Vehicle sampling abnormity early warning method and system based on video monitoring

Abstract

The invention relates to the technical field of vehicle early warning, and discloses a vehicle sampling abnormity early warning method and system based on video monitoring, and the method comprises the steps: obtaining original video monitoring data corresponding to a vehicle to be subjected to abnormal early warning, and determining a to-be-analyzed target monitoring video; dividing a to-be-analyzed target monitoring video into a plurality of sub-monitoring videos, and calculating a sub-monitoring video analysis coefficient of each sub-monitoring video; obtaining a safe target monitoring video of the to-be-analyzed target monitoring video, and determining a corresponding sub-safe video analysis coefficient; performing classification processing on the sub-monitoring video analysis coefficients according to the sub-safety video analysis coefficients, and calculating a vehicle abnormity early warning analysis coefficient; whether the vehicle to be subjected to abnormal early warning is abnormal or not is judged based on the vehicle abnormal early warning analysis coefficient, and when the vehicle to be subjected to abnormal early warning is abnormal, abnormal early warning is given out, so that accurate recognition and rapid early warning of vehicle behaviors are achieved, the safety and reliability of the vehicle are guaranteed, and the method has important significance on the vehicle safety management level.

Description

Technical Field

[0001] The present invention relates to the field of vehicle early warning technology, and in particular to a vehicle sampling abnormality early warning method and system based on video monitoring. Background Art

[0002] Video surveillance system, or VSCS (video surveillance & control system), is a monitoring system composed of cameras, transmission systems and display devices, which is used to achieve security purposes. Video surveillance system is mainly composed of front-end equipment, transmission system, storage device, control part and display part. Front-end equipment includes various types of cameras, such as gun-type, hemispherical, high-speed ball cameras, etc. As an important part of modern security technology, the development and application of video surveillance system is of great significance to improving social security level and optimizing resource allocation. In the future, with the continuous innovation of technology, video surveillance system will play a more critical role in smart city construction and public security protection.

[0003] Traditional traffic monitoring systems mostly rely on manual monitoring or simple line-of-sight detection equipment, which is not only inefficient, but also difficult to achieve comprehensive monitoring and timely warning of complex scenes. In recent years, with the development of deep learning technology and high-performance computing platforms, vehicle warning methods based on video surveillance have gradually become a research hotspot. Summary of the invention

[0004] The embodiment of the present invention provides a vehicle sampling abnormality warning method and system based on video surveillance, which realizes accurate identification and rapid warning of vehicle behavior by optimizing algorithms and enhancing data processing capabilities, thereby ensuring the safety and reliability of the vehicle, which is of great significance to the level of vehicle safety management.

[0005] In order to achieve the above object, the present invention provides a vehicle sampling abnormality warning method based on video monitoring, comprising: Determine a vehicle to be warned of abnormality, obtain original video surveillance data corresponding to the vehicle to be warned of abnormality, and process the original video surveillance data to determine a target surveillance video to be analyzed; Dividing the target surveillance video to be analyzed into a plurality of sub-surveillance videos, analyzing the sub-surveillance videos, and calculating a sub-surveillance video analysis coefficient of each sub-surveillance video based on the analysis result; Acquire a security target monitoring video that is similar to the target monitoring video to be analyzed, and analyze the security target monitoring video to determine a corresponding sub-security video analysis coefficient; Classifying the corresponding sub-monitoring video analysis coefficients according to the sub-safety video analysis coefficients, and calculating the vehicle abnormality warning analysis coefficient of the vehicle to be warned of abnormality based on the classification processing result; Based on the vehicle abnormality warning analysis coefficient, it is judged whether the vehicle to be warned of abnormality has abnormality, and when the vehicle to be warned of abnormality has abnormality, an abnormality warning is issued.

[0006] Furthermore, when determining a vehicle to be warned of abnormality, acquiring original video surveillance data corresponding to the vehicle to be warned of abnormality, and processing the original video surveillance data to determine the target surveillance video to be analyzed, the method includes: Preprocessing the raw video surveillance data, wherein the preprocessing includes denoising, contrast enhancement, and brightness adjustment; Using an image segmentation algorithm to segment the preprocessed raw video surveillance data into multiple video regions, wherein the image segmentation algorithm includes any one or more combinations of a threshold segmentation method, an edge detection-based segmentation method, a region-based segmentation method, or a deep learning-based segmentation method; Determining a target video area to be extracted from multiple video areas according to a preset extraction rule, wherein the preset extraction rule includes shape, size, position, color feature, texture feature or motion feature; Extracting a target video segment from the target video area, and performing post-processing on the extracted target video segment, wherein the post-processing includes smoothing edges and adjusting color balance; The processed target video clip is output as the target surveillance video to be analyzed.

[0007] Furthermore, when the target surveillance video to be analyzed is divided into a plurality of sub-surveillance videos, and the sub-surveillance videos are analyzed, and a sub-surveillance video analysis coefficient of each sub-surveillance video is calculated based on the analysis result, it includes: Extracting a video frame image from the sub-surveillance video and extracting corresponding video pixels; Clustering all video pixel points to determine video pixel point clusters, and determining cluster centers of the video pixel point clusters; Determine the video distance from each video pixel to the cluster center, and construct a video distance set; Calculating the mean and standard deviation of the video distance set; For all video distances greater than the mean and greater than the standard deviation, generate a video distance large flag; For all video distances that are equal to the mean and the standard deviation, generate video distance iso-identifiers; Generate video distance difference identifiers for all remaining video distances; Count the number of large identifiers of the video distance large identifier, count the number of equal identifiers of the video distance equal identifiers, and count the number of difference identifiers of the video distance difference identifiers; The sub-surveillance video analysis coefficient of each sub-surveillance video is calculated based on the number of large identifiers, the number of equal identifiers and the number of difference identifiers.

[0008] Further, when the corresponding sub-monitoring video analysis coefficients are classified according to the sub-safety video analysis coefficients, and the vehicle abnormality warning analysis coefficients of the vehicle to be abnormally warned are calculated based on the classification processing results, it includes: Compare all sub-surveillance video analysis coefficients with the corresponding sub-safety video analysis coefficients one by one, divide all sub-surveillance video analysis coefficients that are smaller than the sub-safety video analysis coefficients into a lower difference coefficient set, and sort the lower difference coefficient set by numerical value; Dividing all sub-surveillance video coefficients equal to the sub-security video analysis coefficients into a set of equal difference coefficients; All sub-surveillance video analysis coefficients that are smaller than the sub-security video analysis coefficient are divided into an upper difference coefficient set, and the lower difference coefficient set is sorted by numerical value; Calculating a vehicle abnormality warning analysis coefficient of the vehicle to be warned according to the lower difference coefficient set, the equal difference coefficient set and the upper difference coefficient set; ; ; ; Among them, w is the vehicle abnormal warning analysis coefficient of the vehicle to be warned, a is the adjustment coefficient of the vehicle abnormal warning analysis coefficient, e(g, k) is the comparison function between the upper difference coefficient set and the lower difference coefficient set, Δd is the weight corresponding to the sub-monitoring video analysis coefficient, and its value is [0.85, 1.25], f is the number of sub-monitoring video analysis coefficients in the upper difference coefficient set, β is the loss factor for calculating the vehicle abnormal warning analysis coefficient, Δt i is the corresponding difference between the sub-monitoring video analysis coefficient in the upper difference coefficient set and the sub-monitoring video analysis coefficient in the lower difference coefficient set, max(Δt i ) is the difference Δt from all i The maximum value obtained in 均 For all differences Δt i The average value, t i is the i-th sub-surveillance video analysis coefficient in the above difference coefficient set, r i is the sub-surveillance video analysis coefficient corresponding to the ith sub-surveillance video analysis coefficient in the lower difference coefficient set; The adjustment coefficient a of the vehicle abnormality warning analysis coefficient is determined according to the following method: Counting the equal number of sub-monitoring video analysis coefficients in the equal difference coefficient set; Counting the sum Q of the sub-monitoring video analysis coefficients in the upper abnormal coefficient set and the lower difference coefficient set; Presetting a first preset adjustment coefficient, a second preset adjustment coefficient, and a third preset adjustment coefficient; When the equal number is less than 0.85Q, the first preset adjustment coefficient is used as the adjustment coefficient a of the vehicle abnormality warning analysis coefficient; When the equal quantity is greater than or equal to 0.85Q and less than 1.25Q, the second adjustment coefficient is used as the adjustment coefficient a of the vehicle abnormality warning analysis coefficient; When the equal number is greater than 1.25Q, the third adjustment coefficient is used as the adjustment coefficient a of the vehicle abnormality warning analysis coefficient.

[0009] Further, when judging whether the vehicle to be warned of abnormality has abnormality based on the vehicle abnormality warning analysis coefficient, it includes: Obtaining a preset abnormal warning analysis coefficient for a preset vehicle, and judging whether the vehicle to be warned of abnormality has an abnormality according to the relationship between the abnormal warning analysis coefficient and the preset abnormal warning analysis coefficient; When the abnormal warning analysis coefficient is less than the preset abnormal warning analysis coefficient, it is determined that the vehicle to be warned of abnormality has an abnormality; When the abnormal warning analysis coefficient is greater than or equal to the preset abnormal warning analysis coefficient, it is determined that there is no abnormality in the vehicle to be warned.

[0010] In order to achieve the above object, the present invention also provides a vehicle sampling abnormality warning system based on video monitoring, comprising: A video processing module is used to determine a vehicle to be warned of abnormality, obtain original video surveillance data corresponding to the vehicle to be warned of abnormality, and process the original video surveillance data to determine a target surveillance video to be analyzed; A first calculation module, used for dividing the target surveillance video to be analyzed into a plurality of sub-surveillance videos, analyzing the sub-surveillance videos, and calculating a sub-surveillance video analysis coefficient of each sub-surveillance video based on the analysis result; A video analysis module, used to obtain a security target monitoring video corresponding to the target monitoring video to be analyzed, and analyze the security target monitoring video to determine a corresponding sub-security video analysis coefficient; A second calculation module is used to classify the corresponding sub-monitoring video analysis coefficient according to the sub-safety video analysis coefficient, and calculate the vehicle abnormality warning analysis coefficient of the vehicle to be abnormally warned based on the classification processing result; The abnormality warning module is used to determine whether the vehicle to be warned of abnormality has abnormality based on the vehicle abnormality warning analysis coefficient, and to issue an abnormality warning when the vehicle to be warned of abnormality has abnormality.

[0011] Furthermore, the video processing module is used for: The video processing module is used to pre-process the raw video surveillance data, wherein the pre-processing includes denoising, contrast enhancement, and brightness adjustment; The video processing module is used to segment the pre-processed raw video surveillance data into multiple video regions using an image segmentation algorithm, wherein the image segmentation algorithm includes any one or more combinations of a threshold segmentation method, an edge detection-based segmentation method, a region-based segmentation method, or a deep learning-based segmentation method; The video processing module is used to determine a target video area to be extracted from multiple video areas according to a preset extraction rule, wherein the preset extraction rule includes shape, size, position, color feature, texture feature or motion feature; The video processing module is used to extract a target video segment from the target video area and perform post-processing on the extracted target video segment, wherein the post-processing includes smoothing edges and adjusting color balance; The video processing module is used to output the post-processed target video segment as the target monitoring video to be analyzed.

[0012] Furthermore, the first calculation module is used for: The first calculation module is used to extract a video frame image from the sub-surveillance video and extract corresponding video pixels; The first calculation module is used to cluster all video pixel points, determine video pixel point clusters, and determine the cluster centers of the video pixel point clusters; The first calculation module is used to determine the video distance from each video pixel to the cluster center and construct a video distance set; The first calculation module is used to calculate the mean and standard deviation of the video distance set; The first calculation module is used to generate a large video distance mark for all video distances greater than the mean and greater than the standard deviation; The first calculation module is used to generate video distance and other identifiers for all video distances that are equal to the mean and the standard deviation; The first calculation module is used to generate a video distance difference identifier for all remaining video distances; The first calculation module is used to count the number of large identifiers of the video distance large identifier, count the number of equal identifiers of the video distance equal identifiers, and count the number of difference identifiers of the video distance difference identifiers; The first calculation module is used to calculate the sub-surveillance video analysis coefficient of each sub-surveillance video based on the number of large identifiers, the number of equal identifiers and the number of difference identifiers.

[0013] Furthermore, the second calculation module is used for: The second calculation module is used to compare all sub-surveillance video analysis coefficients with corresponding sub-safety video analysis coefficients one by one, divide all sub-surveillance video analysis coefficients that are smaller than the sub-safety video analysis coefficients into a lower difference coefficient set, and sort the lower difference coefficient set by numerical value; The second calculation module is used to divide all sub-surveillance video coefficients equal to the sub-security video analysis coefficients into a set of equal difference coefficients; The second calculation module is used to divide all sub-surveillance video analysis coefficients that are smaller than the sub-security video analysis coefficient into an upper difference coefficient set, and sort the lower difference coefficient set by numerical value; The second calculation module is used to calculate the vehicle abnormality warning analysis coefficient of the vehicle to be warned according to the lower difference coefficient set, the equal difference coefficient set and the upper difference coefficient set; ; ; ; Among them, w is the vehicle abnormal warning analysis coefficient of the vehicle to be warned, a is the adjustment coefficient of the vehicle abnormal warning analysis coefficient, e(g, k) is the comparison function between the upper difference coefficient set and the lower difference coefficient set, Δd is the weight corresponding to the sub-monitoring video analysis coefficient, and its value is [0.85, 1.25], f is the number of sub-monitoring video analysis coefficients in the upper difference coefficient set, β is the loss factor for calculating the vehicle abnormal warning analysis coefficient, Δt i is the corresponding difference between the sub-monitoring video analysis coefficient in the upper difference coefficient set and the sub-monitoring video analysis coefficient in the lower difference coefficient set, max(Δt i ) is the difference Δt from all i The maximum value obtained in 均 For all differences Δt i The average value, t i is the i-th sub-surveillance video analysis coefficient in the above difference coefficient set, r iis the sub-surveillance video analysis coefficient corresponding to the ith sub-surveillance video analysis coefficient in the lower difference coefficient set; The second calculation module is used to determine the adjustment coefficient a of the vehicle abnormality warning analysis coefficient according to the following method: The second calculation module is used to count the equal number of sub-monitoring video analysis coefficients in the equal difference coefficient set; The second calculation module is used to count the sum Q of the sub-monitoring video analysis coefficients in the upper abnormal coefficient set and the lower difference coefficient set; The second calculation module is used to pre-set a first preset adjustment coefficient, a second preset adjustment coefficient and a third preset adjustment coefficient; The second calculation module is used for using the first preset adjustment coefficient as the adjustment coefficient a of the vehicle abnormality warning analysis coefficient when the equal quantity is less than 0.85Q; The second calculation module is used for using the second adjustment coefficient as the adjustment coefficient a of the vehicle abnormality warning analysis coefficient when the equal quantity is greater than or equal to 0.85Q and less than 1.25Q; The second calculation module is used for using the third adjustment coefficient as the adjustment coefficient a of the vehicle abnormality warning analysis coefficient when the equal quantity is greater than 1.25Q.

[0014] Furthermore, the abnormal warning module is used to: The abnormal warning module is used to obtain a preset abnormal warning analysis coefficient of a preset vehicle, and judge whether the vehicle to be warned has an abnormality according to the relationship between the abnormal warning analysis coefficient and the preset abnormal warning analysis coefficient; The abnormal warning module is used to determine that the vehicle to be warned has an abnormality when the abnormal warning analysis coefficient is less than the preset abnormal warning analysis coefficient; The abnormal warning module is used to determine that there is no abnormality in the vehicle to be warned when the abnormal warning analysis coefficient is greater than or equal to the preset abnormal warning analysis coefficient.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention discloses a vehicle sampling abnormality warning method and system based on video monitoring, which obtains original video monitoring data corresponding to a vehicle to be warned of abnormality, and determines a target monitoring video to be analyzed; divides the target monitoring video to be analyzed into multiple sub-monitoring videos, and calculates a sub-monitoring video analysis coefficient of each sub-monitoring video; obtains a safety target monitoring video corresponding to the target monitoring video to be analyzed, and determines a corresponding sub-safety video analysis coefficient; classifies the sub-monitoring video analysis coefficients according to the sub-safety video analysis coefficients, and calculates a vehicle abnormality warning analysis coefficient; determines whether the vehicle to be warned of abnormality has abnormality based on the vehicle abnormality warning analysis coefficient, and issues an abnormality warning when the vehicle to be warned of abnormality has abnormality, thereby realizing accurate identification and rapid warning of vehicle behavior, ensuring the safety and reliability of the vehicle, and having important significance for the level of vehicle safety management. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings: Figure 1 A schematic diagram of the process of a vehicle sampling abnormality warning method based on video monitoring in an embodiment of the present invention is shown; Figure 2 A schematic structural diagram of a vehicle sampling abnormality warning system based on video monitoring in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0017] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0018] In the description of the present application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0019] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0020] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0021] The following is a description of preferred embodiments of the present invention with reference to the accompanying drawings.

[0022] like Figure 1 As shown, an embodiment of the present invention discloses a vehicle sampling abnormality warning method based on video monitoring, comprising: S110: Determine a vehicle to be warned of abnormality, obtain original video surveillance data corresponding to the vehicle to be warned of abnormality, and process the original video surveillance data to determine a target surveillance video to be analyzed; In some embodiments of the present application, when determining a vehicle to be warned of abnormality, acquiring original video surveillance data corresponding to the vehicle to be warned of abnormality, and processing the original video surveillance data to determine the target surveillance video to be analyzed, the method includes: Preprocessing the raw video surveillance data, wherein the preprocessing includes denoising, contrast enhancement, and brightness adjustment; Using an image segmentation algorithm to segment the preprocessed raw video surveillance data into multiple video regions, wherein the image segmentation algorithm includes any one or more combinations of a threshold segmentation method, an edge detection-based segmentation method, a region-based segmentation method, or a deep learning-based segmentation method; Determining a target video area to be extracted from multiple video areas according to a preset extraction rule, wherein the preset extraction rule includes shape, size, position, color feature, texture feature or motion feature; Extracting a target video segment from the target video area, and performing post-processing on the extracted target video segment, wherein the post-processing includes smoothing edges and adjusting color balance; The processed target video clip is output as the target surveillance video to be analyzed.

[0023] The beneficial effect of the above technical solution is that the present invention obtains the target monitoring video to be analyzed by performing a series of processing on the original video monitoring data, thereby removing invalid information, improving vehicle warning efficiency, and laying the foundation for vehicle warning.

[0024] S120: Divide the target surveillance video to be analyzed into a plurality of sub-surveillance videos, analyze the sub-surveillance videos, and calculate a sub-surveillance video analysis coefficient of each sub-surveillance video based on the analysis result; In some embodiments of the present application, when the target surveillance video to be analyzed is divided into a plurality of sub-surveillance videos, and the sub-surveillance videos are analyzed, and a sub-surveillance video analysis coefficient of each sub-surveillance video is calculated based on the analysis result, the method includes: Extracting a video frame image from the sub-surveillance video and extracting corresponding video pixels; Clustering all video pixel points to determine video pixel point clusters, and determining cluster centers of the video pixel point clusters; Determine the video distance from each video pixel to the cluster center, and construct a video distance set; Calculating the mean and standard deviation of the video distance set; For all video distances greater than the mean and greater than the standard deviation, generate a video distance large flag; For all video distances that are equal to the mean and the standard deviation, generate video distance iso-identifiers; Generate video distance difference identifiers for all remaining video distances; Count the number of large identifiers of the video distance large identifier, count the number of equal identifiers of the video distance equal identifiers, and count the number of difference identifiers of the video distance difference identifiers; The sub-surveillance video analysis coefficient of each sub-surveillance video is calculated based on the number of large identifiers, the number of equal identifiers and the number of difference identifiers.

[0025] In this embodiment, each frame corresponds to a corresponding image, that is, a video frame image.

[0026] In this embodiment, the method for determining the cluster center is complicated and mature, and will not be introduced in detail here.

[0027] In this embodiment, when calculating the sub-surveillance video analysis coefficient of each sub-surveillance video based on the number of large markers, the number of equal markers, and the number of difference markers, the sub-surveillance video analysis coefficient is calculated according to the following formula: ; Among them, p is the sub-surveillance video analysis coefficient, c1 is the number of equal identifications, c2 is the number of large identifications, and c3 is the number of difference identifications.

[0028] The beneficial effect of the above technical solution is that the present invention can calculate the sub-surveillance video analysis coefficient of each sub-surveillance video based on the number of large identifications, the number of equal identifications and the number of difference identifications, thereby ensuring the calculation precision and accuracy of the sub-surveillance video analysis coefficient and avoiding calculation errors caused by manual participation.

[0029] S130: Acquire a security target monitoring video that is similar to the target monitoring video to be analyzed, and analyze the security target monitoring video to determine a corresponding sub-security video analysis coefficient; In this embodiment, the safety target monitoring video refers to the monitoring video collected when there is no abnormality in the vehicle.

[0030] In this embodiment, the sub-security video analysis coefficient of the security target monitoring video is calculated based on the above method. In order to save space, it will not be repeated here.

[0031] S140: classifying the corresponding sub-surveillance video analysis coefficients according to the sub-safety video analysis coefficients, and calculating the vehicle abnormality warning analysis coefficient of the vehicle to be abnormally warned based on the classification processing result; In some embodiments of the present application, when the corresponding sub-surveillance video analysis coefficient is classified according to the sub-safety video analysis coefficient, and the vehicle abnormality warning analysis coefficient of the vehicle to be abnormally warned is calculated based on the classification processing result, it includes: Compare all sub-surveillance video analysis coefficients with the corresponding sub-safety video analysis coefficients one by one, divide all sub-surveillance video analysis coefficients that are smaller than the sub-safety video analysis coefficients into a lower difference coefficient set, and sort the lower difference coefficient set by numerical value; Dividing all sub-surveillance video coefficients equal to the sub-security video analysis coefficients into a set of equal difference coefficients; All sub-surveillance video analysis coefficients that are smaller than the sub-security video analysis coefficient are divided into an upper difference coefficient set, and the lower difference coefficient set is sorted by numerical value; Calculating a vehicle abnormality warning analysis coefficient of the vehicle to be warned according to the lower difference coefficient set, the equal difference coefficient set and the upper difference coefficient set; ; ; ; Among them, w is the vehicle abnormal warning analysis coefficient of the vehicle to be warned, a is the adjustment coefficient of the vehicle abnormal warning analysis coefficient, e(g, k) is the comparison function between the upper difference coefficient set and the lower difference coefficient set, Δd is the weight corresponding to the sub-monitoring video analysis coefficient, and its value is [0.85, 1.25], f is the number of sub-monitoring video analysis coefficients in the upper difference coefficient set, β is the loss factor for calculating the vehicle abnormal warning analysis coefficient, Δt i is the corresponding difference between the sub-monitoring video analysis coefficient in the upper difference coefficient set and the sub-monitoring video analysis coefficient in the lower difference coefficient set, max(Δt i) is the difference Δt from all i The maximum value obtained in 均 For all differences Δt i The average value, t i is the i-th sub-surveillance video analysis coefficient in the above difference coefficient set, r i is the sub-surveillance video analysis coefficient corresponding to the ith sub-surveillance video analysis coefficient in the lower difference coefficient set; The adjustment coefficient a of the vehicle abnormality warning analysis coefficient is determined according to the following method: Counting the equal number of sub-monitoring video analysis coefficients in the equal difference coefficient set; Counting the sum Q of the sub-monitoring video analysis coefficients in the upper abnormal coefficient set and the lower difference coefficient set; Presetting a first preset adjustment coefficient, a second preset adjustment coefficient, and a third preset adjustment coefficient; When the equal number is less than 0.85Q, the first preset adjustment coefficient is used as the adjustment coefficient a of the vehicle abnormality warning analysis coefficient; When the equal quantity is greater than or equal to 0.85Q and less than 1.25Q, the second adjustment coefficient is used as the adjustment coefficient a of the vehicle abnormality warning analysis coefficient; When the equal number is greater than 1.25Q, the third adjustment coefficient is used as the adjustment coefficient a of the vehicle abnormality warning analysis coefficient.

[0032] In this embodiment, the first preset adjustment coefficient is 0.85, the second preset adjustment coefficient is 1, and the third preset adjustment coefficient is 1.25, which can be adjusted according to actual conditions.

[0033] The beneficial effect of the above technical solution is: the present invention calculates the vehicle abnormal warning analysis coefficient of the vehicle to be warned according to the lower difference coefficient set, the equal difference coefficient set and the upper difference coefficient set, thereby ensuring the calculation accuracy of the vehicle abnormal warning analysis coefficient, providing reliable data support and basis for vehicle warning analysis, and avoiding misjudgment. At the same time, the present application also adjusts the vehicle abnormal warning analysis coefficient according to the preset adjustment coefficient, realizes the dynamic adjustment of the vehicle abnormal warning analysis coefficient, and makes the calculation of the vehicle abnormal warning analysis coefficient more comprehensive.

[0034] S150: judging whether the vehicle to be warned of abnormality has abnormality based on the vehicle abnormality warning analysis coefficient, and issuing an abnormality warning when the vehicle to be warned of abnormality has abnormality.

[0035] In some embodiments of the present application, when judging whether the vehicle to be warned of abnormality has an abnormality based on the vehicle abnormality warning analysis coefficient, it includes: Obtaining a preset abnormal warning analysis coefficient for a preset vehicle, and judging whether the vehicle to be warned of abnormality has an abnormality according to the relationship between the abnormal warning analysis coefficient and the preset abnormal warning analysis coefficient; When the abnormal warning analysis coefficient is less than the preset abnormal warning analysis coefficient, it is determined that the vehicle to be warned of abnormality has an abnormality; When the abnormal warning analysis coefficient is greater than or equal to the preset abnormal warning analysis coefficient, it is determined that there is no abnormality in the vehicle to be warned.

[0036] The beneficial effect of the above technical solution is: the present invention determines whether there is an abnormality in the vehicle to be warned based on the relationship between the abnormal warning analysis coefficient and the preset abnormal warning analysis coefficient, thereby realizing accurate identification and rapid warning of vehicle behavior, ensuring the safety and reliability of the vehicle, and is of great significance to the level of vehicle safety management.

[0037] In order to further explain the technical idea of ​​the present invention, the technical solution of the present invention is now described in combination with specific application scenarios.

[0038] Correspondingly, such as Figure 2 As shown, the present application also provides a vehicle sampling abnormality warning system based on video monitoring, comprising: A video processing module is used to determine a vehicle to be warned of abnormality, obtain original video surveillance data corresponding to the vehicle to be warned of abnormality, and process the original video surveillance data to determine a target surveillance video to be analyzed; A first calculation module, used for dividing the target surveillance video to be analyzed into a plurality of sub-surveillance videos, analyzing the sub-surveillance videos, and calculating a sub-surveillance video analysis coefficient of each sub-surveillance video based on the analysis result; A video analysis module, used to obtain a security target monitoring video corresponding to the target monitoring video to be analyzed, and analyze the security target monitoring video to determine a corresponding sub-security video analysis coefficient; A second calculation module is used to classify the corresponding sub-monitoring video analysis coefficient according to the sub-safety video analysis coefficient, and calculate the vehicle abnormality warning analysis coefficient of the vehicle to be abnormally warned based on the classification processing result; The abnormality warning module is used to determine whether the vehicle to be warned of abnormality has abnormality based on the vehicle abnormality warning analysis coefficient, and to issue an abnormality warning when the vehicle to be warned of abnormality has abnormality.

[0039] In some embodiments of the present application, the video processing module is used to: The video processing module is used to pre-process the raw video surveillance data, wherein the pre-processing includes denoising, contrast enhancement, and brightness adjustment; The video processing module is used to segment the pre-processed raw video surveillance data into multiple video regions using an image segmentation algorithm, wherein the image segmentation algorithm includes any one or more combinations of a threshold segmentation method, an edge detection-based segmentation method, a region-based segmentation method, or a deep learning-based segmentation method; The video processing module is used to determine a target video area to be extracted from multiple video areas according to a preset extraction rule, wherein the preset extraction rule includes shape, size, position, color feature, texture feature or motion feature; The video processing module is used to extract a target video segment from the target video area and perform post-processing on the extracted target video segment, wherein the post-processing includes smoothing edges and adjusting color balance; The video processing module is used to output the post-processed target video segment as the target monitoring video to be analyzed.

[0040] In some embodiments of the present application, the first computing module is used to: The first calculation module is used to extract a video frame image from the sub-surveillance video and extract corresponding video pixels; The first calculation module is used to cluster all video pixel points, determine video pixel point clusters, and determine the cluster centers of the video pixel point clusters; The first calculation module is used to determine the video distance from each video pixel to the cluster center and construct a video distance set; The first calculation module is used to calculate the mean and standard deviation of the video distance set; The first calculation module is used to generate a large video distance mark for all video distances greater than the mean and greater than the standard deviation; The first calculation module is used to generate video distance and other identifiers for all video distances that are equal to the mean and the standard deviation; The first calculation module is used to generate a video distance difference identifier for all remaining video distances; The first calculation module is used to count the number of large identifiers of the video distance large identifier, count the number of equal identifiers of the video distance equal identifiers, and count the number of difference identifiers of the video distance difference identifiers; The first calculation module is used to calculate the sub-surveillance video analysis coefficient of each sub-surveillance video based on the number of large identifiers, the number of equal identifiers and the number of difference identifiers.

[0041] In some embodiments of the present application, the second computing module is used to: The second calculation module is used to compare all sub-surveillance video analysis coefficients with corresponding sub-safety video analysis coefficients one by one, divide all sub-surveillance video analysis coefficients that are smaller than the sub-safety video analysis coefficients into a lower difference coefficient set, and sort the lower difference coefficient set by numerical value; The second calculation module is used to divide all sub-surveillance video coefficients equal to the sub-security video analysis coefficients into a set of equal difference coefficients; The second calculation module is used to divide all sub-surveillance video analysis coefficients that are smaller than the sub-security video analysis coefficient into an upper difference coefficient set, and sort the lower difference coefficient set by numerical value; The second calculation module is used to calculate the vehicle abnormality warning analysis coefficient of the vehicle to be warned according to the lower difference coefficient set, the equal difference coefficient set and the upper difference coefficient set; ; ; ; Among them, w is the vehicle abnormal warning analysis coefficient of the vehicle to be warned, a is the adjustment coefficient of the vehicle abnormal warning analysis coefficient, e(g, k) is the comparison function between the upper difference coefficient set and the lower difference coefficient set, Δd is the weight corresponding to the sub-monitoring video analysis coefficient, and its value is [0.85, 1.25], f is the number of sub-monitoring video analysis coefficients in the upper difference coefficient set, β is the loss factor for calculating the vehicle abnormal warning analysis coefficient, Δt i is the corresponding difference between the sub-monitoring video analysis coefficient in the upper difference coefficient set and the sub-monitoring video analysis coefficient in the lower difference coefficient set, max(Δt i ) is the difference Δt from all i The maximum value obtained in 均 For all differences Δt i The average value, t i is the i-th sub-surveillance video analysis coefficient in the above difference coefficient set, r i is the sub-surveillance video analysis coefficient corresponding to the ith sub-surveillance video analysis coefficient in the lower difference coefficient set; The second calculation module is used to determine the adjustment coefficient a of the vehicle abnormality warning analysis coefficient according to the following method: The second calculation module is used to count the equal number of sub-monitoring video analysis coefficients in the equal difference coefficient set; The second calculation module is used to count the sum Q of the sub-monitoring video analysis coefficients in the upper abnormal coefficient set and the lower difference coefficient set; The second calculation module is used to pre-set a first preset adjustment coefficient, a second preset adjustment coefficient and a third preset adjustment coefficient; The second calculation module is used for using the first preset adjustment coefficient as the adjustment coefficient a of the vehicle abnormality warning analysis coefficient when the equal quantity is less than 0.85Q; The second calculation module is used for using the second adjustment coefficient as the adjustment coefficient a of the vehicle abnormality warning analysis coefficient when the equal quantity is greater than or equal to 0.85Q and less than 1.25Q; The second calculation module is used for using the third adjustment coefficient as the adjustment coefficient a of the vehicle abnormality warning analysis coefficient when the equal quantity is greater than 1.25Q.

[0042] In some embodiments of the present application, the abnormal warning module is used to: The abnormal warning module is used to obtain a preset abnormal warning analysis coefficient of a preset vehicle, and judge whether the vehicle to be warned has an abnormality according to the relationship between the abnormal warning analysis coefficient and the preset abnormal warning analysis coefficient; The abnormal warning module is used to determine that the vehicle to be warned has an abnormality when the abnormal warning analysis coefficient is less than the preset abnormal warning analysis coefficient; The abnormal warning module is used to determine that there is no abnormality in the vehicle to be warned when the abnormal warning analysis coefficient is greater than or equal to the preset abnormal warning analysis coefficient.

[0043] In the description of the above embodiments, specific features, structures, materials or characteristics may be combined in a suitable manner in any one or more embodiments or examples.

[0044] Although the present invention has been described above with reference to the embodiments, various modifications may be made thereto and parts thereof may be replaced with equivalents without departing from the scope of the present invention. In particular, as long as there is no structural conflict, the various features in the embodiments disclosed by the present invention may be used in combination with each other in any manner, and the fact that these combinations are not fully described in this specification is only for the sake of omitting space and saving resources.

[0045] Those skilled in the art can understand that the above are only preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions recorded in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A vehicle sampling abnormality warning method based on video monitoring, characterized in that: include: Determine a vehicle to be warned of abnormality, obtain original video surveillance data corresponding to the vehicle to be warned of abnormality, and process the original video surveillance data to determine a target surveillance video to be analyzed; Dividing the target surveillance video to be analyzed into a plurality of sub-surveillance videos, analyzing the sub-surveillance videos, and calculating a sub-surveillance video analysis coefficient of each sub-surveillance video based on the analysis result; Acquire a security target monitoring video that is similar to the target monitoring video to be analyzed, and analyze the security target monitoring video to determine a corresponding sub-security video analysis coefficient; Classifying the corresponding sub-monitoring video analysis coefficients according to the sub-safety video analysis coefficients, and calculating the vehicle abnormality warning analysis coefficient of the vehicle to be warned of abnormality based on the classification processing result; Based on the vehicle abnormality warning analysis coefficient, it is judged whether the vehicle to be warned of abnormality has abnormality, and when the vehicle to be warned of abnormality has abnormality, an abnormality warning is issued.

2. The vehicle sampling abnormality early warning method based on video monitoring according to claim 1 is characterized in that: When determining a vehicle to be warned of abnormality, acquiring original video surveillance data corresponding to the vehicle to be warned of abnormality, and processing the original video surveillance data to determine a target surveillance video to be analyzed, the method includes: Preprocessing the raw video surveillance data, wherein the preprocessing includes denoising, contrast enhancement, and brightness adjustment; Using an image segmentation algorithm to segment the preprocessed raw video surveillance data into multiple video regions, wherein the image segmentation algorithm includes any one or more combinations of a threshold segmentation method, an edge detection-based segmentation method, a region-based segmentation method, or a deep learning-based segmentation method; Determining a target video area to be extracted from multiple video areas according to a preset extraction rule, wherein the preset extraction rule includes shape, size, position, color feature, texture feature or motion feature; Extracting a target video segment from the target video area, and performing post-processing on the extracted target video segment, wherein the post-processing includes smoothing edges and adjusting color balance; The processed target video clip is output as the target surveillance video to be analyzed.

3. The vehicle sampling abnormality early warning method based on video monitoring according to claim 1 is characterized in that: When the target surveillance video to be analyzed is divided into a plurality of sub-surveillance videos, the sub-surveillance videos are analyzed, and a sub-surveillance video analysis coefficient of each sub-surveillance video is calculated based on the analysis result, the method includes: Extracting a video frame image from the sub-surveillance video and extracting corresponding video pixel points; Clustering all video pixel points to determine video pixel point clusters, and determining cluster centers of the video pixel point clusters; Determine the video distance from each video pixel to the cluster center, and construct a video distance set; Calculating the mean and standard deviation of the video distance set; For all video distances greater than the mean and greater than the standard deviation, generate a video distance large flag; For all video distances that are equal to the mean and the standard deviation, generate video distance iso-identifiers; Generate video distance difference identifiers for all remaining video distances; Count the number of large identifiers of the video distance large identifier, count the number of equal identifiers of the video distance equal identifiers, and count the number of difference identifiers of the video distance difference identifiers; The sub-surveillance video analysis coefficient of each sub-surveillance video is calculated based on the number of large identifiers, the number of equal identifiers and the number of difference identifiers.

4. The vehicle sampling abnormality early warning method based on video monitoring according to claim 1 is characterized in that: When the corresponding sub-monitoring video analysis coefficients are classified according to the sub-safety video analysis coefficients, and the vehicle abnormality warning analysis coefficients of the vehicle to be abnormally warned are calculated based on the classification processing results, the method includes: Compare all sub-surveillance video analysis coefficients with the corresponding sub-safety video analysis coefficients one by one, divide all sub-surveillance video analysis coefficients that are smaller than the sub-safety video analysis coefficients into a lower difference coefficient set, and sort the lower difference coefficient set by numerical value; Dividing all sub-surveillance video coefficients equal to the sub-security video analysis coefficients into a set of equal difference coefficients; All sub-surveillance video analysis coefficients that are smaller than the sub-security video analysis coefficient are divided into an upper difference coefficient set, and the lower difference coefficient set is sorted by numerical value; Calculating a vehicle abnormality warning analysis coefficient of the vehicle to be warned according to the lower difference coefficient set, the equal difference coefficient set and the upper difference coefficient set; ； ； ； Among them, w is the vehicle abnormal warning analysis coefficient of the vehicle to be warned, a is the adjustment coefficient of the vehicle abnormal warning analysis coefficient, e(g, k) is the comparison function between the upper difference coefficient set and the lower difference coefficient set, Δd is the weight corresponding to the sub-monitoring video analysis coefficient, and its value is [0.85, 1.25], f is the number of sub-monitoring video analysis coefficients in the upper difference coefficient set, β is the loss factor for calculating the vehicle abnormal warning analysis coefficient, Δt i is the corresponding difference between the sub-monitoring video analysis coefficient in the upper difference coefficient set and the sub-monitoring video analysis coefficient in the lower difference coefficient set, max(Δt i ) is the sum of all the differences Δt i The maximum value obtained in 均 For all differences Δt i The average value, t i is the i-th sub-surveillance video analysis coefficient in the above difference coefficient set, r i is the sub-surveillance video analysis coefficient corresponding to the ith sub-surveillance video analysis coefficient in the lower difference coefficient set; The adjustment coefficient a of the vehicle abnormality warning analysis coefficient is determined according to the following method: Counting the equal number of sub-monitoring video analysis coefficients in the equal difference coefficient set; Counting the sum Q of the sub-monitoring video analysis coefficients in the upper abnormal coefficient set and the lower difference coefficient set; Presetting a first preset adjustment coefficient, a second preset adjustment coefficient, and a third preset adjustment coefficient; When the equal number is less than 0.85Q, the first preset adjustment coefficient is used as the adjustment coefficient a of the vehicle abnormality warning analysis coefficient; When the equal quantity is greater than or equal to 0.85Q and less than 1.25Q, the second adjustment coefficient is used as the adjustment coefficient a of the vehicle abnormality warning analysis coefficient; When the equal number is greater than 1.25Q, the third adjustment coefficient is used as the adjustment coefficient a of the vehicle abnormality warning analysis coefficient.

5. The vehicle sampling abnormality early warning method based on video monitoring according to claim 1 is characterized in that: When judging whether the vehicle to be warned of abnormality has abnormality based on the vehicle abnormality warning analysis coefficient, it includes: Obtaining a preset abnormal warning analysis coefficient for a preset vehicle, and judging whether the vehicle to be warned of abnormality has an abnormality according to the relationship between the abnormal warning analysis coefficient and the preset abnormal warning analysis coefficient; When the abnormal warning analysis coefficient is less than the preset abnormal warning analysis coefficient, it is determined that the vehicle to be warned of abnormality has an abnormality; When the abnormal warning analysis coefficient is greater than or equal to the preset abnormal warning analysis coefficient, it is determined that there is no abnormality in the vehicle to be warned.

6. A vehicle sampling abnormality warning system based on video monitoring, characterized in that: include: A video processing module is used to determine a vehicle to be warned of abnormality, obtain original video surveillance data corresponding to the vehicle to be warned of abnormality, and process the original video surveillance data to determine a target surveillance video to be analyzed; A first calculation module, used for dividing the target surveillance video to be analyzed into a plurality of sub-surveillance videos, analyzing the sub-surveillance videos, and calculating a sub-surveillance video analysis coefficient of each sub-surveillance video based on the analysis result; A video analysis module, used to obtain a security target monitoring video corresponding to the target monitoring video to be analyzed, and analyze the security target monitoring video to determine a corresponding sub-security video analysis coefficient; A second calculation module is used to classify the corresponding sub-monitoring video analysis coefficient according to the sub-safety video analysis coefficient, and calculate the vehicle abnormality warning analysis coefficient of the vehicle to be abnormally warned based on the classification processing result; The abnormality warning module is used to determine whether the vehicle to be warned of abnormality has abnormality based on the vehicle abnormality warning analysis coefficient, and to issue an abnormality warning when the vehicle to be warned of abnormality has abnormality.

7. The vehicle sampling abnormality warning system based on video monitoring according to claim 6 is characterized in that: The video processing module is used for: The video processing module is used to pre-process the raw video surveillance data, wherein the pre-processing includes denoising, contrast enhancement, and brightness adjustment; The video processing module is used to segment the pre-processed raw video surveillance data into multiple video regions using an image segmentation algorithm, wherein the image segmentation algorithm includes any one or more combinations of a threshold segmentation method, an edge detection-based segmentation method, a region-based segmentation method, or a deep learning-based segmentation method; The video processing module is used to determine a target video area to be extracted from multiple video areas according to a preset extraction rule, wherein the preset extraction rule includes shape, size, position, color feature, texture feature or motion feature; The video processing module is used to extract a target video segment from the target video area and perform post-processing on the extracted target video segment, wherein the post-processing includes smoothing edges and adjusting color balance; The video processing module is used to output the post-processed target video segment as the target monitoring video to be analyzed.

8. The vehicle sampling abnormality warning system based on video monitoring according to claim 6 is characterized in that: The first calculation module is used for: The first calculation module is used to extract a video frame image from the sub-surveillance video and extract corresponding video pixels; The first calculation module is used to cluster all video pixel points, determine video pixel point clusters, and determine the cluster centers of the video pixel point clusters; The first calculation module is used to determine the video distance from each video pixel to the cluster center and construct a video distance set; The first calculation module is used to calculate the mean and standard deviation of the video distance set; The first calculation module is used to generate a large video distance mark for all video distances greater than the mean and greater than the standard deviation; The first calculation module is used to generate video distance and other identifiers for all video distances that are equal to the mean and the standard deviation; The first calculation module is used to generate a video distance difference identifier for all remaining video distances; The first calculation module is used to count the number of large identifiers of the video distance large identifier, count the number of equal identifiers of the video distance equal identifiers, and count the number of difference identifiers of the video distance difference identifiers; The first calculation module is used to calculate the sub-surveillance video analysis coefficient of each sub-surveillance video based on the number of large identifiers, the number of equal identifiers and the number of difference identifiers.

9. The vehicle sampling abnormality warning system based on video monitoring according to claim 6 is characterized in that: The second calculation module is used for: The second calculation module is used to compare all sub-surveillance video analysis coefficients with corresponding sub-safety video analysis coefficients one by one, divide all sub-surveillance video analysis coefficients that are smaller than the sub-safety video analysis coefficients into a lower difference coefficient set, and sort the lower difference coefficient set by numerical value; The second calculation module is used to divide all sub-surveillance video coefficients equal to the sub-security video analysis coefficients into a set of equal difference coefficients; The second calculation module is used to divide all sub-surveillance video analysis coefficients that are smaller than the sub-security video analysis coefficient into an upper difference coefficient set, and sort the lower difference coefficient set by numerical value; The second calculation module is used to calculate the vehicle abnormality warning analysis coefficient of the vehicle to be warned according to the lower difference coefficient set, the equal difference coefficient set and the upper difference coefficient set; ； ； ； Among them, w is the vehicle abnormal warning analysis coefficient of the vehicle to be warned, a is the adjustment coefficient of the vehicle abnormal warning analysis coefficient, e(g, k) is the comparison function between the upper difference coefficient set and the lower difference coefficient set, Δd is the weight corresponding to the sub-monitoring video analysis coefficient, and its value is [0.85, 1.25], f is the number of sub-monitoring video analysis coefficients in the upper difference coefficient set, β is the loss factor for calculating the vehicle abnormal warning analysis coefficient, Δt i is the corresponding difference between the sub-monitoring video analysis coefficient in the upper difference coefficient set and the sub-monitoring video analysis coefficient in the lower difference coefficient set, max(Δt i ) is the sum of all the differences Δt i The maximum value obtained in 均 For all differences Δt i The average value, t i is the i-th sub-surveillance video analysis coefficient in the above difference coefficient set, r i is the sub-surveillance video analysis coefficient corresponding to the ith sub-surveillance video analysis coefficient in the lower difference coefficient set; The second calculation module is used to determine the adjustment coefficient a of the vehicle abnormality warning analysis coefficient according to the following method: The second calculation module is used to count the equal number of sub-monitoring video analysis coefficients in the equal difference coefficient set; The second calculation module is used to count the sum Q of the sub-monitoring video analysis coefficients in the upper abnormal coefficient set and the lower difference coefficient set; The second calculation module is used to pre-set a first preset adjustment coefficient, a second preset adjustment coefficient and a third preset adjustment coefficient; The second calculation module is used for using the first preset adjustment coefficient as the adjustment coefficient a of the vehicle abnormality warning analysis coefficient when the equal quantity is less than 0.85Q; The second calculation module is used for using the second adjustment coefficient as the adjustment coefficient a of the vehicle abnormality warning analysis coefficient when the equal quantity is greater than or equal to 0.85Q and less than 1.25Q; The second calculation module is used for using the third adjustment coefficient as the adjustment coefficient a of the vehicle abnormality warning analysis coefficient when the equal quantity is greater than 1.25Q.

10. The vehicle sampling abnormality warning system based on video monitoring according to claim 6 is characterized in that: The abnormal warning module is used for: The abnormal warning module is used to obtain a preset abnormal warning analysis coefficient of a preset vehicle, and judge whether the vehicle to be warned has an abnormality according to the relationship between the abnormal warning analysis coefficient and the preset abnormal warning analysis coefficient; The abnormal warning module is used to determine that the vehicle to be warned has an abnormality when the abnormal warning analysis coefficient is less than the preset abnormal warning analysis coefficient; The abnormal warning module is used to determine that there is no abnormality in the vehicle to be warned when the abnormal warning analysis coefficient is greater than or equal to the preset abnormal warning analysis coefficient.

Images