A method and system for processing surveillance video data for safe operations in mines.

Abstract

This invention relates to the field of image enhancement technology, specifically to a method and system for processing monitoring video data for mine safety operations. The method includes: processing acquired video images to obtain static images; determining the feature point lost connected components of the t-th and t+1-th static images; determining a first feature point lost connected component and a second feature point lost connected component based on the feature point lost connected components of the t-th and t+1-th static images; determining the illumination influence degree of the first feature point lost connected component based on the degree of grayscale change between the first and second feature point lost connected components; and enhancing the grayscale values ​​of pixels within the first feature point lost connected component using the illumination influence degree as an enhancement coefficient. The method provided by this invention can effectively improve the quality of video images used for safety operation monitoring.

Description

Technical Field

[0001] This invention relates to the field of image enhancement technology, and more specifically to a method and system for processing surveillance video data for safe mining operations. Background Technology

[0002] Mining environments are typically characterized by high risk and danger, involving complex geological conditions, heavy machinery operation, and potential geological disaster risks. To ensure miners' safety, standardize work procedures, improve production efficiency, and facilitate post-accident investigations and liability determination, real-time and effective safety monitoring of mining sites is crucial.

[0003] However, the extremely harsh and complex working environment in mines presents significant challenges to the acquisition of surveillance videos. Underground mines may experience persistent low light levels and flickering light due to unstable lighting equipment; open-pit mines face intense direct sunlight and glare during the day, as well as insufficient lighting in the early morning, late afternoon, or during rainy weather. Furthermore, the complex background of mining areas may contain large amounts of dust, water vapor, heavy machinery, and complex terrain, all of which can interfere with image clarity.

[0004] Under these extreme conditions, the acquired surveillance videos often suffer from serious quality problems. For example, when the lighting changes drastically, the camera's automatic exposure adjustment may not be able to keep up with the speed of the change, resulting in the image being either overexposed (loss of detail in bright areas, making the target appear as a "white blur") or underexposed (overall darkness, loss of detail in dark areas), making important target information such as critical equipment or dangerous areas blurry or even completely lost.

[0005] Commonly used post-processing methods in existing technologies, such as illumination correction, motion blur removal, and intelligent stabilization, often exhibit limitations when facing the extreme and harsh environment of mines, with its drastic changes in lighting and complex backgrounds. These methods may fail to effectively adapt to rapid changes in lighting, struggle to accurately distinguish dynamic interference between the target and the background, resulting in the inability to recover details in overexposed areas, the inability to completely eliminate motion blur, and ultimately, low-quality output video images that cannot meet the demands of mine safety monitoring for clear and reliable visual information. Summary of the Invention

[0006] To address the technical problem that existing video image processing methods cannot effectively improve video image quality, the present invention aims to provide a monitoring video data processing method and system for mine safety operations. The specific technical solution adopted is as follows:

[0007] In a first aspect, embodiments of this application provide a method for processing monitoring video data for mine safety operations, the method comprising:

[0008] The video images collected by the safety operation monitoring are processed to obtain static images of continuous grayscale frames;

[0009] Based on the static image of frame t and the static image of frame (t+1), the feature point loss connected component of the static image of frame t and the feature point loss connected component of the static image of frame (t+1) are determined respectively; where the static image of frame t and the static image of frame (t+1) are two adjacent static images in a series of frames, and t is a positive integer greater than or equal to 1.

[0010] Based on the feature point loss connected components of the static image at frame t and the feature point loss connected components of the static image at frame t+1, the first feature point loss connected component and the second feature point loss connected component are determined; wherein, the first feature point loss connected component and the second feature point loss connected component match, the first feature point loss connected component is the feature point loss connected component of the static image at frame t, and the second feature point loss connected component is the feature point loss connected component of the static image at frame t+1.

[0011] The degree of illumination influence of the lost connected region of the first feature point is determined based on the degree of grayscale change between the lost connected region of the first feature point and the lost connected region of the second feature point, as well as the degree of grayscale change between the connected regions adjacent to the lost connected region of the first feature point and the connected regions adjacent to the lost connected region of the second feature point.

[0012] Using the degree of illumination influence as the enhancement coefficient, the grayscale values ​​of pixels within the lost connected region of the first feature point are enhanced.

[0013] Secondly, embodiments of this application provide a monitoring video data processing system for mine safety operations, the system comprising:

[0014] The first processing module is used to process the video images collected by the safety operation monitoring to obtain static images of continuous grayscale frames;

[0015] The first determining module is used to determine the feature point loss connected component of the t-th frame static image and the feature point loss connected component of the (t+1)-th frame static image, respectively, based on the t-th frame static image and the (t+1)-th frame static image; wherein the t-th frame static image and the (t+1)-th frame static image are two adjacent frames of static images in a continuous frame, and t is a positive integer greater than or equal to 1;

[0016] The second determining module is used to determine the first feature point lost connected component and the second feature point lost connected component based on the feature point lost connected component of the static image of frame t and the feature point lost connected component of the static image of frame t+1; wherein the first feature point lost connected component and the second feature point lost connected component match, the first feature point lost connected component is the feature point lost connected component of the static image of frame t, and the second feature point lost connected component is the feature point lost connected component of the static image of frame t+1.

[0017] The third determining module is used to determine the degree of illumination influence of the lost connected region of the first feature point based on the degree of grayscale change between the lost connected region of the first feature point and the lost connected region of the second feature point, as well as the degree of grayscale change between the connected regions adjacent to the lost connected region of the first feature point and the connected regions adjacent to the lost connected region of the second feature point.

[0018] The second processing module is used to enhance the grayscale values ​​of pixels within the lost connected domain of the first feature point, using the degree of illumination influence as the enhancement coefficient.

[0019] Thirdly, embodiments of this application provide an electronic device including a memory and a processor. The memory stores a computer program that can run on the processor, and when the processor executes the computer program, it can implement the steps in the method provided in the first aspect.

[0020] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps of the method provided in the first aspect.

[0021] The present invention has the following beneficial effects:

[0022] In the video image processing method for safety operation monitoring provided in this application embodiment, by enhancing the grayscale values ​​of pixels within the lost connected region of a feature point, the feature information of the lost connected region can be enhanced. Simultaneously, the surrounding information of the lost connected region is not interfered with, effectively improving the quality of the video captured for safety operation monitoring and providing reliable records and evidence support for safety work. Furthermore, in this application embodiment, the degree of grayscale change between the first and second lost connected regions of a feature point, as well as the degree of grayscale change between connected regions adjacent to the first and second lost connected regions, is comprehensively considered to ultimately determine the degree of illumination influence on the first lost connected region. This effectively avoids contrast imbalance in the first lost connected region due to excessively large parameter variations. Therefore, through the above settings, the method provided in this application embodiment can effectively improve the quality of video images captured for safety operation monitoring. Attached Figure Description

[0023] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 A flowchart illustrating the implementation of a monitoring video data processing method for mine safety operations, provided in an embodiment of this application;

[0025] Figure 2 for Figure 1 The flowchart for implementing step S12 in the process;

[0026] Figure 3 for Figure 1 The flowchart for implementing step S13 in the process;

[0027] Figure 4 A schematic diagram of the composition structure of a monitoring video data processing system for mine safety operations provided in an embodiment of this application;

[0028] Figure 5 This is a schematic diagram of the hardware entity of an electronic device provided in an embodiment of this application. Detailed Implementation

[0029] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a monitoring video data processing method and system for mine safety operations proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0030] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0031] The following description, in conjunction with the accompanying drawings, details a specific scheme for a monitoring video data processing method and system for mine safety operations provided by the present invention.

[0032] It should be noted that the video image processing method for safety operation monitoring provided in this application embodiment can be executed by an electronic device, which can be various types of terminals such as safety operation monitoring devices, laptops, tablets, desktop computers, and mobile devices, or it can be implemented as a server. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0033] Figure 1 This is a schematic diagram illustrating the implementation process of a monitoring video data processing method for mine safety operations provided in an embodiment of this application, as shown below. Figure 1 As shown, the method includes steps S11 to S15, wherein:

[0034] Step S11: Process the video images collected by the safety operation monitoring to obtain static images of continuous grayscale frames.

[0035] It should be noted that, for safety operation monitoring, the video images captured by the monitoring system can be considered as a video of a certain duration consisting of multiple consecutive frames. Therefore, the video can be segmented to obtain a series of still images.

[0036] Based on this, grayscale processing can be performed on the static images of the above consecutive frames to obtain grayscale processed static images.

[0037] In this way, by converting the static image to grayscale, the redundant computational amount of the method provided in this application embodiment for processing static images can be reduced, thereby facilitating the implementation of the method; when implementing the method using a computer program, the running speed of the method can be improved.

[0038] Step S12: Based on the static image of frame t and the static image of frame t+1, determine the feature point loss connected component of the static image of frame t and the feature point loss connected component of the static image of frame t+1, respectively.

[0039] It should be noted that in this embodiment, t is a positive integer greater than or equal to 1. The t-th frame static image and the (t+1)-th frame static image are two adjacent static images in the continuous frame static images after processing the video collected by the safety operation monitoring in the aforementioned step S11.

[0040] In addition, for feature points in static images, feature points can be considered as pixels in static images that are significant and unique. Feature points can be characterized by strong grayscale changes, texture information, or local structures.

[0041] A connected component can be considered a set of connected pixels. By performing superpixel segmentation on a static image frame, several connected components can be formed. The properties of a connected component can be characterized by the number of feature points it contains and the number of pixels it includes. For example, the properties of a connected component can be characterized by the number of feature points it contains, or by the number of feature points per unit area of ​​the connected component. If the number of feature points in a connected component or the number of feature points per unit area of ​​the connected component is less than a certain threshold, the connected component can be considered a feature-loss connected component.

[0042] By analyzing the static image of frame t and the static image of frame (t+1) respectively, we can determine the feature point loss connected components of the static image of frame t and the static image of frame (t+1).

[0043] Step S13: Based on the feature point loss connected components of the t-th frame static image and the feature point loss connected components of the (t+1)-th frame static image, determine the first feature point loss connected component and the second feature point loss connected component.

[0044] It should be noted that, in the embodiments of this application, the first feature point lost connected component is the feature point lost connected component in the static image of frame t, and the second feature point lost connected component is the feature point lost connected component in the static image of frame t+1; and the first feature point lost connected component and the second feature point lost connected component match.

[0045] For example, in this embodiment, firstly, pixels in the t-th frame of the static image can be matched with pixels in the (t+1)-th frame of the static image to establish the correspondence between the pixels, thereby establishing the correspondence between the t-th frame of the static image and the (t+1)-th frame of the static image. Next, feature points in the t-th frame of the static image can be matched with feature points in the (t+1)-th frame of the static image. Finally, the lost connected components of feature points in the t-th frame of the static image can be matched with the lost connected components of feature points in the (t+1)-th frame of the static image.

[0046] Step S14: Based on the degree of grayscale change between the first feature point lost connected component and the second feature point lost connected component, as well as the degree of grayscale change between the connected components adjacent to the first feature point lost connected component and the connected components adjacent to the second feature point lost connected component, determine the degree of illumination influence of the first feature point lost connected component.

[0047] For the feature point lost connected components in the static image of frame t and frame t+1, when the ambient light changes, the gray values ​​of the pixels included in the aforementioned feature point lost connected components will change, which will reduce the matching degree of the feature points in the aforementioned feature point lost connected components and may result in the loss of information contained in the aforementioned feature point lost connected components.

[0048] It should be noted that for a feature point in the t-th frame of the static image that has lost a connected component, there exists a connected component adjacent to that feature point. Similarly, for a feature point in the (t+1)-th frame of the static image that has lost a connected component, there also exists a connected component adjacent to that feature point.

[0049] Based on this, by comparing the gray value of the lost connected region of the first feature point with the gray value of the lost region of the second feature point, the degree of gray value change between the lost connected regions of the feature point can be obtained; and by comparing the gray value of the connected region adjacent to the lost connected region of the first feature point with the gray value of the connected region adjacent to the lost region of the second feature point, the degree of gray value change between the connected regions can be obtained.

[0050] It should be noted that the degree of grayscale variation mentioned above can be characterized by the difference, ratio, variance, or standard deviation of grayscale values. In addition, other parameters can also be introduced for characterization.

[0051] Furthermore, by comprehensively considering the degree of grayscale change between the first feature point's lost connected domain and the second feature point's lost connected domain, as well as the degree of grayscale change between the connected domains adjacent to the first feature point's lost connected domain and the connected domains adjacent to the second feature point's lost connected domain, the degree of illumination influence between the first feature point's lost connected domains can be determined.

[0052] It is understandable that the aforementioned degree of illumination influence can be considered as the degree of change in the gray value of the lost connected region of the first feature point caused by illumination factors.

[0053] Step S15: Using the degree of illumination influence as the enhancement coefficient, enhance the gray values ​​of the pixels included in the lost connected region of the first feature point.

[0054] Based on the illumination influence degree of the lost connected region of the first feature point obtained in the aforementioned steps, the grayscale value of the pixel in the lost connected region of the first feature point can be adjusted by using the illumination influence degree as an enhancement coefficient.

[0055] For example, a curve can be established to correspond to the degree of illumination influence and the gray value of the aforementioned pixel, and a correspondence can be established between the magnitude of the illumination influence and the adjustment magnitude of the gray value of the aforementioned pixel. Based on the aforementioned correspondence, the gray value of the pixel in the lost connected region of the feature point can be adjusted.

[0056] For example, the positive or negative value of the illumination influence degree representation can represent the change in grayscale value of the pixel in the first feature point loss connected region due to increased or decreased illumination intensity, respectively. When the illumination influence degree representation value is positive (equivalent to a certain degree of overexposure), the grayscale value of the pixel in the feature point loss connected region of the current frame's static image can be increased.

[0057] It should be noted that by processing video images collected for safety operation monitoring frame by frame using the above method, the correction of pixels in the connected domains where feature points are lost in each frame of the static image can be completed, thereby completing the processing of the video images.

[0058] Through the above settings, the video image processing method for safety operation monitoring provided in this application embodiment enhances the feature information of the lost connected region by strengthening the grayscale values ​​of pixels within the lost connected region of the feature point. Simultaneously, it avoids interfering with the surrounding information of the lost connected region, effectively improving the quality of the video captured for safety operation monitoring and providing reliable records and evidence support for safety work. Furthermore, in this application embodiment, the degree of grayscale change between the first and second lost connected regions of the feature point, as well as the degree of grayscale change between connected regions adjacent to the first and second lost connected regions, is comprehensively considered to ultimately determine the degree of illumination influence on the first lost connected region. This effectively avoids contrast imbalance in the first lost connected region due to excessively large parameter variations. Therefore, through the above settings, the method provided in this application embodiment can effectively improve the quality of video images captured for safety operation monitoring.

[0059] The targets captured by safety operation monitoring are usually dynamic, and information can be lost due to varying ambient light and complex backgrounds. Especially at night, inconsistent ambient light intensity can cause information loss in certain areas of the captured target. To address this, this application also provides a method for processing monitoring video data in mine safety operations. In this method, reference is made to… Figure 2 The above step S12 can be implemented through steps S121 to S125, wherein:

[0060] Step S121: Construct an optical flow matching model between the static image of frame t and the static image of frame t+1.

[0061] It should be noted that the optical flow field is a two-dimensional vector field that reflects the changing trend of each pixel in a static image. It can be considered as the instantaneous velocity field generated by the motion of pixels with grayscale values ​​on the static image plane. By analyzing the optical flow field, the motion field of pixels that cannot be directly obtained can be approximated from consecutive frames of static images. The information contained in the optical flow field can be considered as the instantaneous velocity vector information of each pixel. By constructing an optical flow matching model between the t-th frame and the (t+1)-th frame, relevant feature points between the t-th and (t+1)-th frames can be found. Based on the optical flow vector formed by these relevant feature points, the relative motion of objects between the t-th and (t+1)-th frames can be inferred.

[0062] For example, in this embodiment of the application, the group characteristics of each pixel can be determined by traversing each pixel in the t-th frame static image and the (t+1)-th frame static image. Furthermore, the local grayscale features of pixels within the four-neighborhood or eight-neighborhood of each pixel can also be obtained.

[0063] Based on this, by combining the cluster features and local grayscale features of the pixels in the t-th frame static image and the (t+1)-th frame static image, an optical flow matching model between the t-th frame static image and the (t+1)-th frame static image can be constructed.

[0064] Step S122: Based on the optical flow matching model, match the pixels of the t-th frame static image and the (t+1)-th frame static image.

[0065] Based on the optical flow matching model constructed in the aforementioned steps between the t-th frame static image and the (t+1)-th frame static image, it is possible to match the pixels of the t-th frame static image with the pixels of the (t+1)-th frame static image.

[0066] Specifically, based on the aforementioned optical flow matching model, and considering the variance of the gray levels in the eight neighborhoods of any pixel in the t-th frame of the static image, the mean of the gray levels in the eight neighborhoods, and the number of pixel groups with the same gray level value; and based on the variance of the gray levels in the eight neighborhoods of any pixel in the (t+1)-th frame of the static image, the same factors can be used to establish a Euclidean distance model between pixels in the t-th frame and pixels in the (t+1)-th frame of the static image. The pixel corresponding to the minimum value of the aforementioned Euclidean distance model is the pixel that is matched between the t-th frame and the (t+1)-th frame of the static image.

[0067] Step S123: Perform superpixel segmentation on the t-th frame static image to obtain the connected components of the t-th frame static image; perform superpixel segmentation on the (t+1)-th frame static image to obtain the connected components of the (t+1)-th frame static image.

[0068] Furthermore, by performing superpixel segmentation on the t-th frame static image, the connected components of the t-th frame static image can be obtained; by performing superpixel segmentation on the (t+1)-th frame static image, the connected components of the (t+1)-th frame static image can be obtained.

[0069] Step S124: Determine the feature point density of each connected component based on the number of feature points and the number of pixels in each connected component.

[0070] By comparing the number of feature points in each connected region of the t-th frame static image with the number of pixels, the feature point density of each connected region of the t-th frame static image can be obtained.

[0071] Similarly, by comparing the number of feature points in each connected region of the (t+1)th frame of the static image with the number of pixels, the feature point density of each connected region of the (t+1)th frame of the static image can be obtained.

[0072] Specifically, the connected components after segmentation of the t-th frame static image can be assigned values. The feature point density of the connected components of the t-th frame of the static image is calculated using the following formula.

[0073] ;

[0074] in, Represents the t-th frame of the still image. The feature point density of a connected region Represents connected components The number of pixels, Represents the corresponding connected components The number of feature points. The feature point density within the connected components of the corresponding t-th frame of the static image can be obtained by the ratio.

[0075] Similarly, the connected components after segmentation of the (t+1)th frame of the static image can be assigned values. By using the above method, the feature point density within the connected components of the (t+1)th frame of the static image can be obtained.

[0076] Step S125: If the feature point density of the connected component in the static image of frame t is less than a preset threshold, the connected component is determined to be a feature point loss connected component of the static image of frame t; if the feature point density of the connected component in the static image of frame t+1 is less than a preset threshold, the connected component is determined to be a feature point loss connected component of the static image of frame t+1.

[0077] Based on the feature point density of each connected component obtained in the above steps, by setting a preset threshold, if the feature point density of a connected component is less than the preset threshold, the connected component can be considered as a feature point loss connected component.

[0078] For example, the feature point density calculated using the above formula In other words, a preset threshold can be set. ,exist Less than The corresponding connected component can be considered as the feature point lost connected component.

[0079] With the above settings, the video image processing method for safety operation monitoring provided in this application embodiment first constructs an optical flow matching model for static images of consecutive frames, performs feature matching on the target captured by the safety operation monitoring, and obtains the corresponding feature points. Based on this, superpixel segmentation is performed on the static image to obtain connected components after superpixel segmentation, and the feature point loss and exposure of each connected component are analyzed. For connected components where overexposure leads to feature point loss, dynamic range compression is performed. By reducing the difference between areas with large brightness changes and areas with small brightness changes in the static image, the dynamic range of the overall image is reduced, preserving the details and information of the static image.

[0080] It should be noted that the optical flow field formed by the static image information of consecutive frames of video graphics captured at night is greatly affected by ambient light, which is constantly changing. Therefore, once the lighting environment changes, the optical flow field obtained solely based on the gray-level similarity between pixels in adjacent static images becomes exceptionally sparse. However, regions with uniform gray-level in adjacent static images are less affected by changes in ambient light. By capturing the optical flow field formed by relevant uniform gray-level regions within a dynamic target area, the optical flow vector of the corresponding target can be inferred.

[0081] Therefore, in some embodiments of this application, step S121 described above can be implemented by steps S1211 to S1213 (not shown in the figure), wherein:

[0082] Step S1211: Traverse each pixel in the t-th frame static image and the (t+1)-th frame static image, and obtain the local grayscale features of the pixels in the eight neighborhoods of each pixel.

[0083] Step S1212: Based on local grayscale features and the group features of each pixel, determine the Euclidean distance model between the pixels of the t-th frame static image and the (t+1)-th frame static image.

[0084] Step S1213: The minimum value of the Euclidean distance model is used as the optical flow matching model between the t-th frame static image and the (t+1)-th frame static image.

[0085] Specifically, by traversing every pixel in the t-th and t+1-th static images, the local grayscale features of the pixels in the eight-neighborhood of each pixel can be obtained. Based on this, and combined with the group features of each pixel, the Euclidean distance model between the pixels in the t-th and t+1-th static images can be determined.

[0086] It should be noted that the cluster feature of a pixel can be considered as the number of times the gray value of that pixel appears in a static image, or as the number of clusters of pixels with the same gray value as that pixel.

[0087] In addition, the local grayscale features of pixels within the eight-neighborhood of a pixel can be characterized by the grayscale variance or grayscale mean of the pixels within the eight-neighborhood of that pixel.

[0088] Based on this, we can establish the following Euclidean distance model between the pixels of the t-th frame of the static image and the pixels of the (t+1)-th frame of the static image.

[0089] ;

[0090] Where t represents the t-th frame of the still image, t+1 represents the (t+1)-th frame of the still image, i represents any pixel in the t-th frame of the still image, and v represents any pixel in the (t+1)-th frame of the still image. Let represent the grayscale variance of the eight neighboring pixels of the i-th pixel in the (t+1)-th frame of the still image. Let V represent the grayscale variance of the eight neighbors of the v-th pixel in the (t+1)-th frame of the static image. Let represent the average gray value of the eight neighboring pixels of the i-th pixel in the t-th frame of the still image. Let represent the average gray value of the eight neighboring pixels of the v-th pixel in the (t+1)-th frame of the static image; This represents the number of clusters in the t-th frame of the still image, where the grayscale value of the i-th pixel is represented. This represents the number of groups of the v-th pixel in the (t+1)-th frame of the static image. The Euclidean distance is represented by the Euclidean distance model between the i-th pixel of the t-th frame static image and the v-th pixel of the (t+1)-th frame static image; min represents the minimum value function.

[0091] The closer the result calculated by the above formula is to 0, the higher the matching degree between the i-th pixel and the v-th pixel. Based on this, the pixel with the smallest Euclidean distance between the pixel of the (t+1)-th frame of the static image and the pixel of the t-th frame of the static image is taken as the matching point.

[0092] Based on this, a threshold can also be set. To determine the credibility of the match between the two pixels mentioned above, specifically, in In this case, it can be considered and The reliability of the matching results between the two pixels is relatively high because the matching results that meet the above conditions are all pixels in the uniform area inside the target rather than unstable pixels in the edge area. The pixel optical flow matching results are considered unreliable.

[0093] With the above settings, in the video image processing method for safety operation monitoring provided in this application embodiment, the characteristic that uniform regions are not sensitive to changes in ambient light, while non-uniform regions are sensitive to changes in ambient light, is utilized. Based on the mean gray level, gray level variance, and the group characteristics of the eight neighboring pixels of the t-th frame static image or the t+1-th frame static image, a corresponding Euclidean distance model is constructed as parameters. By judging the Euclidean distance of the corresponding pixels, the matching degree of related pixels is analyzed, which can make the optical flow field between consecutive frames of static images denser.

[0094] Based on this, refer to Figure 3 In some embodiments of this application, step S13 described above can be implemented by steps S131 to S135, wherein:

[0095] Step S131: Determine the interference edges of the static image t based on the vector mode of all feature points in the static image t.

[0096] Step S132: Eliminate the lost connected components of feature points located within the interference edges of the t-th frame static image;

[0097] Step S133: Determine the interference edges of the static image in frame t+1 based on the vector mode of all feature points in the static image in frame t+1.

[0098] Step S134: Eliminate the lost connected components of feature points located within the interference edges of the (t+1)th frame static image;

[0099] Step S135: Match the remaining feature point lost connected components in the static image of frame t with the remaining feature point lost connected components in the static image of frame t+1 to determine the first feature point lost connected component and the second feature point lost connected component.

[0100] It is understandable that for connected regions where feature points are lost, the loss of feature points may be due to factors such as changes in mine dust, resulting in the loss of feature points in local areas. Therefore, by obtaining the vector mode of all feature points in a static image, the overall movement of the background region can be determined.

[0101] Therefore, in this embodiment, based on the vector mode of all feature points in the t-th frame static image, interference edges in the t-th frame static image can be determined; based on the vector mode of all feature points in the (t+1)-th frame static image, interference edges in the (t+1)-th frame static image can be determined. Based on this, the feature point lost connected components located at the interference edges in the t-th frame static image can be excluded, and the feature point lost connected components located at the interference edges in the (t+1)-th frame static image can also be excluded. Then, the remaining feature point lost connected components in the t-th frame static image and the remaining feature point lost connected components in the (t+1)-th frame static image are matched to obtain the first and second feature point lost connected components.

[0102] That is, in the embodiments of this application, the first feature point lost connected domain is the feature point lost connected domain after excluding the feature point lost connected domain within its interference edge in the static image of frame t, and the second feature point lost connected domain is the feature point lost connected domain after excluding the feature point lost connected domain within its interference edge in the static image of frame t+1, and the first feature point lost connected domain and the second feature point lost connected domain match.

[0103] Based on this, in some embodiments of this application, step S135 described above can be implemented through the following steps:

[0104] Based on the gray-level variance, mean edge gradient, and area of ​​the remaining feature points lost connected components in the t-th frame static image and the (t+1)-th frame static image, a cosine similarity model is constructed between the remaining feature points lost connected components in the t-th frame static image and the (t+1)-th frame static image. When the cosine similarity model reaches its maximum value, the two feature point lost connected components are the first feature point lost connected component and the second feature point lost connected component.

[0105] It is understood that, in the embodiments of this application, the gray-level variance, mean edge gradient, and area of ​​the feature point lost connected region of the t-th frame static image and the feature point lost connected region of the (t+1)-th frame static image can be used as judgment factors to determine the matching of the first feature point lost connected region and the second feature point lost connected region.

[0106] Based on this, the following cosine similarity model can be established:

[0107] ;

[0108] in, This indicates that any feature point in the t-th frame of the static image has lost its connected components. This indicates that any feature point in the (t+1)th frame of the static image has lost its connected components. This represents the gray-level variance within the lost connected component of the B-th feature point in the t-th frame of the static image. This represents the gray-level variance within the lost connected component of the C-th feature point in the (t+1)-th frame of the static image. This represents the mean edge gradient of the lost connected component at the B-th feature point in the t-th frame of the static image. This represents the mean edge gradient of the lost connected component at the Cth feature point in the (t+1)th frame of the static image. This represents the area of ​​the B-th feature missing connected component within the t-th frame of the static image. This represents the area of ​​the connected component where the Cth feature is lost within the (t+1)th frame of the static image; denoted by , represents the cosine value of the cosine similarity model between the lost connected component of the Bth feature point in the t-th frame of the static image and the lost connected component of the Cth feature point in the (t+1)-th frame of the static image; max represents the maximum value function.

[0109] It should be noted that, The cosine similarity model is constructed using three attribute parameters: the gray-level variance, the mean edge gradient, and the area feature of the lost connected components of the B-th feature point in the t-th frame and the C-th feature point in the (t+1)-th frame. The closer the calculated cosine value is to 1, the higher the matching degree between the B-th and C-th feature point connected components. The lost connected component of the feature point with the highest matching degree between the (t+1)-th frame and the B-th feature point connected component in the t-th frame can be taken as its corresponding matched lost connected component.

[0110] In addition, in this embodiment of the application, a threshold can be set to determine the reliability of the matching result of the two feature points losing connected components. When it is greater than or equal to this threshold, it can be considered that The matching results between two feature points with missing connected components have high reliability because the corresponding missing connected components of the feature points participating in the matching are the filtered missing connected components of the feature points. If the value is less than this threshold, the matching result of the two feature points losing connected components is considered unreliable, and there may be motion or occlusion. For example, the threshold can be set to 0.5.

[0111] With the above settings, in the method provided in this application embodiment, a cosine similarity model is constructed using the gray-level variance, mean edge gradient, and area of ​​the connected regions of the feature points lost in consecutive frame static images as matching conditions to match the feature point lost connected regions of the t-th frame static image and the t+1-th frame static image, obtaining the matched first and second feature point lost connected regions. By filtering feature point lost connected regions caused by ambient lighting issues based on the gray-level feature changes within the successfully matched first and second feature point lost connected regions and the gray-level changes of adjacent connected regions, the corresponding parameters are adjusted to process the pixels of the feature point lost connected regions according to the gray-level changes. This further enhances the feature information of the feature point lost connected regions provided in this application embodiment, while not interfering with the surrounding information of the feature point lost connected regions, effectively improving the quality of the video captured for safety operation monitoring.

[0112] In addition, in some embodiments of this application, step S14 described above can be implemented by the following steps:

[0113] Obtain the N nearest neighbor connected components for each connected component; establish a function to determine the degree of illumination influence on the connected components where the first feature point is lost.

[0114] ;

[0115] Where q represents the lost connected component of the first feature point. This indicates the degree of illumination impact caused by the loss of connected components at the first feature point. Let represent the lost connected component of the second feature point in the (t+1)th frame of the static image, and the mean gray value of the vth connected component adjacent to the lost connected component of the second feature point. Let represent the average grayscale value of the first feature point lost connected component and the vth connected component adjacent to the first feature point lost connected component of the static image in frame t. Let represent the distance between the v-th connected component and the centroid of the lost connected component q at the first feature point; N represents the number of adjacent connected components for each connected component; and e represents the natural constant. It should be noted that when v = q, The distance represented is 0. The corresponding value is the largest, which can be considered as It represents a multiplication coefficient.

[0116] Based on this, it can be considered that This represents the degree of illumination influence obtained by multiplying the gray level difference between the v-th connected components by the multiplication coefficient. When v=q, it represents the gray level difference between adjacent frames of the connected component where the feature point is lost, and the multiplication coefficient is the largest at this time.

[0117] when This indicates that the loss of feature points is caused by increased illumination intensity in the connected regions. This indicates that the loss of feature points is due to the weakening of the illumination intensity in the connected region.

[0118] With the above settings, the method provided in this application embodiment determines the final degree of illumination influence by utilizing the difference in grayscale changes between the associated region and the surrounding region. When processing the grayscale of the region using the illumination influence degree obtained through joint judgment, it can effectively avoid regional contrast imbalance caused by excessively large changes in parameters.

[0119] In addition, in some embodiments of this application, step S15 described above can be implemented by the following steps:

[0120] The degree of light influence was normalized to the range of [-1, 1].

[0121] Using the normalized illumination effect as the enhancement coefficient, linear enhancement processing is performed on the pixels in the lost connected region of the feature point of the t-th frame static image.

[0122] For example, the pixels within the lost connected region of the feature points in the t-th frame of the static image can be linearly enhanced according to the following formula:

[0123] ;

[0124] This represents the enhanced grayscale value of the u-th pixel in the connected component where feature points of the t-th frame of the static image are lost. This indicates the degree of illumination impact on the loss of connected components at feature points. This represents the grayscale value of the u-th pixel in the lost connected component of the feature points of the t-th frame of the static image.

[0125] Understandably, by processing each frame of static image in the video footage collected from safety operation monitoring in the manner described above, the degree of information loss caused by lighting effects can be reduced, providing reliable records and evidence support for safety work.

[0126] Based on this, embodiments of this application also provide a monitoring video data processing system for mine safety operations, such as... Figure 4 As shown, the monitoring video data processing system 500 for mine safety operations includes a first processing module 501, a first determining module 502, a second determining module 503, a third determining module 504, and a second processing module 505. Wherein:

[0127] The first processing module 501 is used to process the video images collected by the safety operation monitoring to obtain static images of continuous grayscale frames;

[0128] The first determining module 502 is used to determine the feature point loss connected component of the static image of frame t and the feature point loss connected component of the static image of frame t+1, respectively, based on the static image of frame t and the static image of frame t+1; wherein, the static image of frame t and the static image of frame t+1 are two adjacent static images in a series of static images, and t is a positive integer greater than or equal to 1;

[0129] The second determining module 503 is used to determine the first feature point lost connected component and the second feature point lost connected component based on the feature point lost connected component of the static image of frame t and the feature point lost connected component of the static image of frame t+1; wherein the first feature point lost connected component and the second feature point lost connected component match, the first feature point lost connected component is the feature point lost connected component of the static image of frame t, and the second feature point lost connected component is the feature point lost connected component of the static image of frame t+1.

[0130] The third determining module 504 is used to determine the degree of illumination influence of the first feature point lost connected domain based on the degree of grayscale change between the first feature point lost connected domain and the second feature point lost connected domain, as well as the degree of grayscale change between the connected domains adjacent to the first feature point lost connected domain and the connected domains adjacent to the second feature point lost connected domain.

[0131] The second processing module 505 is used to enhance the grayscale values ​​of pixels within the lost connected region of the first feature point, using the degree of illumination influence as the enhancement coefficient.

[0132] The descriptions of the apparatus embodiments above are similar to those of the method embodiments above, and have similar beneficial effects. In some embodiments, the functions or modules included in the apparatus provided in this application can be used to perform the methods described in the method embodiments above. For technical details not disclosed in the apparatus embodiments of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0133] It should be noted that, in the embodiments of this application, if the above methods are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to related technologies, can be embodied in the form of software products. These software products are stored in a storage medium and include several instructions to cause an electronic device (which may be a personal computer, server, network device, or vehicle-mounted computer, etc.) to execute all or part of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks. Thus, the embodiments of this application are not limited to any specific hardware, software, or firmware, or any combination of hardware, software, and firmware.

[0134] This application also provides an electronic device, which includes a processor, a memory, a communication interface and a communication bus. The memory stores a computer program that can run on the processor. When the processor executes the program, it implements some or all of the steps in the above method.

[0135] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements some or all of the steps in the above-described method. The computer-readable storage medium can be transient or non-transient.

[0136] This application also provides a computer program including computer-readable code. When the computer-readable code runs in an electronic device, the processor in the electronic device executes some or all of the steps in the above-described method.

[0137] This application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program. When the computer program is read and executed by a computer, it implements some or all of the steps in the above-described method. This computer program product can be implemented specifically through hardware, software, or a combination thereof. In some embodiments, the computer program product is specifically embodied as a computer storage medium; in other embodiments, the computer program product is specifically embodied as a software product, such as a software development kit (SDK), etc.

[0138] It should be noted that the descriptions of the various embodiments above tend to emphasize the differences between them, while their similarities or commonalities can be referred to interchangeably. The descriptions of the above embodiments of the device, storage medium, computer program, and computer program product are similar to the descriptions of the above method embodiments and have similar beneficial effects. For technical details not disclosed in the embodiments of the device, storage medium, computer program, and computer program product of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0139] It should be noted that, referring to Figure 5 This is a schematic diagram of the hardware entity of the electronic device in the embodiments of this application, such as... Figure 5 As shown, the hardware entity of the electronic device 600 includes: a processor 601, a communication interface 602, a memory 603, and a communication bus 604, wherein:

[0140] Processor 601 typically controls the overall operation of electronic device 600;

[0141] Communication interface 602 and communication bus 604 enable electronic device 600 to communicate with other terminals or servers through a network;

[0142] The memory 603 is configured to store instructions and applications executable by the processor 601, and can also cache data to be processed or already processed (e.g., image data, audio data, voice communication data, and video communication data) in the processor 601 and various modules in the electronic device 600. It can be implemented using flash memory or random access memory (RAM). Data transfer between the processor 601, the communication interface 602, and the memory 603 can be performed via bus 604.

[0143] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of this application, the sequence numbers of the above steps / processes do not imply a sequential order of execution; the execution order of each step / process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. The sequence numbers of the above embodiments of this application are merely descriptive and do not represent the superiority or inferiority of the embodiments.

[0144] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0145] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0146] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0147] In addition, each functional unit in the various embodiments of this application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0148] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0149] Alternatively, if the integrated units described above are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence or the part that contributes to related technologies, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an electronic device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0150] The above are merely embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for processing monitoring video data for safe operation of a mine, characterized by, The method comprises the following steps: processing video images collected by safety operation monitoring to obtain gray-scale continuous frame static images; based on the tth frame static image and the t+1th frame static image, respectively determining the feature point loss connected domain of the tth frame static image and the feature point loss connected domain of the t+1th frame static image; wherein, the tth frame static image and the t+1th frame static image are adjacent two frame static images in the continuous frame static images, and t is a positive integer greater than or equal to 1; based on the feature point loss connected domain of the tth frame static image and the feature point loss connected domain of the t+1th frame static image, determining the first feature point loss connected domain and the second feature point loss connected domain; wherein, the first feature point loss connected domain and the second feature point loss connected domain are matched, the first feature point loss connected domain is the feature point loss connected domain of the tth frame static image, and the second feature point loss connected domain is the feature point loss connected domain of the t+1th frame static image; based on the gray scale change degree between the first feature point loss connected domain and the second feature point loss connected domain, and the gray scale change degree between the connected domain adjacent to the first feature point loss connected domain and the connected domain adjacent to the second feature point loss connected domain, determining the illumination influence degree of the first feature point loss connected domain; taking the illumination influence degree as the enhancement coefficient, enhancing the gray value of the pixel points in the first feature point loss connected domain; the method based on the tth frame static image and the t+1th frame static image, respectively determining the feature point loss connected domain of the tth frame static image and the feature point loss connected domain of the t+1th frame static image, comprises: constructing an optical flow matching model between the tth frame static image and the t+1th frame static image; based on the optical flow matching model, matching the pixel points of the tth frame static image and the t+1th frame static image; performing superpixel segmentation on the tth frame static image to obtain the connected domain of the tth frame static image; performing superpixel segmentation on the t+1th frame static image to obtain the connected domain of the t+1th frame static image; based on the number of feature points and the number of pixel points in each connected domain, determining the feature point density of each connected domain; in the case that the feature point density of the connected domain of the tth frame static image is less than a preset threshold, determining that the connected domain is the feature point loss connected domain of the tth frame static image; in the case that the feature point density of the connected domain of the t+1th frame static image is less than a preset threshold, determining that the connected domain is the feature point loss connected domain of the t+1th frame static image.

2. The method for processing monitoring video data for safe operation of a mine according to claim 1, characterized in that, the method of constructing the optical flow matching model between the tth frame static image and the t+1th frame static image comprises: traversing each pixel point in the tth frame static image and the t+1th frame static image to obtain the local gray scale features of the pixel points in the eight-neighbor domain of each pixel point; based on the local gray scale features and the population characteristics of each pixel point, determining the Euclidean distance model between the pixel points of the tth frame static image and the t+1th frame static image; Take the minimum value of the Euclidean distance model as a light flow matching model between the tth frame static image and the t+1th frame static image.

3. The method for processing monitoring video data for safe operation of a mine according to claim 2, characterized in that, The light flow matching model is: ； wherein t represents the tth frame of static image, t+1 represents the t+1th frame of static image, i represents any one pixel point of the tth frame of static image, v represents any one pixel point of the t+1th frame of static image, a gray scale variance of the pixel points of the eight-neighborhood of the ith pixel point of the t+1th frame of static image, a gray scale variance of the pixel points of the eight-neighborhood of the vth pixel point of the t+1th frame of static image; a gray scale mean of the pixel points of the eight-neighborhood of the ith pixel point of the tth frame of static image, a gray scale mean of the pixel points of the eight-neighborhood of the vth pixel point of the t+1th frame of static image; a population number of the gray scale value of the ith pixel point of the tth frame of static image in the tth frame of static image, a population number of the vth pixel point of the t+1th frame of static image in the t+1th frame of static image; a Euclidean distance of the Euclidean distance model between the ith pixel point of the tth frame of static image and the vth pixel point of the t+1th frame of static image; and min represents a minimum function.

4. The method for processing monitoring video data for safe operation of a mine according to claim 1, characterized in that, The first feature point loss connected domain and the second feature point loss connected domain are determined based on the feature point loss connected domain of the tth frame static image and the feature point loss connected domain of the t+1th frame static image, and the determination includes: An interference edge of the tth frame static image is determined based on the vector mode of all feature points in the tth frame static image. The feature point loss connected domain in the tth frame static image located in the interference edge of the tth frame static image is excluded. An interference edge of the t+1th frame static image is determined based on the vector mode of all feature points in the t+1th frame static image. The feature point loss connected domain in the t+1th frame static image located in the interference edge of the t+1th frame static image is excluded. The first feature point loss connected domain and the second feature point loss connected domain are determined by matching the remaining feature point loss connected domain in the tth frame static image and the remaining feature point loss connected domain in the t+1th frame static image.

5. The method for processing monitoring video data for safe operation of a mine according to claim 4, characterized in that, The first feature point loss connected domain and the second feature point loss connected domain are determined by matching the remaining feature point loss connected domain in the tth frame static image and the remaining feature point loss connected domain in the t+1th frame static image, and the determination includes: A cosine similarity model between the remaining feature point loss connected domain in the tth frame static image and the remaining feature point loss connected domain in the t+1th frame static image is constructed based on the gray variance, the edge gradient mean and the area of the remaining feature point loss connected domain in the tth frame static image and the remaining feature point loss connected domain in the t+1th frame static image. The two feature point loss connected domains corresponding to the maximum value of the cosine similarity model are the first feature point loss connected domain and the second feature point loss connected domain.

6. The method for processing monitoring video data for safe operation of a mine according to claim 5, wherein, The cosine similarity model is: ； wherein, represents a loss of connectivity domain of any one feature point in the tth frame of static image, represents a loss of connectivity domain of any one feature point in the t+1th frame of static image, represents a gray scale variance inside the Bth feature point loss of connectivity domain in the tth frame of static image, represents a gray scale variance inside the Cth feature point loss of connectivity domain in the t+1th frame of static image, represents an edge gradient mean value of the Bth feature point loss of connectivity domain in the tth frame of static image, represents an edge gradient mean value of the Cth feature point loss of connectivity domain in the t+1th frame of static image, represents an area of the Bth feature loss of connectivity domain in the tth frame of static image, represents an area of the Cth feature loss of connectivity domain in the t+1th frame of static image; represents a cosine value of a cosine similarity model between the Bth feature point loss of connectivity domain in the tth frame of static image and the Cth feature point loss of connectivity domain in the t+1th frame of static image; max represents a maximum value function.

7. The method for processing monitoring video data for safe operation of a mine according to claim 1, wherein, The illumination influence degree of the first feature point loss connected domain is determined based on the gray change degree between the first feature point loss connected domain and the second feature point loss connected domain, and the gray change degree between the connected domain adjacent to the first feature point loss connected domain and the connected domain adjacent to the second feature point loss connected domain, and the determination includes: The N adjacent connected domains of each connected domain are obtained. An illumination influence degree function of the first feature point loss connected domain is established: ； wherein q represents a first feature point loss connected domain, represents a degree of illumination influence of the first feature point loss connected domain, represents a gray mean value of a second feature point loss connected domain of the t+1th frame of static image and a vth connected domain adjacent to the second feature point loss connected domain, represents a gray mean value of a first feature point loss connected domain of the tth frame of static image and a vth connected domain adjacent to the first feature point loss connected domain, represents a distance between the vth connected domain and a centroid of the first feature point loss connected domain q; N represents a number of adjacent connected domains of each connected domain; and e represents a natural constant.

8. The method for processing monitoring video data for safe operation of a mine according to claim 1, wherein, The gray value of the pixel point included in the first feature point loss connected domain is enhanced by taking the illumination influence degree as an enhancement coefficient, and the enhancement processing includes: The illumination influence degree is normalized in the [-1, 1] interval. The pixel point in the feature point loss connected domain of the tth frame static image is linearly enhanced by taking the normalized illumination influence degree as the enhancement coefficient.

9. A monitoring video data processing system for use in a mine safety operation, characterized by, The system includes: A first processing module is configured to process video images collected by safety operation monitoring to obtain gray-scale continuous frame static images. A second processing module is configured to determine a feature point loss connected domain in each frame static image based on the gray-scale continuous frame static images. The first determining module is configured to determine a feature point loss connected domain of the tth frame of static image and a feature point loss connected domain of the (t+1)th frame of static image based on the tth frame of static image and the (t+1)th frame of static image, wherein the tth frame of static image and the (t+1)th frame of static image are two adjacent frames of static image in the continuous frames of static image, and t is a positive integer greater than or equal to 1. The second determining module is configured to determine a first feature point loss connected domain and a second feature point loss connected domain based on the feature point loss connected domain of the tth frame of static image and the feature point loss connected domain of the (t+1)th frame of static image, wherein the first feature point loss connected domain and the second feature point loss connected domain are matched, the first feature point loss connected domain is the feature point loss connected domain of the tth frame of static image, and the second feature point loss connected domain is the feature point loss connected domain of the (t+1)th frame of static image. The third determining module is configured to determine a light influence degree of the first feature point loss connected domain based on a gray level change degree between the first feature point loss connected domain and the second feature point loss connected domain and a gray level change degree between a connected domain adjacent to the first feature point loss connected domain and a connected domain adjacent to the second feature point loss connected domain. The second processing module is configured to perform enhancement processing on a gray value of a pixel point in the first feature point loss connected domain by taking the light influence degree as an enhancement coefficient. The first determining module is configured to determine a feature point loss connected domain of the tth frame of static image and a feature point loss connected domain of the (t+1)th frame of static image based on the tth frame of static image and the (t+1)th frame of static image, wherein the tth frame of static image and the (t+1)th frame of static image are two adjacent frames of static image in the continuous frames of static image, and t is a positive integer greater than or equal to 1. The second determining module is configured to determine a first feature point loss connected domain and a second feature point loss connected domain based on the feature point loss connected domain of the tth frame of static image and the feature point loss connected domain of the (t+1)th frame of static image, wherein the first feature point loss connected domain and the second feature point loss connected domain are matched, the first feature point loss connected domain is the feature point loss connected domain of the tth frame of static image, and the second feature point loss connected domain is the feature point loss connected domain of the (t+1)th frame of static image. The third determining module is configured to determine a light influence degree of the first feature point loss connected domain based on a gray level change degree between the first feature point loss connected domain and the second feature point loss connected domain and a gray level change degree between a connected domain adjacent to the first feature point loss connected domain and a connected domain adjacent to the second feature point loss connected domain. The second processing module is configured to perform enhancement processing on a gray value of a pixel point in the first feature point loss connected domain by taking the light influence degree as an enhancement coefficient. The first determining module is configured to determine a feature point loss connected domain of the tth frame of static image and a feature point loss connected domain of the (t+1)th frame of static image based on the tth frame of static image and the (t+1)th frame of static image, wherein the tth frame of static image and the (t+1)th frame of static image are two adjacent frames of static image in the continuous frames of static image, and t is a positive integer greater than or equal to 1. The second determining module is configured to determine a first feature point loss connected domain and a second feature point loss connected domain based on the feature point loss connected domain of the tth frame of static image and the feature point loss connected domain of the (t+1)th frame of static image, wherein the first feature point loss connected domain and the second feature point loss connected domain are matched, the first feature point loss connected domain is the feature point loss connected domain of the tth frame of static image, and the second feature point loss connected domain is the feature point loss connected domain of the (t+1)th frame of static image. The third determining module is configured to determine a light influence degree of the first feature point loss connected domain based on a gray level change degree between the first feature point loss connected domain and the second feature point loss connected domain and a gray level change degree between a connected domain adjacent to the first feature point loss connected domain and a connected domain adjacent to the second feature point loss connected domain. The second processing module is configured to perform enhancement processing on a gray value of a pixel point in the first feature point loss connected domain by taking the light influence degree as an enhancement coefficient.

Images