center to obtain the motion vectors and matching errors of each image block in each frame of the monitoring video using a motion estimation search algorithm. For any frame, the weights of each image block's nearest neighbor image blocks are obtained by analyzing the distribution of matching errors of all nearest neighbor image blocks of each image block. The consistency coefficient of each image block is obtained by combining the similarity of motion vectors between each image block and its nearest neighbor image blocks with the weights. By analyzing the distribution characteristics of the matching errors of all image blocks within any given frame, and combining this with the consistency coefficient of the corresponding image blocks in the adjacent previous frame within each image block in any given frame, the feature value of any given frame is obtained. Based on the aforementioned feature values, and combined with a filtering algorithm, the image of any given frame is filtered. The video storage and management module is used to store the filtered surveillance video, as well as the information and index of the surveillance video data; The monitoring screen display and operation module is used to display monitoring screens and customize the patrol scheme.

2. The industrial video surveillance system with automatic configuration function as described in claim 1, characterized in that, The system also includes an automatic device discovery and configuration module, which is used to deploy industrial cameras with automatic configuration functions at each preset monitoring point. The industrial cameras are connected to the monitoring system via industrial Ethernet. The automatic configuration server in the monitoring system automatically assigns network parameters and video parameters to the industrial cameras according to the preset configuration rules. The video parameters include at least video encoding format, resolution and frame rate.

3. The industrial video surveillance system with automatic configuration function as described in claim 1, characterized in that, The video acquisition and transmission module adopts a redundant link design and compresses the monitoring video during transmission.

4. An industrial video surveillance system with automatic configuration function as described in claim 1, characterized in that, The monitoring center is equipped with a firewall and intrusion detection system at its network entry point to perform security checks on the transmitted surveillance videos.

5. An industrial video surveillance system with automatic configuration function as described in claim 1, characterized in that, The weights are calculated as follows: Calculate the normalized value of the matching error for each image block; calculate the difference between 1 and the normalized value; Calculate the sum of the differences between all neighboring image blocks of each image block; The weights of each neighboring image block of each image block are directly proportional to the difference between the neighboring image blocks and inversely proportional to the accumulated value, wherein the sum of the weights of all neighboring image blocks of each image block is 1.

6. An industrial video surveillance system with automatic configuration function as described in claim 5, characterized in that, The weight is the ratio of the difference between each neighboring image block to the accumulated value.

7. An industrial video surveillance system with automatic configuration function as described in claim 1, characterized in that, The consistency coefficient is obtained by: The consistency coefficient is a weighted sum of the similarities between each image patch and all its nearest neighbor image patches, wherein the weight of the similarity between each image patch and its nearest neighbor image patches is the weight of each nearest neighbor image patch.

8. An industrial video surveillance system with automatic configuration function as described in claim 1, characterized in that, The process of obtaining the feature values ​​is as follows: Calculate the mean of the matching error between each image patch and all its nearest neighbor image patches; Calculate the product of the consistency coefficient of each image block in any frame and the corresponding image block in the adjacent previous frame, and the mean value; The feature value is the sum of the products of all image blocks within each frame.

9. An industrial video surveillance system with automatic configuration function as described in claim 1, characterized in that, The filtering of any one of the frame images includes: A Gaussian filter is used to filter each frame of the image. The standard deviation of the Gaussian filter is the sum of the normalized value of the feature value of each frame and the preset standard deviation, which is used as the standard deviation when the Gaussian filter is used to filter each frame of the image.

10. An industrial video surveillance system with automatic configuration function as described in claim 1, characterized in that, The video storage and management module uses redundant independent disk array technology to configure the disk array, stores the filtered surveillance video to the disk array, and uses a database management system to store the information and index of the surveillance video data.

Images