Video monitoring management method based on intranet computer

Abstract

The invention discloses a video monitoring management method based on an intranet computer, and relates to the technical field of computer video monitoring, and the system comprises the following components: a video analysis module which carries out the deep analysis of collected video data, recognizes a target object in a video through a target detection algorithm, and carries out the recognition of the target object through a behavior analysis algorithm; whether the behavior of the target object is abnormal or not is judged, key information is provided for subsequent intelligent tracking and positioning, the video data is processed and analyzed through the video analysis module, the abnormal behavior or the specific target is quickly and accurately recognized through the target detection algorithm and the behavior analysis algorithm, and the automatic video analysis process has the advantages of being high in efficiency and high in reliability. The efficiency and the accuracy of video monitoring are greatly improved, so that a manager can obtain key information more quickly and make a timely decision.

Description

Technical Field

[0001] The invention relates to the technical field of computer video monitoring, in particular to a video monitoring management method based on an intranet computer. Background Art

[0002] With the rapid development of information technology, video surveillance has become an important part of the security management system of modern society. Due to the rapid development of information technology, computer vision and artificial intelligence, the performance of digital cameras, network transmission equipment, and storage media hardware equipment has been continuously improved, providing high-quality images and video data for video surveillance. However, traditional video surveillance mainly relies on manual real-time monitoring, which is not only labor-intensive, but also prone to monitoring errors due to human factors. In addition, most traditional video surveillance management methods lack the function of recording and analyzing the target's movement trajectory. When a security incident occurs, it often takes a lot of time and energy to replay and analyze video data to find key clues. This post-processing method is not only inefficient, but also may miss the best time for emergency response.

[0003] Therefore, how to combine computer technology and video analysis technology to realize a video surveillance management method that can automatically track and locate targets and improve the real-time, accuracy and intelligence level of video surveillance is an important research direction in the current field of video surveillance technology. An ideal video surveillance management method based on an intranet computer should have the following characteristics: it can detect targets through automatic processing and analysis of video data; it can combine map visualization technology with alarm notification mechanism to provide managers with intuitive and efficient monitoring and management means.

[0004] However, most of the intranet computer-based video surveillance management methods on the market can only meet some of the requirements of the above-mentioned characteristics and cannot achieve target positioning and automatic tracking at the same time. Therefore, developing a video surveillance management method based on an intranet computer that can comprehensively realize the above-mentioned characteristics is of great significance for improving the real-time, accuracy and intelligence level of video surveillance. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the prior art and provide a video surveillance management method based on an intranet computer. The video data of the monitored area is collected in real time through a camera connected to the intranet computer, and the video data is processed and analyzed using a video surveillance management system. The system can automatically detect and identify abnormal behavior or specific targets in the video. When the system detects abnormal behavior or specific targets, it will immediately start the intelligent tracking function. Through an efficient target tracking algorithm, the system can achieve continuous tracking of the target object and accurately record its movement trajectory, providing managers with real-time and accurate monitoring information, which helps to promptly discover and handle abnormal situations.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a video surveillance management method, characterized in that the method comprises the following modules:

[0007] Video acquisition module: responsible for real-time acquisition of video data in the monitoring area. Through camera equipment, it captures dynamic images of the monitoring area and transmits them to the video monitoring management system to provide raw materials for subsequent video analysis.

[0008] Video analysis module: conducts in-depth analysis of the collected video data, identifies the target object in the video through the target detection algorithm, and determines whether the target object's behavior is abnormal through the behavior analysis algorithm, providing key information for subsequent intelligent tracking and positioning;

[0009] Intelligent tracking module: responsible for continuously tracking these targets after the video analysis module finds abnormal behaviors or specific targets. It uses the target tracking algorithm to ensure that the target objects can be tracked stably and accurately and record their movement trajectories.

[0010] Positioning and display module: Match the mobile trajectory data obtained by the intelligent tracking module with the map data to achieve real-time positioning of the target. Through map visualization technology, the location information of the target can be intuitively displayed on the map, helping managers to quickly understand the monitoring situation and make corresponding processing decisions;

[0011] Alarm and notification module: When abnormal behavior or specific targets are detected, the alarm mechanism can be automatically triggered and an alarm notification can be sent to relevant personnel to ensure that abnormal situations are handled quickly and improve the speed and efficiency of emergency response.

[0012] Furthermore, the specific steps of the video acquisition module are:

[0013] (1) Device initialization: Initialize the camera device and set the camera resolution and frame rate parameters;

[0014] (2) Data acquisition: collect video data in real time and transmit it to the video surveillance management system.

[0015] Furthermore, the video analysis module specifically comprises the following steps:

[0016] (1) Target recognition: Use target detection algorithms to identify target objects in the video;

[0017] (2) Feature extraction: Apply behavioral analysis algorithms to extract and classify the target's behavior to determine whether its behavior is abnormal.

[0018] Furthermore, the smart tracking module specifically comprises the following steps:

[0019] (1) Target locking: After the video analysis module detects abnormal behavior or a specific target, it selects the target to be tracked;

[0020] (2) Intelligent tracking: Utilize the target tracking algorithm to ensure that the target object can be tracked stably and accurately and its movement trajectory can be recorded.

[0021] Furthermore, the positioning and display module comprises the following specific steps:

[0022] (1) Real-time positioning: Match the movement trajectory data obtained by the intelligent tracking module with the map data to achieve real-time positioning of the target;

[0023] (2) Target display: Through map visualization technology, the location information of the target is intuitively displayed on the map, helping managers to quickly understand the monitoring situation and make corresponding processing decisions;

[0024] Furthermore, the alarm and notification module includes the following specific steps:

[0025] (1) Automatic alarm: When abnormal behavior or specific targets are detected, the alarm mechanism can be automatically triggered;

[0026] (2) Alarm notification: When the alarm mechanism is triggered, a notification will be sent to relevant personnel to ensure that abnormal situations are handled quickly, thereby improving the speed and efficiency of emergency response.

[0027] Furthermore, the video analysis module performs target detection through the FasterR-CNN algorithm. The formula of the FasterR-CNN algorithm is: a convolutional neural network (CNN) is used to extract features from the input image frame, which usually uses a convolution operation, and its formula is: S(i,j)=(I*K)(i,j)=∑ m ∑ n I(im,jn)K(m,n), where (I) is the input image, (K) is the convolution kernel, and (S) is the feature map after convolution. Next, the region proposal network (RPN) generates a series of candidate regions (anchors) on the extracted feature map. RPN performs binary classification (foreground or background) and bounding box regression on each anchor to predict the area where the target may exist and its precise location. Logistic regression is usually used for binary classification, and its formula is: Where (x) is the feature vector, (β0) and (β1) are model parameters, and the goal of bounding box regression is to predict the offset between the anchor and the true target box, which is usually achieved through linear regression, and its formula is: * =W T φ(A), where (t *) is the predicted offset, W is the regression parameter, (φ(A)) is the feature representation of the anchor. In order to eliminate overlapping candidate boxes, non-maximum suppression is usually used. The candidate boxes are sorted based on their scores and other boxes that overlap too much with the highest scoring box are suppressed. Finally, the video analysis unit outputs the result of target detection.

[0028] Furthermore, the video surveillance management method based on the intranet computer according to claim 4 is characterized in that the intelligent tracking module intelligently tracks the target through the optical flow method, and the formula of the optical flow method is: extract feature points, and the feature point set is: F = {f1, f2, ..., f n}, where f i represents the i-th feature point, n is the total number of feature points, and the Lucas-Kanade method is used to calculate the optical flow. The basic equation of the optical flow can be expressed as: I(x+Δx,y+Δy,t+Δt)=I(x,y,t), where I is the image intensity function, (x,y) is the pixel position, t is the time, (Δx,Δy) is the optical flow vector, and by Taylor expansion and ignoring high-order terms, we get: I x Δx+I y Δy+I t Δt=0, where Ix, Iy, It are the partial derivatives of the image intensity with respect to x, y, and t, respectively. For multiple points in a small window, we can obtain an overdetermined linear equation system and use the least squares method to solve the optical flow vector (Δx, Δy). For target tracking, assuming that we have obtained the feature point set Ft in the previous frame and the feature point set Ft+1 in the current frame, the matching process can be expressed as finding a mapping: M:F t →F t+1 , the trajectory record can simply store the position of the feature points in each frame. Assume represents the position of the i-th feature point in the t-th frame, then the trajectory can be expressed as: Where T is the total number of frames in the video. Anomaly detection is usually based on trajectory analysis. For example, we can calculate the velocity and acceleration of the trajectory and compare them with a preset threshold: if or If the motion exceeds a certain threshold, the motion of the feature point is considered abnormal.

[0029] Furthermore, the system also includes a storage module, and the storage module adopts a cloud storage method.

[0030] Furthermore, the specific steps of the method are:

[0031] (1) System initialization and setup: Connect the camera to the intranet computer to ensure that the video data can be transmitted to the video surveillance management system on the intranet computer in real time. Install and configure the video surveillance management system on the intranet computer.

[0032] (2) Video acquisition and transmission: Turn on the camera and start collecting video data of the monitored area. The video data is transmitted to the video surveillance management system in real time via the intranet to ensure the accuracy and timeliness of the data.

[0033] (3) Video analysis and anomaly detection

[0034] After receiving the video data, the video surveillance management system uses video analysis technology to process and analyze the data, identifies the target object in the video through the target detection algorithm, and uses the behavior analysis algorithm to analyze the behavior of the target object to determine whether there is any abnormal behavior.

[0035] (4) Intelligent tracking and motion trajectory recording: When the video surveillance management system detects abnormal behavior or a specific target, it automatically starts the intelligent tracking function and uses the target tracking algorithm to continuously track the target object, ensuring that it can still be accurately identified when the target moves or changes its posture, and recording the movement trajectory of the target object.

[0036] (5) Positioning display and map matching: The movement trajectory of the tracked target object is matched with the map data stored in the intranet computer, and the current position of the target and the historical record of the movement trajectory are displayed on the map in real time. Through map visualization technology, managers can intuitively understand the movement of the target, which facilitates rapid positioning and handling of abnormal situations.

[0037] (6) Alarm and notification: When the video surveillance management system detects abnormal behavior or specific targets, the alarm mechanism is automatically triggered and an alarm notification is sent to relevant personnel based on the preset alarm conditions. After receiving the alarm notification, the relevant personnel can respond quickly and take appropriate measures.

[0038] (7) Data storage and query: The video surveillance management system stores the collected video data, analysis results, and tracking trajectories for subsequent query and analysis. Managers can use the query function provided by the system to provide strong support for security management.

[0039] Compared with the prior art, the video surveillance management method based on the intranet computer has the following beneficial effects:

[0040] 1. The present invention adopts a target tracking algorithm through an intelligent analysis module to continuously and stably track the detected target object and record its movement trajectory, so that managers can clearly understand the target's action path and movement, which is helpful for in-depth analysis and judgment of abnormal behavior or specific targets. At the same time, the intelligent tracking function can also be combined with an alarm mechanism. Once the target leaves the preset safety area or triggers other alarm conditions, the system can immediately issue an alarm, thereby improving the timeliness and effectiveness of security monitoring.

[0041] 2. The present invention processes and analyzes video data through a video analysis module, and quickly and accurately identifies abnormal behaviors or specific targets through target detection algorithms and behavior analysis algorithms. This automated video analysis process greatly improves the efficiency and accuracy of video surveillance, allowing managers to obtain key information more quickly and make timely decisions.

[0042] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 This is a flow chart of the video surveillance management method based on an intranet computer. DETAILED DESCRIPTION

[0045] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] Example 1

[0047] A video surveillance management method based on an intranet computer, the method comprises the following modules:

[0048] Video acquisition module: responsible for real-time acquisition of video data in the monitoring area. Through camera equipment, it captures dynamic images of the monitoring area and transmits them to the video monitoring management system to provide raw materials for subsequent video analysis.

[0049] Video analysis module: conducts in-depth analysis of the collected video data, identifies the target object in the video through the target detection algorithm, and determines whether the target object's behavior is abnormal through the behavior analysis algorithm, providing key information for subsequent intelligent tracking and positioning;

[0050] Intelligent tracking module: responsible for continuously tracking these targets after the video analysis module finds abnormal behaviors or specific targets. It uses the target tracking algorithm to ensure that the target objects can be tracked stably and accurately and record their movement trajectories.

[0051] Positioning and display module: Match the mobile trajectory data obtained by the intelligent tracking module with the map data to achieve real-time positioning of the target. Through map visualization technology, the location information of the target can be intuitively displayed on the map, helping managers to quickly understand the monitoring situation and make corresponding processing decisions;

[0052] Alarm and notification module: When abnormal behavior or specific targets are detected, the alarm mechanism can be automatically triggered and an alarm notification can be sent to relevant personnel to ensure that abnormal situations are handled quickly and improve the speed and efficiency of emergency response.

[0053] The video analysis module performs target detection through the FasterR-CNN algorithm. The formula of the FasterR-CNN algorithm is: Use a convolutional neural network (CNN) to extract features from the input image frame, which usually uses a convolution operation. The formula is: S(i,j)=(I*K)(i,j)=∑ m ∑ n I(im,jn)K(m,n), where (I) is the input image, (K) is the convolution kernel, and (S) is the feature map after convolution. Next, the region proposal network (RPN) generates a series of candidate regions (anchors) on the extracted feature map. RPN performs binary classification (foreground or background) and bounding box regression on each anchor to predict the area where the target may exist and its precise location. Logistic regression is usually used for binary classification, and its formula is: Where (x) is the feature vector, (β0) and (β1) are model parameters, and the goal of bounding box regression is to predict the offset between the anchor and the true target box, which is usually achieved through linear regression, and its formula is: * =W T φ(A), where (t *) is the predicted offset, W is the regression parameter, (φ(A)) is the feature representation of the anchor. In order to eliminate overlapping candidate boxes, non-maximum suppression is usually used. The candidate boxes are sorted based on their scores and other boxes that overlap too much with the highest scoring box are suppressed. Finally, the video analysis unit outputs the result of target detection.

[0054] The algorithm adopted by the video analysis module can better handle multi-scale and small target problems, while achieving a relatively high frame rate to meet the needs of real-time video surveillance. When the target is affected by multiple factors such as lighting changes, occlusion, and perspective changes, it can maintain a high detection accuracy in these complex environments and can adapt to different application scenarios and needs.

[0055] Example 2

[0056] A video surveillance management method based on an intranet computer, the method comprises the following modules:

[0057] Video acquisition module: responsible for real-time acquisition of video data in the monitoring area. Through camera equipment, it captures dynamic images of the monitoring area and transmits them to the video monitoring management system to provide raw materials for subsequent video analysis.

[0058] Video analysis module: conducts in-depth analysis of the collected video data, identifies the target object in the video through the target detection algorithm, and determines whether the target object's behavior is abnormal through the behavior analysis algorithm, providing key information for subsequent intelligent tracking and positioning;

[0059] Intelligent tracking module: responsible for continuously tracking these targets after the video analysis module finds abnormal behaviors or specific targets. It uses the target tracking algorithm to ensure that the target objects can be tracked stably and accurately and record their movement trajectories.

[0060] Positioning and display module: Match the mobile trajectory data obtained by the intelligent tracking module with the map data to achieve real-time positioning of the target. Through map visualization technology, the location information of the target can be intuitively displayed on the map, helping managers to quickly understand the monitoring situation and make corresponding processing decisions;

[0061] Alarm and notification module: When abnormal behavior or specific targets are detected, the alarm mechanism can be automatically triggered and an alarm notification can be sent to relevant personnel to ensure that abnormal situations are handled quickly and improve the speed and efficiency of emergency response.

[0062] The intelligent tracking module uses the optical flow method to intelligently track the target. The formula of the optical flow method is: Extract feature points, and the feature point set is: F = {f1, f2, ..., f n}, where f irepresents the i-th feature point, n is the total number of feature points, and the Lucas-Kanade method is used to calculate the optical flow. The basic equation of the optical flow can be expressed as: I(x+Δx,y+Δy,t+Δt)=I(x,y,t), where I is the image intensity function, (x,y) is the pixel position, t is the time, (Δx,Δy) is the optical flow vector, and by Taylor expansion and ignoring high-order terms, we get: I x Δx+I y Δy+I t Δt=0, where Ix, Iy, It are the partial derivatives of the image intensity with respect to x, y, and t, respectively. For multiple points in a small window, we can obtain an overdetermined linear equation system and use the least squares method to solve the optical flow vector (Δx, Δy). For target tracking, assuming that we have obtained the feature point set Ft in the previous frame and the feature point set Ft+1 in the current frame, the matching process can be expressed as finding a mapping: M:F t →F t+1 , the trajectory record can simply store the position of the feature points in each frame. Assume represents the position of the i-th feature point in the t-th frame, then the trajectory can be expressed as: Where T is the total number of frames in the video. Anomaly detection is usually based on trajectory analysis. For example, we can calculate the velocity and acceleration of the trajectory and compare them with a preset threshold: if or If the motion exceeds a certain threshold, the motion of the feature point is considered abnormal.

[0063] The intelligent tracking module can track multiple targets in the video at the same time. By processing the tracking tasks of multiple targets in parallel, the efficiency and accuracy of the monitoring system can be greatly improved. At the same time, it can perform anomaly detection based on trajectory analysis. Once abnormal behavior or specific targets are detected, the alarm mechanism can be automatically triggered and an alarm notification can be sent to relevant personnel. This helps to promptly discover and deal with potential safety hazards and improve the speed and efficiency of emergency response.

[0064] Example 3

[0065] The specific steps of the method based on Example 1 and Example 2 are as follows:

[0066] (1) System initialization and setup: Connect the camera to the intranet computer to ensure that the video data can be transmitted to the video surveillance management system on the intranet computer in real time. Install and configure the video surveillance management system on the intranet computer.

[0067] (2) Video acquisition and transmission: Turn on the camera and start collecting video data of the monitored area. The video data is transmitted to the video surveillance management system in real time via the intranet to ensure the accuracy and timeliness of the data.

[0068] (3) Video analysis and anomaly detection: After receiving video data, the video surveillance management system uses video analysis technology to process and analyze the data, identifies the target object in the video through the target detection algorithm, and uses the behavior analysis algorithm to analyze the behavior of the target object to determine whether there is any abnormal behavior.

[0069] (4) Intelligent tracking and motion trajectory recording: When the video surveillance management system detects abnormal behavior or a specific target, it automatically starts the intelligent tracking function and uses the target tracking algorithm to continuously track the target object, ensuring that it can still be accurately identified when the target moves or changes its posture, and recording the movement trajectory of the target object.

[0070] (5) Positioning display and map matching: The movement trajectory of the tracked target object is matched with the map data stored in the intranet computer, and the current position of the target and the historical record of the movement trajectory are displayed on the map in real time. Through map visualization technology, managers can intuitively understand the movement of the target, which facilitates rapid positioning and handling of abnormal situations.

[0071] (6) Alarm and notification: When the video surveillance management system detects abnormal behavior or specific targets, the alarm mechanism is automatically triggered and an alarm notification is sent to relevant personnel based on the preset alarm conditions. After receiving the alarm notification, the relevant personnel can respond quickly and take appropriate measures.

[0072] (7) Data storage and query: The video surveillance management system stores the collected video data, analysis results, and tracking trajectories for subsequent query and analysis. Managers can use the query function provided by the system to provide strong support for security management.

[0073] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A video monitoring management method based on an intranet computer, characterized in that: The method consists of the following modules: Video acquisition module: responsible for real-time acquisition of video data in the monitoring area. Through camera equipment, it captures dynamic images of the monitoring area and transmits them to the video monitoring management system to provide raw materials for subsequent video analysis. Video analysis module: conducts in-depth analysis of the collected video data, identifies the target object in the video through the target detection algorithm, and determines whether the target object's behavior is abnormal through the behavior analysis algorithm, providing key information for subsequent intelligent tracking and positioning; Intelligent tracking module: responsible for continuously tracking these targets after the video analysis module finds abnormal behaviors or specific targets. It uses the target tracking algorithm to ensure that the target objects can be tracked stably and accurately and record their movement trajectories. Positioning and display module: Match the mobile trajectory data obtained by the intelligent tracking module with the map data to achieve real-time positioning of the target. Through map visualization technology, the location information of the target can be intuitively displayed on the map, helping managers to quickly understand the monitoring situation and make corresponding processing decisions; Alarm and notification module: When abnormal behavior or specific targets are detected, the alarm mechanism can be automatically triggered and an alarm notification can be sent to relevant personnel to ensure that abnormal situations are handled quickly and improve the speed and efficiency of emergency response.

2. The video surveillance management method based on an intranet computer according to claim 1, characterized in that: The specific steps of the video acquisition module are as follows: (1) Device initialization: Initialize the camera device and set the camera resolution and frame rate parameters; (2) Data acquisition: collect video data in real time and transmit it to the video surveillance management system.

3. The video monitoring management method based on an intranet computer according to claim 1 is characterized in that: The specific steps of the video analysis module are as follows: (1) Target recognition: Use target detection algorithms to identify target objects in the video; (2) Feature extraction: Apply behavioral analysis algorithms to extract and classify the target's behavior to determine whether its behavior is abnormal.

4. The video surveillance management method based on an intranet computer according to claim 1, characterized in that: The specific steps of the intelligent tracking module are as follows: (1) Target locking: After the video analysis module detects abnormal behavior or a specific target, it selects the target to be tracked; (2) Intelligent tracking: Utilize the target tracking algorithm to ensure that the target object can be tracked stably and accurately and its movement trajectory can be recorded.

5. The video monitoring management method based on an intranet computer according to claim 1 is characterized in that: The specific steps of the positioning and display module are as follows: (1) Real-time positioning: Match the movement trajectory data obtained by the intelligent tracking module with the map data to achieve real-time positioning of the target; (2) Target display: Through map visualization technology, the location information of the target is intuitively displayed on the map, helping managers to quickly understand the monitoring situation and make corresponding processing decisions.

6. The video surveillance management method based on an intranet computer according to claim 1, characterized in that: The specific steps of the alarm and notification module are as follows: (1) Automatic alarm: When abnormal behavior or specific targets are detected, the alarm mechanism can be automatically triggered; (2) Alarm notification: When the alarm mechanism is triggered, a notification will be sent to relevant personnel to ensure that abnormal situations are handled quickly, thereby improving the speed and efficiency of emergency response.

7. The video monitoring management method based on an intranet computer according to claim 3 is characterized in that: The video analysis module performs target detection through the FasterR-CNN algorithm. The formula of the FasterR-CNN algorithm is: a convolutional neural network (CNN) is used to extract features from the input image frame, which usually uses a convolution operation. The formula is: S(i,j)=(I*K)(i,j)=∑ m ∑ n I(im,jn)K(m,n), where (I) is the input image, (K) is the convolution kernel, and (S) is the feature map after convolution. Next, the region proposal network (RPN) generates a series of candidate regions (anchors) on the extracted feature map. RPN performs binary classification (foreground or background) and bounding box regression on each anchor to predict the area where the target may exist and its precise location. Logistic regression is usually used for binary classification, and its formula is: Where (x) is the feature vector, (β0) and (β1) are model parameters, and the goal of bounding box regression is to predict the offset between the anchor and the true target box, which is usually achieved through linear regression, and its formula is: * =W T φ(A), where (t * ) is the predicted offset, W is the regression parameter, (φ(A)) is the feature representation of the anchor. In order to eliminate overlapping candidate boxes, non-maximum suppression is usually used. The candidate boxes are sorted based on their scores and other boxes that overlap too much with the highest scoring box are suppressed. Finally, the video analysis unit outputs the result of target detection.

8. The video monitoring management method based on an intranet computer according to claim 4 is characterized in that: The intelligent tracking module intelligently tracks the target by using the optical flow method. The formula of the optical flow method is: extract feature points, and the feature point set is: F = {f1, f2, ..., f n }, where f i represents the i-th feature point, n is the total number of feature points, and the Lucas-Kanade method is used to calculate the optical flow. The basic equation of the optical flow can be expressed as: I(x+Δx,y+Δy,t+Δt)=I(x,y,t), where I is the image intensity function, (x,y) is the pixel position, t is the time, (Δx,Δy) is the optical flow vector, and by Taylor expansion and ignoring high-order terms, we get: I x Δx+I y Δy+I t Δt=0, where Ix, Iy, It are the partial derivatives of the image intensity with respect to x, y, and t, respectively. For multiple points in a small window, we can obtain an overdetermined linear equation system and use the least squares method to solve the optical flow vector (Δx, Δy). For target tracking, assuming that we have obtained the feature point set Ft in the previous frame and the feature point set Ft+1 in the current frame, the matching process can be expressed as finding a mapping: M:F t →F t+1 , the trajectory record can simply store the position of the feature points in each frame. Assume represents the position of the i-th feature point in the t-th frame, then the trajectory can be expressed as: Where T is the total number of frames in the video. Anomaly detection is usually based on trajectory analysis. For example, we can calculate the velocity and acceleration of the trajectory and compare them with a preset threshold: if or If the motion exceeds a certain threshold, the motion of the feature point is considered abnormal.

9. The video monitoring management method based on an intranet computer according to claim 1, characterized in that: The system also includes a storage module, which adopts cloud storage.

10. The video surveillance management method based on an intranet computer according to claim 1, wherein the specific steps of the method are: (1) System initialization and setup: Connect the camera to the intranet computer to ensure that the video data can be transmitted to the video surveillance management system on the intranet computer in real time. Install and configure the video surveillance management system on the intranet computer. (2) Video acquisition and transmission: Turn on the camera and start collecting video data of the monitored area. The video data is transmitted to the video surveillance management system in real time via the intranet to ensure the accuracy and timeliness of the data. (3) Video analysis and anomaly detection: After receiving video data, the video surveillance management system uses video analysis technology to process and analyze the data, identifies the target object in the video through the target detection algorithm, and uses the behavior analysis algorithm to analyze the behavior of the target object to determine whether there is any abnormal behavior. (4) Intelligent tracking and motion trajectory recording: When the video surveillance management system detects abnormal behavior or a specific target, it automatically starts the intelligent tracking function and uses the target tracking algorithm to continuously track the target object, ensuring that it can still be accurately identified when the target moves or changes its posture, and recording the movement trajectory of the target object. (5) Positioning display and map matching: The movement trajectory of the tracked target object is matched with the map data stored in the intranet computer, and the current position of the target and the historical record of the movement trajectory are displayed on the map in real time. Through map visualization technology, managers can intuitively understand the movement of the target, which facilitates rapid positioning and handling of abnormal situations. (6) Alarm and notification: When the video surveillance management system detects abnormal behavior or specific targets, the alarm mechanism is automatically triggered and an alarm notification is sent to relevant personnel based on the preset alarm conditions. After receiving the alarm notification, the relevant personnel can respond quickly and take appropriate measures. (7) Data storage and query: The video surveillance management system stores the collected video data, analysis results, and tracking trajectories for subsequent query and analysis. Managers can use the query function provided by the system to provide strong support for security management.

Images