Active and passive combined video monitoring system fault positioning method

Abstract

The invention discloses an active and passive combined video monitoring system fault positioning method, which comprises the following steps of: S1, acquiring a real-time monitoring video stream, and acquiring video frame data, a video frame statistical index and video stream session information from the real-time monitoring video stream; s2, analyzing the video frame data to obtain a video fault index; s3, detecting a video server and a network path condition to obtain a network degradation condition of the network node; and S4, carrying out association analysis by taking the video fault index, the network path and the network degradation condition as data sources, and positioning the fault. According to the invention, from user perception, accurate positioning of video faults is realized, network camera faults, network transmission node faults and server faults can be positioned, an accurate monitoring means is provided for rapid repair of the video faults, and the video fault inspection and repair cost is greatly reduced.

Description

Technical Field

[0001] The invention relates to the technical field of video fault detection, and in particular to a fault locating method for a video monitoring system combining active and passive functions. Background Art

[0002] Video surveillance systems are widely used. When a video surveillance system fails, it is often difficult to accurately locate whether it is a camera failure, a network failure, or a server failure. The current technical solutions for locating video surveillance system faults only partially use network diagnostic technology or server diagnostic technology. This can only find problems with the network or server, but is separated from the diagnosis and analysis of the actual business video stream of video surveillance. It is impossible to establish a complete monitoring link, and therefore it is impossible to truly accurately locate the fault of the video surveillance system. Summary of the invention

[0003] In order to solve the above problems, the present invention provides a method for locating faults in a video surveillance system combining active and passive functions, comprising the following steps: S1. acquiring a real-time monitoring video stream, and acquiring video frame data, video frame statistical indicators and video stream session information from the real-time monitoring video stream; S2. analyzing the video frame data to acquire video fault indicators; S3. detecting the video server and the network path conditions to acquire the network degradation conditions of the network nodes; S4. performing correlation analysis based on the video fault indicators, the network path and the network degradation conditions as data sources to locate the fault.

[0004] Furthermore, the video frame data in the S1 step specifically includes: key frames, non-key frames; the video frame statistical indicators specifically include: key frame loss rate, non-key frame loss rate, key frame packet loss rate, loss duration; the video stream session information specifically includes: source address, source port, destination address, destination port.

[0005] Furthermore, the step S1 specifically includes the following sub-steps: S11. Obtain video stream IP packet data by real-time packet capture; S12. Perform data analysis on the captured IP packet data: protocol identification, data analysis, and indicator calculation; The protocol identification specifically includes: identifying the transport layer type of the data packet, obtaining the video stream session information, extracting the characteristic fields of the video data packet and calculating the unique ID data identifier of each video stream; The data analysis specifically includes: generating multiple types of video data packet feature codes according to the feature field combination of the extracted video data packet, and establishing the feature video data packet data analysis mode function according to the data packet feature code; The calculation formula in the indicator calculation specifically includes: Theoretical number of key frames = 300*frame rate / group length; Key frame loss rate = (theoretical key frame number - actual key frame number) / theoretical key frame number; Theoretical number of non-key frames = 300*frame rate-300*frame rate / group length; Non-key frame loss rate = (theoretical non-key frame number - actual non-key frame number) / theoretical non-key frame number; Key frame subpacket loss rate = number of key frame subpacket losses / total number of key frame subpackets, where the number of key frame subpacket losses = number of missing subpacket sequence numbers; Frame drop duration = time of receiving a new frame - time of frame drop start.

[0006] Furthermore, the calculation formula for generating the video data packet feature code is specifically: (low 5 bits of the first byte of the data packet & the low 5 bits of the second byte of the data packet) | (high 3 bits of the first byte of the data packet & the high 3 bits of the second byte of the data packet).

[0007] Furthermore, the established characteristic video data packet data parsing mode function is used to complete data parsing of a specific video data packet, and reverse data parsing according to the data encapsulation rules of the characteristic video data packet specifically includes the following steps: A1. Based on the characteristic field, determine whether the data packet is single-packet data or sub-packet data; A2. Cache the single-packet data and cache the sub-packet data after extracting the data body; A3. Based on the characteristic field, determine the frame type, and accumulate the actual number of key frames and the actual number of non-key frames.

[0008] Furthermore, the video fault indicators in the step S2 specifically include: video interruption, video freeze, video distorted screen, and video black screen.

[0009] Furthermore, the steps for judging each video fault indicator in step S2 are respectively: Video interruption: Record the appearance time of the first frame. When the video data cannot be decoded, start accumulating time until a new normal frame is obtained. When the accumulated time is greater than the standard key frame interval, it is judged as a video interruption. Video freeze: Record the appearance time of the first frame. When the time intervals before and after the appearance of 5 consecutive decoded frames are less than the standard I-frame interval, it is judged as video freeze. When the non-key frame loss rate is greater than the set threshold and the frame loss duration is less than the set threshold, it is also judged as video freeze. Video distorted screen: Detects whether a specific area of ​​the decoded image is a pure color area, while the remaining area is a non-pure color area. If both conditions are met, it is judged as a video distorted screen. When the upper or lower half of an image is a pure color area and the other half is a non-pure color area, it is perceived as partial loss of image content, and it is judged as a video distorted screen. Video black screen: Detects whether the decoded image is a solid color image. If it is a solid color image, it is perceived that all image content is lost and the video is judged to be black screen.

[0010] Furthermore, the specific steps of determining whether an image is a pure color area include: Z1. binarizing and inverting the grayscale image after grayscale transformation; Z2. counting the number and proportion of white pixels in a specific area: when the white pixel ratio is greater than 0.999, the area is determined to be a pure color area; when the white pixel ratio is less than 0.001, the area is determined to be a pure color area; when the white pixel ratio is greater than 0.001 and less than 0.999, the area is determined to be a non-pure color area.

[0011] Furthermore, the association rules for the association analysis in step S4 are specifically: When the fault is video interruption, find the associated I frame loss rate and frame loss duration, and then query the end-to-end path information, node delay and node packet loss. If the delay of all nodes on the path on the network camera side and the server side is less than 50ms, and the node packet loss rate is less than 5%, it can be determined that the network is normal and the state of data frames arriving at the target server is normal. It is determined that the network camera is abnormal and caused the interruption; when any network node on the end-to-end path has a delay of >100ms, or a node packet loss rate of >10%, it is directly determined that the network is abnormal and the abnormal network node on the end-to-end path is located. The abnormality of a certain network node causes the interruption; When the fault is video freeze, find the associated P frame loss rate and frame loss duration, and then query the end-to-end path information, node delay and node packet loss. If the delay of all nodes on the path of the network camera side and the server side is less than 50ms, and the node packet loss rate is less than 5%, the network is judged to be normal, and the state of data frames arriving at the target server is normal. It is determined that the freeze is caused by abnormal operation of the network camera; when any network node on the end-to-end path has a delay of >100ms, or a node packet loss rate of >10%, the network is judged to be abnormal, and the abnormal network node on the end-to-end path is located. The freeze is caused by an abnormal network node; When the fault is a black screen of video, find the associated I frame loss rate and frame loss duration. The I frame loss rate is further associated with the end-to-end path. The complete end-to-end path is queried, and the delay and packet loss rate of each network node are queried. If an abnormal network node is found, it is determined that the network abnormality causes the black screen, and the abnormal network node is located. If all network nodes on the end-to-end path are normal, it is determined whether the video frame arrival end is working normally, the video server is normal, and the fault is located as a network camera fault and a video server abnormality. The fault is located as a video server abnormality that causes the black screen of video. When the fault is video distorted screen, find the associated I-frame packet loss rate. If the I-frame packet loss rate is less than 5%, the network is considered normal and the fault is located as a video server fault. If the I-frame packet loss rate is greater than 5%, further query the network node delay and node packet loss rate on the end-to-end path. If all node delays are less than 50ms and the node packet loss rate is less than 5%, the network is considered normal and the fault is located as a video server fault. If there are network nodes with node delays greater than 100ms or node packet loss rates greater than 10%, the network is considered abnormal.

[0012] The present invention provides a method for locating faults in a video surveillance system combining active and passive functions, which has the following beneficial effects: The present invention starts from user perception, that is, video failures such as freezes, interruptions, screen distortion, and black screens that users intuitively feel. By collecting and analyzing real-time video frames, and triggering network path detection, network indicator calculation, and server indicator calculation, the present invention realizes accurate positioning of video failures based on correlation analysis of multiple indicators. It can locate network camera failures, network transmission node failures, and server failures, providing an accurate monitoring means for the rapid repair of video failures, and greatly reducing the cost of video failure inspection and repair. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] 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 description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0014] Figure 1 A flow chart of the method provided by the present invention; Figure 2 This is a schematic diagram of the association rules provided by the present invention. DETAILED DESCRIPTION

[0015] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0016] The following is a detailed description of the implementation method of the present invention in conjunction with the accompanying drawings. Only some embodiments are described, not all embodiments. For the purpose of clarity, representations and descriptions that are not related to the present invention are omitted in the drawings and descriptions.

[0017] In order to have a clearer understanding of the technical features, purposes and beneficial effects of the present invention, the technical solution of the present invention is now described in detail below. Obviously, the implementation cases described are part of the embodiments of the present invention, not all of the embodiments, and cannot be understood as limiting the scope of the implementation of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the protection scope of the present invention.

[0018] The real-time monitoring video stream is obtained through switch mirroring and other means. The real-time monitoring video stream is used as the only data source. After collection and analysis, video frame data (including key frames and non-key frames) and video frame statistical indicators (key frame loss rate, non-key frame loss rate, key frame packet loss rate, and loss duration) are obtained; in the process of collecting the real-time monitoring video stream, the session information of the video stream (including: source address, source port, destination address, and destination port) is obtained, and the obtained video frame data is subjected to algorithm analysis to obtain video fault indicators (including video disconnection, video freeze, video flower screen, and video black screen). When the analysis and detection shows that the video stream has a video stream transmission failure caused by disconnection and freeze, the network degradation of the network nodes in the end-to-end network between the network camera and the video server (delay and packet loss rate) is actively detected; the end-to-end network path information, network node degradation indicators, and video stream failures are obtained and used as the data source of the fault correlation analysis module for correlation analysis, so as to accurately locate the root cause of the video stream failure.

[0019] like Figure 1 As shown, the present invention provides a fault locating method for a video surveillance system combining active and passive functions, comprising the following steps: S1. acquiring a real-time monitoring video stream, and acquiring video frame data, video frame statistical indicators, and video stream session information from the real-time monitoring video stream; S2. analyzing the video frame data to acquire video fault indicators; S3. detecting the video server and the network path conditions to acquire the network degradation conditions of the network nodes; S4. performing correlation analysis based on the video fault indicators, the network path, and the network degradation conditions as data sources to locate the fault.

[0020] The video frame data in step S1 specifically includes: key frames, non-key frames; the video frame statistical indicators specifically include: key frame loss rate, non-key frame loss rate, key frame packet loss rate, loss duration; the video stream session information specifically includes: source address, source port, destination address, destination port.

[0021] Step S1 specifically includes the following sub-steps: S11. Acquire video stream IP packet data by real-time packet capture; S12. Perform data analysis on the captured IP packet data: protocol identification, data parsing, and indicator calculation.

[0022] Among them, the protocol identification specifically includes: identifying the transport layer type of the data packet, obtaining the video stream session information, extracting the characteristic fields of the video data packet and calculating the unique ID data identifier of each video stream; data analysis specifically includes: generating multiple types of video data packet characteristic codes according to the characteristic field combination of the extracted video data packet, and establishing the characteristic video data packet data analysis mode function according to the data packet characteristic code.

[0023] The calculation formulas in the indicator calculation include: Theoretical number of key frames = 300*frame rate / group length; Key frame loss rate = (theoretical key frame number - actual key frame number) / theoretical key frame number; Theoretical number of non-key frames = 300*frame rate-300*frame rate / group length; Non-key frame loss rate = (theoretical non-key frame number - actual non-key frame number) / theoretical non-key frame number; Key frame subpacket loss rate = number of key frame subpacket losses / total number of key frame subpackets, where the number of key frame subpacket losses = number of missing subpacket sequence numbers; Frame drop duration = time of receiving a new frame - time of frame drop start.

[0024] The calculation formula for generating the video data packet feature code is specifically: (low 5 bits of the first byte of the data packet & low 5 bits of the second byte of the data packet) | (high 3 bits of the first byte of the data packet & high 3 bits of the second byte of the data packet).

[0025] The established characteristic video data packet data parsing mode function is used to complete the data parsing of a specific video data packet. The reverse data parsing according to the data encapsulation rule of the characteristic video data packet specifically includes the following steps: A1. According to the characteristic field, determine whether the data packet is single-packet data or sub-packet data; A2. Cache the single-packet data and cache the sub-packet data after extracting the data body; A3. According to the characteristic field, determine the frame type, and accumulate the actual number of key frames and the actual number of non-key frames.

[0026] The video fault indicators in step S2 specifically include: video interruption, video freeze, video distorted screen, and video black screen.

[0027] The judgment steps for each video fault indicator in step S2 are: Video interruption: Record the appearance time of the first frame. When the video data cannot be decoded, start accumulating time until a new normal frame is obtained. When the accumulated time is greater than the standard key frame interval (= group length / theoretical frame rate) (common group length = 50 frames, theoretical frame rate 25 frames / second, so the calculated key frame interval is 2 seconds), it is judged as video interruption.

[0028] Video freeze: Record the appearance time of the first frame. When the time intervals between five consecutive decoded frames are less than the standard I-frame interval (= group length / theoretical frame rate), it is considered as video freeze. When the non-key frame loss rate is greater than the set threshold and the frame loss duration is less than the set threshold, it is also considered as video freeze.

[0029] Video distorted screen: Detects whether a specific area of ​​the decoded image is a pure color area, while the remaining area is a non-pure color area. When both conditions are met, it is judged as a video distorted screen. When the upper or lower half of an image is a pure color area and the other half is a non-pure color area, it is perceived as partial loss of image content, and it is judged as a video distorted screen.

[0030] Video black screen: Detects whether the decoded image is a solid color image. If it is a solid color image, it is perceived that all image content is lost and the video is judged to be black screen.

[0031] The specific steps of judging whether an image is a pure color area include: Z1. binarizing and inverting the grayscale image after grayscale transformation; Z2. counting the number and proportion of white pixels in a specific area: when the white pixel proportion is greater than 0.999, the area is judged to be a pure color area; when the white pixel proportion is less than 0.001, the area is judged to be a pure color area; when the white pixel proportion is greater than 0.001 and less than 0.999, the area is judged to be a non-pure color area.

[0032] Analyze the video stream data arriving at the target IP address of the video stream (corresponding to the video server). When decoding the video, record the abnormal decoding status as the video server status. The decoding success is 1 and the decoding failure is 0.

[0033] Get the session information of the video stream, including the source IP, source port, target IP, and target port of the video stream. Construct ICMP packets and TCP SYN packets. Both network detection packets are active detection packets. When the initial TTL value of the detection packet is 32, it starts to send active detection packets to the source IP and source port of the video stream, as well as the target IP and target port of the video stream. Each time the detection packet passes through a network node, the TTL value decreases by 1 (until it is 0). At the same time, the network node sends a feedback packet, which contains the network node's own IP and other information. After receiving the feedback packet, the network node is added to the source path table; similarly, the target path table is obtained by iteration. In order to obtain the information of each hop network node on the network path to the greatest extent, ensure that each network node can try its best to send out detection feedback packets, and send out ICMP packets and TCP SYN packets at the same time. When the two packets arrive at a certain network node, if the network node blocks the response feedback of one of the packets due to security settings and other reasons, then at least one response feedback packet can be sent. As long as at least one response feedback packet is received, the IP and other information of the node can be obtained, thereby constructing a path table. If multiple response feedback packets are received at the same time, the module merges the feedback data therein. The processing principle is: retain different data information and merge the same data information. Finally, the source path table and the target path table are obtained.

[0034] The network path detection initiates a maximum of 25 detections for each target IP address every 300 seconds to maximize the path discovery and timely update.

[0035] When a video streaming session is detected to have a disconnection or freeze, network degradation detection is performed. Based on the path information obtained from network path detection, 25 active detection ICMP packets are sent evenly to each IP address on the path within each detection cycle of 300 seconds. The feedback of each detection is analyzed. If there is an abnormal situation of delay >= 1000ms (delay = time to receive feedback packet - time to send detection packet) or packet loss rate = 100%, an active detection TCP SYN packet is sent to the IP to complete the remaining active detection of this probe cycle. The average value of the delay obtained by calculating the 25 feedback packets is the node delay, and the node packet loss rate is the number of lost packets / 25.

[0036] The association rules for association analysis in step S4 are as follows: The association rules are divided into user perception layer, video stream perception layer, and network perception layer. The user perception layer is the fault detection result generated based on the user's intuitive perception, the video stream perception layer is composed of the perception indicators obtained by passive analysis of the video stream, and the network perception layer is composed of the perception indicators obtained by active network detection, such as Figure 2 shown.

[0037] When the fault is video interruption, find the associated I frame loss rate and frame loss duration, and then query the end-to-end path information, node delay and node packet loss. If the delay of all nodes on the path on the network camera side and the server side is less than 50ms, and the node packet loss rate is less than 5%, it can be determined that the network is normal and the status of data frames arriving at the target server is normal. It is determined that the network camera is abnormal and caused the interruption. When any network node on the end-to-end path has a delay of >100ms, or a node packet loss rate of >10%, it is directly determined that the network is abnormal and the abnormal network node on the end-to-end path is located. The abnormality of a network node causes the interruption.

[0038] When the fault is video freeze, find the associated P frame loss rate and frame loss duration, and then query the end-to-end path information, node delay and node packet loss. If the delay of all nodes on the path on the network camera side and the server side is less than 50ms, and the node packet loss rate is less than 5%, the network is judged to be normal, and the status of data frames arriving at the target server is normal. It is determined that the freeze is caused by abnormal operation of the network camera; when any network node on the end-to-end path has a delay of >100ms, or a node packet loss rate of >10%, the network is judged to be abnormal, and the abnormal network node on the end-to-end path is located. The freeze is caused by an abnormal network node.

[0039] When the fault is a black screen of video, find the associated I frame loss rate and frame loss duration. The I frame loss rate is further associated with the end-to-end path. Query the complete end-to-end path, and query the delay and packet loss rate of each network node. If there is an abnormal network node, determine that the network abnormality causes the black screen, and locate the abnormal network node; if all network nodes on the end-to-end path are normal, determine whether the video frame arrival end is working normally, the video server is normal, and the fault is located as a network camera failure and a video server abnormality. The fault is located as the video server abnormality causing the black screen of video.

[0040] When the fault is video distorted screen, find the associated I-frame packet loss rate. If the I-frame packet loss rate is less than 5%, the network is considered normal and the fault is located as a video server fault. If the I-frame packet loss rate is greater than 5%, further query the network node delay and node packet loss rate on the end-to-end path. If all node delays are less than 50ms and the node packet loss rate is less than 5%, the network is considered normal and the fault is located as a video server fault. If there are network nodes with node delays greater than 100ms or node packet loss rates greater than 10%, the network is considered abnormal.

[0041] The present invention starts from user perception, that is, video failures such as freezes, interruptions, screen distortion, and black screens that users intuitively feel. By collecting and analyzing real-time video frames, and triggering network path detection, network indicator calculation, and server indicator calculation, the present invention realizes accurate positioning of video failures based on correlation analysis of multiple indicators. It can locate network camera failures, network transmission node failures, and server failures, providing an accurate monitoring means for the rapid repair of video failures, and greatly reducing the cost of video failure inspection and repair.

[0042] The above is only a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art shall not deviate from the spirit and scope of the present invention, and shall be within the scope of protection of the claims attached to the present invention.

Claims

1. A method for locating faults in a video surveillance system combining active and passive functions, characterized in that: The following steps are involved: S1. Obtain real-time monitoring video stream, and obtain video frame data, video frame statistics and video stream session information from the real-time monitoring video stream; S2. Analyze the video frame data to obtain video fault indicators; S3. Detect the video server and network path to obtain the network degradation of the network nodes; S4. Perform correlation analysis based on video fault indicators, network paths, and network degradation as data sources to locate the fault.

2. The fault location method of the active and passive combined video surveillance system according to claim 1 is characterized in that: The video frame data in step S1 specifically includes: key frames and non-key frames; The video frame statistical indicators specifically include: key frame loss rate, non-key frame loss rate, key frame packet loss rate, and loss duration; The video stream session information specifically includes: source address, source port, destination address, and destination port.

3. The fault location method of the active and passive combined video surveillance system according to claim 2 is characterized in that: The S1 step specifically includes the following sub-steps: S11. Obtain video stream IP packet data by real-time packet capture; S12. Perform data analysis on the captured IP packet data: protocol identification, data analysis, and indicator calculation; The protocol identification specifically includes: identifying the transport layer type of the data packet, obtaining the video stream session information, extracting the characteristic fields of the video data packet and calculating the unique ID data identifier of each video stream; The data analysis specifically includes: generating multiple types of video data packet feature codes according to the feature field combination of the extracted video data packet, and establishing the feature video data packet data analysis mode function according to the data packet feature code; The calculation formula in the indicator calculation specifically includes: Theoretical number of key frames = 300*frame rate / group length; Key frame loss rate = (theoretical key frame number - actual key frame number) / theoretical key frame number; Theoretical number of non-key frames = 300*frame rate-300*frame rate / group length; Non-key frame loss rate = (theoretical non-key frame number - actual non-key frame number) / theoretical non-key frame number; Key frame subpacket loss rate = number of key frame subpacket losses / total number of key frame subpackets, where the number of key frame subpacket losses = number of missing subpacket sequence numbers; Frame drop duration = time of receiving a new frame - time of frame drop start.

4. The fault location method of the active and passive combined video surveillance system according to claim 3 is characterized in that: The calculation formula for generating the video data packet feature code is specifically: (low 5 bits of the first byte of the data packet & low 5 bits of the second byte of the data packet) | (high 3 bits of the first byte of the data packet & high 3 bits of the second byte of the data packet).

5. The fault location method of the active and passive combined video surveillance system according to claim 3 is characterized in that: The established characteristic video data packet data parsing mode function is used to complete the data parsing of a specific video data packet, and reverse data parsing according to the data encapsulation rule of the characteristic video data packet specifically includes the following steps: A1. According to the characteristic field, determine whether the data packet is single packet data or sub-packet data; A2. Cache single-packet data and cache sub-packet data after extracting the data body; A3. Determine the frame type based on the feature field, and accumulate the actual number of key frames and the actual number of non-key frames.

6. The fault location method of the active and passive combined video surveillance system according to claim 1, characterized in that: The video fault indicators in step S2 specifically include: video interruption, video freeze, video distorted screen, and video black screen.

7. The fault location method of the active and passive combined video surveillance system according to claim 6 is characterized in that: The steps for judging each video fault indicator in step S2 are: Video interruption: Record the appearance time of the first frame. When the video data cannot be decoded, start accumulating time until a new normal frame is obtained. When the accumulated time is greater than the standard key frame interval, it is judged as a video interruption. Video freeze: Record the appearance time of the first frame. When the time intervals before and after the appearance of 5 consecutive decoded frames are less than the standard I-frame interval, it is judged as video freeze. When the non-key frame loss rate is greater than the set threshold and the frame loss duration is less than the set threshold, it is also judged as video freeze. Video distorted screen: Detects whether a specific area of ​​the decoded image is a pure color area, while the remaining area is a non-pure color area. If both conditions are met, it is judged as a video distorted screen. When the upper or lower half of an image is a pure color area and the other half is a non-pure color area, it is perceived as partial loss of image content, and it is judged as a video distorted screen. Video black screen: Detects whether the decoded image is a solid color image. If it is a solid color image, it is perceived that all image content is lost and the video is judged to be black screen.

8. The fault location method of the active and passive combined video surveillance system according to claim 7 is characterized in that: The specific steps of determining whether an image is a pure color area include: Z1. binarizing and inverting the grayscale image after grayscale transformation; Z2. counting the number and proportion of white pixels in a specific area: when the white pixel ratio is greater than 0.999, the area is determined to be a pure color area; when the white pixel ratio is less than 0.001, the area is determined to be a pure color area; when the white pixel ratio is greater than 0.001 and less than 0.999, the area is determined to be a non-pure color area.

9. The fault location method of the active and passive combined video surveillance system according to claim 1, characterized in that: The association rules for association analysis in step S4 are specifically: When the fault is video interruption, find the associated I frame loss rate and frame loss duration, and then query the end-to-end path information, node delay and node packet loss. If the delay of all nodes on the path on the network camera side and the server side is less than 50ms, and the node packet loss rate is less than 5%, it can be determined that the network is normal and the state of data frames arriving at the target server is normal. It is determined that the network camera is abnormal and caused the interruption; when any network node on the end-to-end path has a delay of >100ms, or a node packet loss rate of >10%, it is directly determined that the network is abnormal and the abnormal network node on the end-to-end path is located. The abnormality of a certain network node causes the interruption; When the fault is video freeze, find the associated P frame loss rate and frame loss duration, and then query the end-to-end path information, node delay and node packet loss. If the delay of all nodes on the path of the network camera side and the server side is less than 50ms, and the node packet loss rate is less than 5%, the network is judged to be normal, and the state of data frames arriving at the target server is normal. It is determined that the freeze is caused by abnormal operation of the network camera; when any network node on the end-to-end path has a delay of >100ms, or a node packet loss rate of >10%, the network is judged to be abnormal, and the abnormal network node on the end-to-end path is located. The freeze is caused by an abnormal network node; When the fault is a black screen of video, find the associated I frame loss rate and frame loss duration. The I frame loss rate is further associated with the end-to-end path. The complete end-to-end path is queried, and the delay and packet loss rate of each network node are queried. If an abnormal network node is found, it is determined that the network abnormality causes the black screen, and the abnormal network node is located. If all network nodes on the end-to-end path are normal, it is determined whether the video frame arrival end is working normally, the video server is normal, and the fault is located as a network camera fault and a video server abnormality. The fault is located as a video server abnormality that causes the black screen of video. When the fault is video distorted screen, find the associated I-frame packet loss rate. If the I-frame packet loss rate is less than 5%, the network is considered normal and the fault is located as a video server fault. If the I-frame packet loss rate is greater than 5%, further query the network node delay and node packet loss rate on the end-to-end path. If all node delays are less than 50ms and the node packet loss rate is less than 5%, the network is considered normal and the fault is located as a video server fault. If there are network nodes with node delays greater than 100ms or node packet loss rates greater than 10%, the network is considered abnormal.

Images