A video-based river flow monitoring method, device, medium and equipment

Abstract

This invention discloses a video-based method, device, medium, and equipment for river flow monitoring, belonging to the fields of hydrological flow measurement and computer vision technology. Based on a continuous image frame sequence, this invention generates a region-optimized mask that matches the strongly reflective areas in the video image. The video image is then subjected to specular suppression and enhancement processing according to the region-optimized mask to obtain an enhanced image sequence. The region-optimized mask precisely defines the strongly reflective areas, and specular suppression is then applied to these areas to maximize the highlighting of the image texture, thereby generating an accurate river surface motion displacement field. This reduces the impact of complex lighting conditions on river video velocity measurement, improves the accuracy and stability of river surface velocity inversion and flow estimation, and solves the problem that strongly reflective areas in river video images can cause distortion in local optical flow field estimation and inaccurate calculation results.

Description

Technical Field

[0001] This invention relates to a video-based method, apparatus, medium, and equipment for monitoring river flow, belonging to the fields of hydrological flow measurement and computer vision technology. Background Technology

[0002] River flow velocity and discharge measurement are fundamental tasks in hydrological monitoring, water resource management, flood control and early warning, and the construction of smart water conservancy. Traditional contact flow measurement methods, such as rotor current meters and acoustic Doppler current profilers, usually require the deployment of sensors in the water body, which has problems such as high dependence on manual labor, complex deployment and maintenance, high equipment costs, and difficulty in implementation in dangerous scenarios such as floods and fast currents.

[0003] Currently, non-contact river velocity measurement methods based on video images are gradually becoming a research hotspot. These methods acquire continuous image sequences of the river surface using a fixed camera, and then utilize the motion information of ripples, floating objects, or surface textures between adjacent frames to invert the river's surface velocity. They offer advantages such as low cost, flexible deployment, suitability for continuous monitoring, and low risk to personnel safety. In particular, deep optical flow networks can densely estimate pixel-level displacements between adjacent image frames, providing an effective technical path for inverting river surface velocity.

[0004] However, water surface images in natural river scenes are often affected by complex lighting conditions such as direct sunlight, sky reflection, and specular reflection, resulting in highly reflective areas in the images. These areas typically exhibit high brightness, low color saturation, and a lack of local texture details, and their distribution changes continuously with lighting conditions and water surface fluctuations. Highly reflective areas significantly reduce the discernibility of local image textures, disrupting the brightness consistency and texture continuity upon which deep optical flow networks rely for feature extraction and motion matching. This leads to distortion in local optical flow field estimation, ultimately resulting in increased deviations in flow velocity calculations and unstable flow rate estimation results. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a video-based method, device, medium and equipment for river flow monitoring. This reduces the impact of complex lighting conditions on river video velocity measurement, improves the accuracy and stability of river surface velocity inversion and flow estimation, and solves the problem that strong reflective areas in river video images can cause local optical flow field estimation distortion and inaccurate calculation results.

[0006] To solve the above-mentioned technical problems, the present invention is implemented using the following technical solution: This invention provides a video-based method for monitoring river flow, comprising: Obtain a continuous sequence of image frames from the video of the river channel to be tested; Based on a continuous sequence of image frames, a region-optimized mask is generated that matches the highly reflective areas in the video image. The video image is subjected to specular suppression and enhancement processing based on the region optimization mask to obtain an enhanced image sequence; Based on the enhanced image sequence, a displacement field of the river surface motion in the river channel under test is generated; The total river flow is calculated based on the displacement field of the river surface and the pre-defined velocity measurement points.

[0007] Furthermore, the step of generating a region-optimized mask that matches the highly reflective areas in the video image based on a continuous sequence of image frames includes: Obtain the brightness and saturation information of each frame in a continuous image frame sequence; Based on the brightness and saturation information of each frame, the areas of strong reflection in the video image are determined; Generate an initial mask for the region corresponding to the highly reflective area; The initial mask for the region is optimized to obtain the optimized mask for the region.

[0008] Furthermore, the generation of the initial mask corresponding to the highly reflective area includes: expressing the initial mask corresponding to the highly reflective area using the following formula: ; in, Indicates the video image in coordinates Initial mask for the area; Indicates the video image in coordinates The brightness value at that location, Indicates the video image in coordinates The saturation value at that location; Indicates the brightness threshold. Indicates the saturation threshold; ; in, This represents the average brightness of the video image. The standard deviation of the brightness of a video image. Indicates the brightness adjustment coefficient; This represents the mean saturation value of the video image. The standard deviation of saturation in a video image. This represents the saturation adjustment coefficient.

[0009] Furthermore, the step of performing specular suppression enhancement processing on the video image based on the region-optimized mask to obtain an enhanced image sequence includes: enhancing the pixels at the region-optimized mask location in the video image using the following formula: ; in, Indicates the video image in coordinates Enhanced pixel values; Indicates the video image in coordinates Pixel value at that location, Indicates the specular suppression coefficient; Indicates the video image in coordinates Optimize the mask for the region at the coordinates When the pixel at that location is in a reflective area, When coordinates When the pixel at that location is in a non-reflective area, ; Indicates the video image in coordinates Local contrast enhancement within the pixel neighborhood. This represents the texture enhancement coefficient.

[0010] Furthermore, the step of generating the river surface motion displacement field of the river channel to be tested based on the enhanced image sequence includes: using a deep optical flow network to perform motion estimation on the enhanced image sequence, outputting a two-dimensional motion displacement vector of each pixel in the enhanced image sequence between adjacent frames, and forming the river surface motion displacement field of the river channel to be tested.

[0011] Furthermore, the calculation of the total river flow based on the river surface displacement field and pre-defined velocity measurement points includes: Projecting the displacement field of the river surface motion onto the world coordinate system, the river surface velocity at each velocity measurement point is obtained; Based on the surface velocity of the river, calculate the average vertical velocity and cross-sectional area between each velocity measuring point; the average vertical velocity between each velocity measuring point is calculated using the following formula: ; in, Indicates speed measurement point To the speed camera The average velocity along the vertical line between them; Indicates speed measurement point Vertical velocity at the point Indicates speed measurement point Vertical velocity at the point; , Indicates the surface velocity coefficient. Indicates speed measurement point Surface velocity at the location; The cross-sectional area of ​​the water passage between each speed measuring point is calculated using the following formula: ; in, Indicates speed measurement point Speed ​​camera The cross-sectional area of ​​the water passage between them; Indicates speed measurement point The water depth at the vertical position is Indicates speed measurement point The water depth at the vertical position is Indicates speed measurement point Speed ​​camera The distance between; Calculate the total river flow based on the average vertical velocity between each velocity measurement point and the cross-sectional area of ​​the water passage; The total flow of the river is calculated using the following formula: ; in, Indicates the total flow of the river. express Traffic, Indicates the number of speed measurement points. .

[0012] Furthermore, the projection of the river surface displacement field onto the world coordinate system to obtain the river surface velocity at each velocity measurement point includes: Obtain the pixel coordinate arrays of at least four marker points in the video image and their corresponding world coordinate arrays; Construct the transmission transformation matrix based on the pixel coordinate array of the marker points and their corresponding world coordinate array; Select a diagonal speed measurement line between at least four marker points, select multiple equidistant speed measurement points on the cross section of the speed measurement line, and calculate the pixel coordinate displacement of each speed measurement point between adjacent frames in the video image. The pixel coordinate displacement of each velocity measurement point is converted into world coordinates using a transmission transformation matrix to obtain the actual displacement between each velocity measurement point. The surface velocity of the river at each velocity measurement point is calculated based on the actual displacement between each velocity measurement point and the time interval between adjacent frames in the enhanced image sequence.

[0013] A second aspect of the present invention provides a video-based river flow monitoring device, comprising: The image frame sequence acquisition module is used to acquire a continuous image frame sequence of the video of the river channel under test. The mask generation module is used to generate a region-optimized mask that matches the highly reflective areas in the video image based on a continuous sequence of image frames. The image sequence enhancement module is used to perform highlight suppression enhancement processing on video images based on a region-optimized mask to obtain an enhanced image sequence. The water surface displacement calculation module is used to generate the river surface motion displacement field of the river channel under test based on the enhanced image sequence. The flow calculation module is used to calculate the total flow of a river based on the displacement field of the river surface and pre-defined velocity measurement points.

[0014] A third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the video-based river flow monitoring method described above.

[0015] The present invention also provides a computer device, comprising: Memory, used to store instructions; A processor is configured to execute the instructions, causing the device to perform the video-based river flow monitoring method described above.

[0016] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: 1. This invention generates a region-optimized mask that matches the strongly reflective areas in a video image based on a continuous image frame sequence; it then performs specular suppression and enhancement processing on the video image according to the region-optimized mask to obtain an enhanced image sequence; by precisely defining the strongly reflective areas through the region-optimized mask, and then performing specular suppression and enhancement processing on the strongly reflective areas, it enhances the details of the water surface texture while suppressing the reflective brightness, thereby generating an accurate river surface motion displacement field. This reduces the impact of complex lighting conditions on river video velocity measurement, improves the accuracy and stability of river surface velocity inversion and flow estimation, and solves the problem that strongly reflective areas in river video images can cause local optical flow field estimation distortion and inaccurate calculation results.

[0017] 2. This invention uses an adaptive threshold for brightness and saturation to generate an initial mask for the region corresponding to the highly reflective area, which can better adapt to different lighting environments and improve the stability of reflective detection.

[0018] 3. This invention constructs a transmission transformation matrix based on the pixel coordinate array of at least four marker points in a video image and their corresponding world coordinate array, selects velocity measurement points by relying on diagonal velocity measurement lines, and then uses the transmission transformation matrix to transform the pixel coordinates of the velocity measurement points to obtain the actual displacement of the river surface between adjacent frames in the enhanced image sequence, thus obtaining accurate river surface flow velocity. Attached Figure Description

[0019] Figure 1 This is a flowchart of a video-based river flow monitoring method provided in an embodiment of the present invention; Figure 2 This is a flowchart of obtaining the river surface velocity at each velocity measurement point provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of selecting marker points provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of a video-based river flow monitoring device provided in an embodiment of the present invention. Detailed Implementation

[0020] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention. Example

[0021] like Figure 1 As shown, the present invention provides a video-based method for monitoring river flow, comprising: Step 1: Obtain a continuous image frame sequence from the video of the river to be tested; specifically: perform frame segmentation on the input river video and generate a continuous image frame sequence at fixed time intervals.

[0022] Step 2: Based on a continuous sequence of image frames, generate a region-optimized mask that matches the highly reflective areas in the video image; specifically: Obtain the brightness and saturation information of each frame in a continuous image frame sequence; Based on the brightness and saturation information of each frame, the areas of strong reflection in the video image are determined; Generate an initial mask for the region corresponding to the highly reflective area; The initial mask for the region corresponding to the highly reflective area is represented by the following formula: ; in, Indicates the video image in coordinates Initial mask for the area; Indicates the video image in coordinates The brightness value at that location, Indicates the video image in coordinates The saturation value at that location; Indicates the brightness threshold. Indicates the saturation threshold; ; in, This represents the average brightness of the video image. The standard deviation of the brightness of a video image. Indicates the brightness adjustment coefficient; This represents the mean saturation value of the video image. The standard deviation of saturation in a video image. This represents the saturation adjustment coefficient; It should be noted that the brightness and saturation information of each frame of the image in this embodiment is obtained by converting each frame of the image from RGB color space to HSV color space and extracting the brightness component and saturation component respectively; wherein, the brightness component is used to describe the local brightness of the image and the saturation component is used to describe the color purity of the pixel. The initial mask for the region is optimized to obtain the optimized mask for the region. Specifically: First, adjacent reflective areas are connected and local gaps are repaired through closing operations. Then, discrete noise points are removed through opening operations. Subsequently, hole filling is performed to fill the voids inside the reflective areas. Finally, boundary smoothing is used to reduce the jagged edges of the mask, thereby obtaining a continuous and stable region-optimized mask.

[0023] Step 3: Perform specular suppression enhancement processing on the video image based on the region optimization mask to obtain an enhanced image sequence; use the following formula to enhance the pixels at the region optimization mask location in the video image: ; in, Indicates the video image in coordinates Enhanced pixel values; Indicates the video image in coordinates Pixel value at that location, Indicates the specular suppression coefficient. ; Indicates the video image in coordinates Optimize the mask for the region at the coordinates When the pixel at that location is in a reflective area, When coordinates When the pixel at that location is in a non-reflective area, ; Indicates the video image in coordinates Local contrast enhancement within the pixel neighborhood. This represents the texture enhancement factor; in this embodiment, it refers to the local contrast enhancement term. The standard deviation of grayscale within the pixel neighborhood or the gradient magnitude is used.

[0024] Step 4: Generate the displacement field of the river surface motion of the river channel to be tested based on the enhanced image sequence. Specifically, a deep optical flow network is used to estimate the motion of the enhanced image sequence and output the two-dimensional displacement vector of each pixel in the enhanced image sequence between adjacent frames to form the displacement field of the river surface motion of the river channel to be tested. In this process, the deep optical flow network performs feature encoding on adjacent frames, extracts multi-scale feature representations, then constructs pixel correlation relationships and gradually optimizes the displacement estimation through an iterative update module, and finally outputs the two-dimensional displacement vector of each pixel between adjacent frames.

[0025] Step 5, as follows Figure 2 As shown, the total river flow is calculated based on the displacement field of the river surface and pre-defined velocity measurement points; specifically: By projecting the displacement field of the river surface motion onto the world coordinate system, the river surface velocity at each velocity measurement point is obtained, including: Obtain the pixel coordinate arrays of at least four marker points in the video image and their corresponding world coordinate arrays; Construct a transmission transformation matrix based on the pixel coordinate array of the marker points and their corresponding world coordinate array; such as Figure 3 As shown, in this embodiment, the river channel in the video image is used as the boundary, and four marker points A, B, C, and D are selected on the riverbank. The pixel coordinate array of marker points A, B, C, and D is represented as follows: ; The corresponding world coordinate system coordinate array is represented as follows: ; The mapping relationship from the world coordinate system to the pixel coordinate system is established based on the perspective projection transformation model, and the transmission transformation matrix is ​​calculated by the least squares method. The mapping relationship is represented as follows: ; in, Represents world coordinates, Represents pixel coordinates, Represents the transmission transformation matrix; The calculated transmission transformation matrix is: ; in, To indicate scientific notation, , ;For example, ,but , ; ,but , .

[0026] Select a diagonal speed measurement line between at least four marker points, and select multiple equidistant speed measurement points on the cross section of the speed measurement line. Calculate the pixel coordinates of each speed measurement point, including: Using AD as the speed measuring line, and taking D as the starting point on the cross section of AD, select speed measuring points at distances of 2m, 3m, 4m, 5m, 6m, 7m, 8m, 9m, 10m, and 11m from the starting point D. The coordinates of these speed measuring points in the world coordinate system are then represented as follows: , , , , , , , , , ; The pixel coordinates of each velocity measurement point are calculated using the transmission transformation matrix: , , , , , , , , , ; The pixel coordinates of each speed measurement point are input into the depth optical flow network to calculate the pixel coordinate displacement between each speed measurement point in adjacent frames of the video image. Then, the pixel coordinate displacement between each speed measurement point is converted into the actual displacement between each speed measurement point using the transmission transformation matrix. The surface velocity of the river at each velocity measurement point is calculated based on the actual displacement between each velocity measurement point and the time interval between adjacent frames. Based on the surface velocity of the river, calculate the average vertical velocity and cross-sectional area of ​​the water flow between each velocity measurement point; The vertical average velocity between each velocity measuring point is calculated using the following formula: ; in, Indicates speed measurement point To the speed camera The average velocity along the vertical line between them; Indicates speed measurement point Vertical velocity at the point Indicates speed measurement point Vertical velocity at the point; , Indicates the surface velocity coefficient. Indicates speed measurement point Surface velocity at the location; The cross-sectional area of ​​the water passage between each speed measuring point is calculated using the following formula: ; in, Indicates speed measurement point Speed ​​camera The cross-sectional area of ​​the water passage between them; Indicates speed measurement point The water depth at the vertical position is Indicates speed measurement point The water depth at the vertical position is Indicates speed measurement point Speed ​​camera The distance between; Calculate the total river flow based on the average vertical velocity between each velocity measurement point and the cross-sectional area of ​​the water passage; The total flow of the river is calculated using the following formula: ; in, Indicates the total flow of the river. express Traffic, Indicates the number of speed measurement points. .

[0027] Example 2 like Figure 4 As shown, this embodiment provides a video-based river flow monitoring device, including: The image frame sequence acquisition module is used to acquire a continuous image frame sequence of the video of the river channel under test. The mask generation module is used to generate a region-optimized mask that matches the highly reflective areas in the video image based on a continuous sequence of image frames. The image sequence enhancement module is used to perform highlight suppression enhancement processing on video images based on a region-optimized mask to obtain an enhanced image sequence. The water surface displacement calculation module is used to generate the river surface motion displacement field of the river channel under test based on the enhanced image sequence. The flow calculation module is used to calculate the total flow of a river based on the displacement field of the river surface and pre-defined velocity measurement points.

[0028] The specific functions of each module described above are explained in the relevant content of the method in Embodiment 1, and will not be repeated here.

[0029] Example 3 This embodiment provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the following method steps: Obtain a continuous sequence of image frames from the video of the river channel to be tested; Based on a continuous sequence of image frames, a region-optimized mask is generated that matches the highly reflective areas in the video image. The video image is subjected to specular suppression and enhancement processing based on the region optimization mask to obtain an enhanced image sequence; Based on the enhanced image sequence, a displacement field of the river surface motion in the river channel under test is generated; The total river flow is calculated based on the displacement field of the river surface and the pre-defined velocity measurement points.

[0030] Example 4 This embodiment also provides a computer device, including: Memory, used to store instructions; A processor is configured to execute the instructions, causing the device to perform the following method steps: Obtain a continuous sequence of image frames from the video of the river channel to be tested; Based on a continuous sequence of image frames, a region-optimized mask is generated that matches the highly reflective areas in the video image. The video image is subjected to specular suppression and enhancement processing based on the region optimization mask to obtain an enhanced image sequence; Based on the enhanced image sequence, a displacement field of the river surface motion in the river channel under test is generated; The total river flow is calculated based on the displacement field of the river surface and the pre-defined velocity measurement points.

[0031] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0032] This application is described with reference to flowchart illustrations of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each step in the flowchart can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing device, generate instructions for implementing the process. Figure 1 One or more processes or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0033] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 The function specified in one or more processes.

[0034] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 Steps of a specified function in one or more processes.

[0035] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A method for river flow monitoring based on video, characterized in that, include: Obtain a continuous sequence of image frames from the video of the river channel to be tested; Based on a continuous sequence of image frames, a region-optimized mask is generated that matches the highly reflective areas in the video image. The video image is subjected to specular suppression and enhancement processing based on the region optimization mask to obtain an enhanced image sequence; Based on the enhanced image sequence, a displacement field of the river surface motion in the river channel under test is generated; The total river flow is calculated based on the displacement field of the river surface and the pre-defined velocity measurement points.

2. The video-based river flow monitoring method according to claim 1, wherein, The process of generating a region-optimized mask that matches strongly reflective areas in a video image based on a continuous sequence of image frames includes: Obtain the brightness and saturation information of each frame in a continuous image frame sequence; Based on the brightness and saturation information of each frame, the areas of strong reflection in the video image are determined; Generate an initial mask for the region corresponding to the highly reflective area; The initial mask for the region is optimized to obtain the optimized mask for the region.

3. The video-based river flow monitoring method of claim 2, wherein, The generation of the initial mask corresponding to the highly reflective area includes: expressing the initial mask corresponding to the highly reflective area using the following formula: ； in, Indicates the video image in coordinates Initial mask for the area; Indicates the video image in coordinates The brightness value at that location, Indicates the video image in coordinates The saturation value at that location; Indicates the brightness threshold. Indicates the saturation threshold; ； in, This represents the average brightness of the video image. The standard deviation of the brightness of a video image. Indicates the brightness adjustment coefficient; This represents the mean saturation value of the video image. The standard deviation of saturation in a video image. This represents the saturation adjustment coefficient.

4. The video-based river flow monitoring method according to claim 1, characterized in that, The process of performing highlight suppression and enhancement processing on the video image based on the region optimization mask to obtain an enhanced image sequence includes: enhancing the pixels at the region optimization mask location in the video image using the following formula: ; in, Indicates the video image in coordinates Enhanced pixel values; Indicates the video image in coordinates Pixel value at that location, Indicates the specular suppression coefficient; Indicates the video image in coordinates Optimize the mask for the region at the coordinates When the pixel at that location is in a reflective area, When coordinates When the pixel at that location is in a non-reflective area, ; Indicates the video image in coordinates Local contrast enhancement within the pixel neighborhood This represents the texture enhancement coefficient.

5. The video-based river flow monitoring method according to claim 1, characterized in that, The step of generating the river surface motion displacement field of the river channel to be tested based on the enhanced image sequence includes: using a deep optical flow network to perform motion estimation on the enhanced image sequence, outputting a two-dimensional motion displacement vector of each pixel in the enhanced image sequence between adjacent frames, and forming the river surface motion displacement field of the river channel to be tested.

6. The video-based river flow monitoring method according to claim 1, characterized in that, The calculation of the total river flow based on the river surface displacement field and pre-defined velocity measurement points includes: Projecting the displacement field of the river surface motion onto the world coordinate system, the river surface velocity at each velocity measurement point is obtained; Based on the surface velocity of the river, calculate the average vertical velocity and cross-sectional area between each velocity measuring point; the average vertical velocity between each velocity measuring point is calculated using the following formula: ; in, Indicates speed measurement point To the speed camera The average velocity along the vertical line between; Indicates speed measurement point The vertical flow velocity at that point Indicates speed measurement point Vertical velocity at the point; , Indicates the surface velocity coefficient. Indicates speed measurement point Surface velocity at the location; The cross-sectional area of ​​the water passage between each speed measuring point is calculated using the following formula: ; in, Indicates speed measurement point Speed ​​camera The cross-sectional area of ​​the water passage between them; Indicates speed measurement point The water depth at the vertical position is Indicates speed measurement point The water depth at the vertical position is Indicates speed measurement point Speed ​​camera The distance between; Calculate the total river flow based on the average vertical velocity between each velocity measurement point and the cross-sectional area of ​​the water passage; The total flow of the river is calculated using the following formula: ; in, Indicates the total flow of the river. express Traffic, Indicates the number of speed measurement points. .

7. The video-based river flow monitoring method according to claim 6, characterized in that, The process of projecting the displacement field of the river surface motion onto the world coordinate system to obtain the river surface velocity at each velocity measurement point includes: Obtain the pixel coordinate arrays of at least four marker points in the video image and their corresponding world coordinate arrays; Construct the transmission transformation matrix based on the pixel coordinate array of the marker points and their corresponding world coordinate array; Select a diagonal speed measurement line between at least four marker points, select multiple equidistant speed measurement points on the cross section of the speed measurement line, and calculate the pixel coordinate displacement of each speed measurement point between adjacent frames in the video image. The pixel coordinate displacement of each velocity measurement point is converted into world coordinates using a transmission transformation matrix to obtain the actual displacement between each velocity measurement point. The surface velocity of the river at each velocity measurement point is calculated based on the actual displacement between each velocity measurement point and the time interval between adjacent frames in the enhanced image sequence.

8. A video-based river flow monitoring device, characterized in that, include: The image frame sequence acquisition module is used to acquire a continuous image frame sequence of the video of the river channel under test. The mask generation module is used to generate a region-optimized mask that matches the highly reflective areas in the video image based on a continuous sequence of image frames. The image sequence enhancement module is used to perform highlight suppression enhancement processing on video images based on a region-optimized mask to obtain an enhanced image sequence. The water surface displacement calculation module is used to generate the river surface motion displacement field of the river channel under test based on the enhanced image sequence. The flow calculation module is used to calculate the total flow of a river based on the displacement field of the river surface and pre-defined velocity measurement points.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the video-based river flow monitoring method as described in any one of claims 1 to 7.

10. A computer device, characterized in that, include: Memory, used to store instructions; A processor for executing the instructions, causing the device to perform the video-based river flow monitoring method as described in any one of claims 1 to 7.

Images