An intelligent monitoring method for the coverage area of submerged plants

Abstract

The present invention provides an intelligent monitoring method for the coverage area of submerged plants. While obtaining the projection image of the submerged plants, the present invention obtains the turbidity data corresponding to the projection image; and based on the turbidity data, different image stretching methods are used to stretch the projection image to enhance the contrast of the projection image and obtain an enhanced image. Then, edge detection and optimization are performed on the enhanced image to obtain an edge-optimized image. Finally, the coverage area of the submerged plants is obtained according to the edge-optimized image. Since in the process of monitoring the coverage area of submerged plants in the present invention, based on turbidity, the contrast before edge detection is enhanced by the image stretching method, the clarity of the captured image of the submerged plants is improved, and thus the accuracy of detecting the coverage area of the submerged plants is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image recognition, and in particular to an intelligent monitoring method for the coverage area of submerged plants. Background Art

[0002] Submerged plants have well-developed intercellular spaces, a strong ventilation tissue, and functions such as purifying water quality, providing habitats for aquatic animals, improving substrate soil, and inhibiting the growth of algae. They play an important role in the removal of pollutants in water bodies and the maintenance of clean water quality. Therefore, submerged plants are an important part of the monitoring of water ecology and aquatic plants.

[0003] In ecology, coverage is an important concept for measuring the growth status of vegetation. By using underwater camera technology, the ratio of the projected area of submerged plants to the bottom area of the water can be used to monitor the growth status of submerged plants. However, due to the presence of suspended solids in the water, the phenomenon of image degradation is serious, such as color fading, low contrast, and blurred details. Therefore, there are currently problems with low clarity of the captured images of submerged plants and inaccurate detection of the coverage area of submerged plants. Summary of the Invention

[0004] The present invention provides an intelligent monitoring method for the coverage area of submerged plants, which can improve the clarity of the captured images of submerged plants and further improve the accuracy of detecting the coverage area of submerged plants.

[0005] In a first aspect, the present invention provides an intelligent monitoring method for the coverage area of submerged plants, including: obtaining a projected image of the submerged plants and the turbidity data corresponding to the projected image; determining the turbidity level corresponding to the projected image based on the turbidity data corresponding to the projected image; performing image stretching on the projected image based on the turbidity level corresponding to the projected image to obtain an enhanced image; wherein, the image stretching methods corresponding to different turbidity levels are different; performing edge detection and optimization on the enhanced image to obtain an edge-optimized image; and determining the coverage area of the submerged plants based on the edge-optimized image.

[0006] In a possible implementation manner, before performing image stretching on the projected image based on the turbidity level corresponding to the projected image to obtain an enhanced image, it further includes: extracting the image blur feature and image structure feature of the projected image; dividing the projected image into multiple regions based on the image blur feature and image structure feature; wherein, the clarity of each region of the projected image is different; performing attention analysis on the clarity of the projected image to obtain an attention matrix; the weight of the clear pixels in the attention matrix is higher than the weight of the blurred pixels; and reconstructing the projected image based on the multiple regions of the projected image and the attention matrix to implement preprocessing of the projected image.

[0007] In a possible implementation, based on the turbidity level corresponding to the projection image, the projection image is stretched to obtain an enhanced image, including: determining a stretching scheme for the projection image based on the turbidity level corresponding to the projection image; the stretching scheme includes one of the following: histogram stretching, Retinex stretching, logarithmic transformation stretching, exponential transformation stretching, maximum-minimum stretching, and piecewise linear stretching; stretching the projection image based on the stretching scheme of the projection image to obtain an enhanced image.

[0008] In a possible implementation, determining a stretching scheme for the projection image based on the turbidity level corresponding to the projection image includes: if the turbidity level corresponding to the projection image is low turbidity, the stretching scheme for the projection image is histogram stretching or Retinex stretching; correspondingly, stretching the projection image based on the stretching scheme of the projection image to obtain an enhanced image includes: performing a spatial transformation on the projection image to convert the RGB space to the HSV space to obtain the H component, S component, and V component of the HSV space; for any component in the HSV space, calculating the total number of pixels of this component and the gray values of each pixel in this component; based on the total number of pixels of this component and the gray values of each pixel in this component, performing level division to obtain multiple gray levels; calculating the cumulative distribution values corresponding to each gray level; performing histogram equalization on this component based on the cumulative distribution values corresponding to each gray level to obtain the stretched component corresponding to this component; based on the stretched components corresponding to each component in the HSV space, performing a spatial transformation to convert the HSV space to the RGB space to obtain an enhanced image.

[0009] In a possible implementation, performing edge detection and optimization on the enhanced image to obtain an edge-optimized image includes: performing a block operation on the enhanced image to obtain multiple block images; for any block image, using the Canny operator to perform edge detection on the submerged plants in this block image to determine the block image with edge contours; performing foreground and background differentiation on each block image with edge contours to determine the foreground area and background area in each block image; based on the edge contours, foreground areas, and background areas of each block image, converting each block image into each binary image; performing inter-block merging based on each binary image to obtain an edge-optimized image.

[0010] In a possible implementation, for any sub-block image, the Canny operator is used to detect the edges of the submerged plants in the sub-block image to determine the sub-block image with edge contours, including: performing Gaussian smoothing on the any sub-block image to remove high-frequency noise in the any sub-block image, obtaining a Gaussian smoothed image; using the soble horizontal operator and the soble vertical operator to perform convolution processing on the Gaussian smoothed image to obtain the edge intensity and direction of each pixel point in the Gaussian smoothed image; based on the edge intensity and direction of each pixel point in the Gaussian smoothed image, using a double-threshold algorithm to determine the first edge and the second edge in the Gaussian smoothed image, where the first edge is the edge determined by the lower threshold in the double threshold, and the second edge is the edge determined by the higher threshold in the double threshold; in the Gaussian smoothed image, based on the second edge, select new edge points between the first edge and the second edge to complete the breakpoints in the second edge, forming a closed edge contour, and obtaining the sub-block image with edge contours.

[0011] In a possible implementation, for each sub-block image with edge contours, foreground and background discrimination is performed to determine the foreground area and the background area in each sub-block image, including: for any sub-block image with edge contours, based on the edge contours, calculate the average gray value of the area inside the contour and the area outside the contour respectively; based on the average gray value of the area inside the contour and the area outside the contour, perform gray generalization on each pixel point in the sub-block image to obtain the generalized sub-block image; based on the gray value of each pixel point in the generalized sub-block image, determine the foreground area and the background area of the sub-block image.

[0012] In a second aspect, an intelligent monitoring device for the coverage area of submerged plants provided by an embodiment of the present invention includes a communication module and a processing module; the communication module is used to obtain the projection image of the submerged plants and the turbidity data corresponding to the projection image; the processing module is used to determine the turbidity level corresponding to the projection image based on the turbidity data corresponding to the projection image; perform image stretching on the projection image based on the turbidity level corresponding to the projection image to obtain an enhanced image; where the image stretching methods corresponding to different turbidity levels are different; perform edge detection and optimization on the enhanced image to obtain an edge-optimized image; and determine the coverage area of the submerged plants based on the edge-optimized image.

[0013] In a possible implementation, the processing module is further used to extract the image blur feature and the image structure feature of the projection image; divide the projection image into multiple regions based on the image blur feature and the image structure feature; where the clarity of each region of the projection image is different; perform attention analysis on the clarity of the projection image to obtain an attention matrix; the weight of the clear pixels in the attention matrix is higher than the weight of the blurred pixels; and reconstruct the projection image based on the multiple regions of the projection image and the attention matrix to implement preprocessing of the projection image.

[0014] In a possible implementation manner, the processing module is specifically configured to determine a stretching scheme for the projection image based on the turbidity level corresponding to the projection image; the stretching scheme includes one of the following: histogram stretching, Retinex stretching, logarithmic transformation stretching, exponential transformation stretching, maximum-minimum stretching, and piecewise linear stretching; and perform image stretching on the projection image based on the stretching scheme of the projection image to obtain an enhanced image.

[0015] In a possible implementation manner, the processing module is specifically configured to, if the turbidity level corresponding to the projection image is low turbidity, the stretching scheme of the projection image is histogram stretching or Retinex stretching; correspondingly, the processing module is specifically configured to perform a spatial transformation on the projection image, convert the RGB space to the HSV space, and obtain the H component, S component, and V component of the HSV space; for any component of the HSV space, calculate the total number of pixels of the component and the gray values of each pixel in the component; perform level division based on the total number of pixels of the component and the gray values of each pixel in the component to obtain multiple gray levels; calculate the cumulative distribution values corresponding to each gray level; perform histogram equalization on the component based on the cumulative distribution values corresponding to each gray level to obtain the stretched component corresponding to the component; and perform a spatial transformation based on the stretched components corresponding to each component in the HSV space, convert the HSV space to the RGB space, and obtain an enhanced image.

[0016] In a possible implementation manner, the processing module is specifically configured to perform a block operation on the enhanced image to obtain multiple block images; for any block image, use the Canny operator to perform edge detection on the submerged plants in the block image to determine the block image with edge contours; perform foreground and background distinction on each block image with edge contours to determine the foreground area and background area in each block image; convert each block image into each binary image based on the edge contours, foreground area, and background area of each block image; and perform inter-block merging based on each binary image to obtain an edge-optimized image.

[0017] In a possible implementation, the processing module is specifically configured to perform Gaussian smoothing on any of the segmented images to remove high-frequency noise in the segmented image and obtain a Gaussian-smoothed image; perform convolution processing on the Gaussian-smoothed image by using a Sobel horizontal operator and a Sobel vertical operator to obtain the edge intensity and direction of each pixel point in the Gaussian-smoothed image; based on the edge intensity and direction of each pixel point in the Gaussian-smoothed image, adopt a double-threshold algorithm to determine the first edge and the second edge in the Gaussian-smoothed image, where the first edge is the edge determined by using the lower threshold in the double thresholds, and the second edge is the edge determined by using the higher threshold in the double thresholds; in the Gaussian-smoothed image, based on the second edge, select new edge points between the first edge and the second edge to complete the breakpoints in the second edge and form a closed edge contour, thereby obtaining a segmented image with an edge contour.

[0018] In a possible implementation, the processing module is specifically configured to, for any segmented image with an edge contour, calculate the average gray value of the area inside the contour and the area outside the contour respectively based on the edge contour; perform gray generalization on each pixel point in the segmented image based on the average gray value of the area inside the contour and the area outside the contour to obtain a generalized segmented image; determine the foreground area and the background area of the segmented image based on the gray value of each pixel point in the generalized segmented image.

[0019] In a third aspect, an embodiment of the present invention provides an electronic device, where the electronic device includes a memory and a processor, the memory stores a computer program, and the processor is configured to call and run the computer program stored in the memory to execute the steps of the method described in the first aspect and any possible implementation manner in the first aspect as above.

[0020] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and is characterized in that when the computer program is executed by a processor, it implements the steps of the method described in the first aspect and any possible implementation manner in the first aspect as above.

[0021] The present invention provides an intelligent monitoring method for the coverage area of submerged plants. While obtaining the projection image of the submerged plants, the present invention obtains the turbidity data corresponding to the projection image; and based on the turbidity data, performs image stretching on the projection image by using different image stretching methods to enhance the contrast of the projection image and obtain an enhanced image. Then, edge detection and optimization are performed on the enhanced image to obtain an edge-optimized image, and finally, the coverage area of the sleeping plants is obtained according to the edge-optimized image. Since in the process of monitoring the coverage area of submerged plants, the present invention enhances the contrast before edge detection by means of image stretching based on turbidity, improves the clarity of the captured image of submerged plants, and further improves the accuracy of detecting the coverage area of submerged plants. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0023] Figure 1 It is a schematic flowchart of an intelligent monitoring method for the coverage area of submerged plants provided by an embodiment of the present invention;

[0024] Figure 2 It is a schematic structural diagram of an intelligent monitoring device for the coverage area of submerged plants provided by an embodiment of the present invention;

[0025] Figure 3 It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] In the following description, specific details such as specific system structures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present invention. However, those skilled in the art should clearly understand that the present invention can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present invention.

[0027] In the description of the present invention, unless otherwise specified, " / " means "or". For example, A / B can represent A or B. The "and / or" herein is only a description of the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, "at least one" and "multiple" mean two or more. The words such as "first" and "second" do not limit the quantity and execution order, and the words such as "first" and "second" do not necessarily limit being different.

[0028] In the embodiments of the present application, words such as "exemplary" or "for example" are used to represent examples, illustrations, or explanations. Any embodiment or design solution described as "exemplary" or "for example" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Exactly speaking, using words such as "exemplary" or "for example" aims to present relevant concepts in a specific manner for easy understanding.

[0029] In addition, the terms "including" and "having" and any variations thereof mentioned in the description of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or modules is not limited to the listed steps or modules, but may optionally further include other unlisted steps or modules, or may optionally further include other steps or modules inherent to these processes, methods, products, or devices.

[0030] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will be described through specific embodiments in conjunction with the accompanying drawings of the present invention.

[0031] Figure 1 The figure is a schematic flowchart of an intelligent monitoring method for the coverage area of submerged plants provided by an embodiment of the present invention. The execution subject of this method is an intelligent monitoring device for the coverage area of submerged plants. This method includes steps S101 - S105.

[0032] S101. Obtain the projection image of the submerged plants and the turbidity data corresponding to the projection image.

[0033] As a possible implementation, the present invention can collect video data of submerged plants in real time; and use a turbidity sensor to collect turbidity data at each moment in real time.

[0034] Among them, the projection image is an image at any moment in the video data. The turbidity data corresponding to the projection image is the turbidity data at the moment corresponding to the projection image.

[0035] Optionally, before processing the projection image, the intelligent monitoring device provided by the embodiment of the present invention for the intelligent monitoring method of the coverage area of submerged plants can also preprocess the projection image. Exemplarily, the intelligent monitoring device can implement the preprocessing of the projection image through steps A1 - A5.

[0036] A1. Extract the image blur feature and image structure feature of the projection image.

[0037] A2. Divide the projection image into multiple regions based on the image blur feature and image structure feature.

[0038] Among them, the clarity of each region of the projection image is different.

[0039] A3. Perform attention analysis on the clarity of the projection image to obtain an attention matrix.

[0040] In some embodiments, the weight of clear pixels in the attention matrix is higher than the weight of blurred pixels.

[0041] A4. Reconstruct the projection image based on the multiple regions of the projection image and the attention matrix to implement the preprocessing of the projection image.

[0042] It should be noted that during the data acquisition process of the present invention, due to changing factors such as water environment flow, there is an aliasing phenomenon in the spatial position information of the leaves of submerged plants. The longer the exposure time, the more severe the blurring of the vegetation edge in the image. Therefore, in the data enhancement preprocessing stage, first, the STDANet deformable attention-based video deblurring network can be used. A lightweight motion estimation branch is used to estimate the rough optical flow, and at the same time, the spatio-temporal deformable attention module can assign higher weights to the clear regions in the picture, thereby effectively extracting the clear pixel information in the video segment to reconstruct and restore the blurred pixels, effectively solving the blurring problem caused by the movement during the underwater submerged vegetation photography.

[0043] Exemplarily, the present invention uses a motion deblurring preprocessing algorithm to enhance the edge blurring of submerged vegetation caused by water flow. For example, three consecutive frames of images are selected as the input, and the STDANet network method is used. The pixel-level blurring degree of the consecutive image frames is used to extract the information of the clear pixels in the consecutive frames, and the blurred intermediate frame image is restored to complete the deblurring reconstruction of the submerged vegetation image in the intermediate frame. Given three consecutive video frames Bi = {Bk}k = i - 1, i, i + 1 as the input, first, the consecutive video frames are input into the feature extractor to obtain the features of the consecutive video frames. Then, the motion estimation branch estimates the rough optical flow between the consecutive frames according to the extracted features. The STDA module extracts the clear pixel information of the multi-frame features under the guidance of the estimated rough optical flow to obtain the reconstructed feature Fif. Finally, the reconstruction network restores Fif to the RGB image space to complete the reconstruction.

[0044] S102. Determine the turbidity level corresponding to the projection image based on the turbidity data corresponding to the projection image.

[0045] As a possible implementation manner, the present invention can query the preselected turbidity level mapping table based on the turbidity data corresponding to the projection image to obtain the turbidity level corresponding to the projection image.

[0046] S103. Perform image stretching on the projection image based on the turbidity level corresponding to the projection image to obtain an enhanced image.

[0047] Among them, the image stretching methods corresponding to different turbidity levels are different.

[0048] As a possible implementation manner, the intelligent monitoring device can obtain the enhanced image through steps S1031 - S1032.

[0049] S1031. Determine the stretching scheme of the projection image based on the turbidity level corresponding to the projection image.

[0050] In some embodiments, the stretching scheme includes one of the following: histogram stretching, Retinex stretching, logarithmic transformation stretching, exponential transformation stretching, maximum-minimum stretching, and piecewise linear stretching.

[0051] Exemplarily, if the turbidity level corresponding to the projected image is low turbidity, the stretching scheme for the projected image is determined to be histogram stretching or Retinex stretching;

[0052] In another example, if the turbidity level corresponding to the projected image is medium turbidity, the stretching scheme for the projected image is determined to be logarithmic transformation stretching or exponential transformation stretching.

[0053] In another example, if the turbidity level corresponding to the projected image is high turbidity, the stretching scheme for the projected image is determined to be maximum-minimum stretching or piecewise linear stretching.

[0054] S1032. Based on the stretching scheme of the projected image, perform image stretching on the projected image to obtain an enhanced image.

[0055] Exemplarily, when the stretching scheme is histogram stretching, the intelligent monitoring device can perform histogram stretching on the projected image through steps B11 - B16 to obtain an enhanced image.

[0056] B11. Perform a spatial transformation on the projected image to convert the RGB space to the HSV space, obtaining the H component, S component, and V component in the HSV space.

[0057] B12. For any component in the HSV space, calculate the total number of pixels of this component and the gray values of each pixel in this component.

[0058] Exemplarily, taking the V component as an example, the present invention can calculate the total number of pixels in the V component and the gray values of each pixel in the V component.

[0059] B13. Based on the total number of pixels of this component and the gray values of each pixel in this component, perform level division to obtain multiple gray levels.

[0060] Exemplarily, taking the V component as an example, the present invention can count the proportion of the pixel amount in different gray levels in the total pixel amount in the V component to determine the probability of each gray level appearing.

[0061] Exemplarily, the present invention can determine each gray level based on the following formula.

[0062] p x (i) = p(x = i) = n i / n;

[0063] Wherein, p x (i) is the probability corresponding to the gray level with a gray value of i, ni The number of pixels with a gray value of i; n is the total number of pixels in the V component.

[0064] It should be noted that 0 ≤ i < L, where L is the gray value corresponding to the highest gray level. For example, for the V - component channel image, an equal - interval quantization level can be established with a spacing of 0.005. Then the number of gray levels is 1 / 0.005 = 200 levels.

[0065] In this way, the present invention can quantize and classify all gray values in the V component, determine multiple gray levels, and the probabilities corresponding to each gray level.

[0066] B14. Calculate the cumulative distribution value corresponding to each gray level.

[0067] Exemplarily, the present invention can determine the cumulative distribution value corresponding to each gray level based on the following formula.

[0068]

[0069] Among them, cdf(i) is the cumulative distribution value corresponding to the gray level with a gray value of i, and p x (j) is the probability corresponding to the gray level with a gray value of j.

[0070] B15. Based on the cumulative distribution values corresponding to each gray level, perform histogram equalization on this component to obtain the stretched component corresponding to this component.

[0071] Exemplarily, the present invention can achieve histogram equalization of this component based on the following formula.

[0072]

[0073] Among them, h(i) is the gray value of the pixel point with a gray value of i after stretching, cdf(i) is the cumulative distribution value corresponding to the gray level with a gray value of i, and cdf min is the minimum value among the cumulative distribution values corresponding to each gray level, M is the number of pixels in the row direction of this component image, N is the number of pixels in the column direction of this component image, and L is the gray value corresponding to the highest gray level.

[0074] B16. Based on the stretched components corresponding to each component in the HSV space, perform a space conversion to convert the HSV space to the RGB space to obtain an enhanced image.

[0075] Exemplarily, when the stretching scheme is Retinex stretching, the intelligent monitoring device can achieve Retinex stretching of the projection image through steps B21 - B26 to obtain an enhanced image.

[0076] B21. Perform a spatial transformation on the projected image, convert the RGB space to the HSV space, and obtain the H component, S component, and V component in the HSV space.

[0077] B22. For any component in the HSV space, construct different scale parameters based on the Gaussian surround function.

[0078] Exemplarily, the Gaussian surround function can be expressed by the following formula.

[0079]

[0080] Where K is the normalization constant, G(x, y) is the Gaussian surround function, x is the number of pixels in the row direction of the component image, y is the number of pixels in the column direction of the component image, and σ is the scale parameter.

[0081] Among them, each scale parameter satisfies ∫∫G(x, y)dxdy = 1.

[0082] B23. Use the Gaussian surround function to perform convolution filtering on the component to obtain the illumination component;

[0083] L(x, y) = I(x, y) * G(x, y);

[0084] Where L(x, y) is the illumination component, I(x, y) is the original input image, and G(x, y) is the Gaussian surround function.

[0085] B24. Perform image stretching on the component based on the illumination component to obtain the logarithmic component corresponding to the component.

[0086] Exemplarily, the present invention can determine the stretched component corresponding to the component based on the following formula.

[0087]

[0088] Where r(x, y) is the logarithmic domain reflection component, R(x, y) is the real domain reflection component, I(x, y) is the original input image, and L(x, y) is the illumination component.

[0089] B25. Calculate the maximum value and minimum value of the logarithmic component, and perform linearization based on the maximum value and minimum value of the logarithmic component to obtain the stretched component corresponding to the component.

[0090] Exemplarily, the present invention can obtain the stretched component corresponding to the component based on the following formula.

[0091]

[0092] Where R d(x, y) is the reflected component after linear stretching, that is, the stretching component corresponding to this component. Value is the logarithmic domain reflected component value of pixel (x, y), Max is the maximum value of the logarithmic domain reflected component in the image of this component; Min is the minimum value of the logarithmic domain reflected component in the image of this component.

[0093] B26. Based on the stretching components corresponding to each component in the HSV space, perform a space conversion to convert the HSV space to the RGB space to obtain an enhanced image.

[0094] Exemplarily, when the stretching scheme is logarithmic change stretching, the intelligent monitoring device can perform logarithmic change stretching on the projection image through steps B31 - B36 to obtain an enhanced image.

[0095] B31. Perform a space conversion on the projection image to convert the RGB space to the HSV space to obtain the H component, S component, and V component of the HSV space.

[0096] B32. For any component in the HSV space, use the logarithmic change formula to perform image stretching processing on this component to obtain the stretching component corresponding to this component.

[0097] Exemplarily, the logarithmic change formula is as follows.

[0098] y = λlog(1 + x)

[0099] Where x is the gray level of each pixel point in the image of this component channel, λ is an adjustment constant used to adjust the gray level value after image transformation, and y is the output value after linear transformation, that is, the gray level of each pixel point in the stretching component corresponding to this component.

[0100] B33. Based on the stretching components corresponding to each component in the HSV space, perform a space conversion to convert the HSV space to the RGB space to obtain an enhanced image.

[0101] Exemplarily, when the stretching scheme is exponential change stretching, the intelligent monitoring device can perform exponential change stretching on the projection image through steps B41 - B46 to obtain an enhanced image.

[0102] B41. Perform a space conversion on the projection image to convert the RGB space to the HSV space to obtain the H component, S component, and V component of the HSV space.

[0103] B42. For any component in the HSV space, use the exponential change formula to perform image stretching processing on this component to obtain the stretching component corresponding to this component.

[0104] Exemplarily, the exponential change formula is as follows.

[0105] y = C * Rr ;

[0106] Where y is the output after exponential transformation, that is, the grayscale of each pixel point in the stretching component corresponding to this component, R is the grayscale of each pixel point in the image of this component channel, C is the exponential change coefficient, and γ is the exponent.

[0107] B43. Based on the stretching components corresponding to each component in the HSV space, perform a space conversion to convert the HSV space into the RGB space to obtain an enhanced image.

[0108] Exemplarily, when the stretching scheme is maximum-minimum stretching, the intelligent monitoring device can perform maximum-minimum stretching on the projected image through steps B51 - B56 to obtain an enhanced image.

[0109] B51. Perform a space conversion on the projected image to convert the RGB space into the HSV space to obtain the H component, S component, and V component of the HSV space.

[0110] B52. For any component in the HSV space, use the following formula to perform image stretching processing on this component to obtain the stretching component corresponding to this component.

[0111]

[0112] Where I (x,y) is the grayscale of each pixel point in the image of this component channel, I max is the maximum grayscale value of each pixel point in this component, I min is the minimum grayscale value of each pixel point in this component, and M (x,y) is the grayscale value of each pixel point in the stretching component.

[0113] B53. Based on the stretching components corresponding to each component in the HSV space, perform a space conversion to convert the HSV space into the RGB space to obtain an enhanced image.

[0114] Exemplarily, when the stretching scheme is piecewise linear stretching, the intelligent monitoring device can perform piecewise linear stretching on the projected image through steps B61 - B66 to obtain an enhanced image.

[0115] B61. Perform a space conversion on the projected image to convert the RGB space into the HSV space to obtain the H component, S component, and V component of the HSV space.

[0116] B62. For any component in the HSV space, determine the first segmentation point and the second segmentation point, as well as the first enhanced grayscale and the second enhanced grayscale, and use the following formula to calculate three stretching coefficients.

[0117] k1 = y1 / x1

[0118] k2 = (y2 - y1) / (x2 - x1);

[0119] k3 = (1 - y2) / (1 - x2)

[0120] Wherein, k1 is the first stretching coefficient, k2 is the second stretching coefficient, k3 is the third stretching coefficient, x1 is the first segmentation point, x2 is the second segmentation point, y1 is the first enhanced gray level, and y2 is the second enhanced gray level.

[0121] B63. Calculate three stretching offset values according to the three stretching coefficients.

[0122] B64. Implement image stretching processing of this component according to the following formula to obtain the stretched component corresponding to this component.

[0123]

[0124] Wherein, x1 is the first segmentation point, x2 is the second segmentation point, k1 is the first stretching coefficient, k2 is the second stretching coefficient, k3 is the third stretching coefficient, b1 is the first stretching offset value, b2 is the second stretching offset value, b3 is the third stretching offset value, y is the gray value of each pixel point in the stretched component, and x is the gray value of each pixel point in this component.

[0125] B65. Based on the stretched components corresponding to each component in the HSV space, perform space conversion to convert the HSV space to the RGB space to obtain the enhanced image.

[0126] S104. Perform edge detection and optimization on the enhanced image to obtain the edge-optimized image.

[0127] As a possible implementation manner, the intelligent monitoring device can obtain the enhanced image through steps S1041 - S1045.

[0128] S1041. Perform a blocking operation on the enhanced image to obtain a plurality of blocked images.

[0129] Exemplarily, the present invention can determine the number of segmentation blocks of the enhanced image based on the following formula, and then complete the blocking operation on the enhanced image based on the number of segmentation blocks.

[0130] n = INT[1 / FTU];

[0131] Wherein, FTU is the turbidity corresponding to the projection image, the value range is (0, 1), INT is the rounding operation, and n is the number of segmentation blocks of the enhanced image.

[0132] S1042. For any blocked image, use the Canny operator to perform edge detection on the submerged plants in the blocked image to determine the blocked image with edge contours.

[0133] Exemplarily, the intelligent monitoring device can determine a segmented image with an edge contour through steps C11 - C14.

[0134] C11. Perform Gaussian smoothing on the arbitrary segmented image to remove high - frequency noise in the arbitrary segmented image, obtaining a Gaussian - smoothed image.

[0135] C12. Use the soble horizontal operator and the soble vertical operator to perform convolution processing on the Gaussian - smoothed image, obtaining the edge intensity and direction of each pixel point in the Gaussian - smoothed image.

[0136] C13. Based on the edge intensity and direction of each pixel point in the Gaussian - smoothed image, adopt a double - threshold algorithm to determine the first edge and the second edge in the Gaussian - smoothed image.

[0137] Among them, the first edge is the edge determined by using the lower threshold in the double - threshold, and the second edge is the edge determined by using the higher threshold in the double - threshold.

[0138] C14. In the Gaussian - smoothed image, based on the second edge, select new edge points between the first edge and the second edge, complement the breakpoints in the second edge, and form a closed edge contour to obtain a segmented image with an edge contour.

[0139] It should be noted that in the present invention, within the neighborhood range of an arbitrary segmented image, a Gaussian filtering operator generated by using the Gaussian formula is used for convolution operation to implement weighted average operation, effectively removing those high - frequency noises in the image and completing Gaussian smoothing. The soble horizontal and vertical operators are selected to perform convolution operation within the neighborhood range of an arbitrary segmented image to obtain the gradient magnitude and direction, and estimate the edge intensity and direction at each pixel point. According to the gradient direction, non - maximum suppression is performed on the gradient magnitude. The double - threshold algorithm is used to detect and connect edges. Two thresholds are respectively selected, and usually the ratio of the magnitudes of these two thresholds is 1:2 or 1:3. Pixel points smaller than the lower threshold are removed and determined as false edges; pixel points greater than the higher threshold are retained and marked as strong edges; while pixel points between the two thresholds need to be further determined according to specific rules. Based on the pixel points formed after being determined by the higher threshold, those pixel points are connected into an edge. When reaching the end points of the edge, new edge points that meet the lower threshold are re - determined among their neighboring pixel points, and then based on this, new edge points are continuously detected and connected until the contour is closed.

[0140] S1043. Perform foreground - background distinction on each segmented image with an edge contour to determine the foreground region and the background region in each segmented image.

[0141] Exemplarily, the intelligent monitoring device can determine a segmented image with an edge contour through steps C21 - C23.

[0142] C21. For any segmented image with an edge contour, based on the edge contour, calculate the average gray value of the area inside the contour and the area outside the contour respectively.

[0143] C22. Based on the average gray values of the area inside the contour and the area outside the contour, perform gray generalization on each pixel point in the segmented image to obtain the generalized segmented image.

[0144] C23. Based on the gray values of each pixel point in the generalized segmented image, determine the foreground area and the background area of the segmented image.

[0145] It should be noted that in the present invention, the contour edge B extracted according to the Canny operator within the block C , is the local initial value B c0 , calculate the contour length penalty term, length constraint, which is used to regularize the evolution curve C to ensure that the obtained curve is short enough. The closed edge contour line B c0 divides the local picture into two parts, the area inside the contour inside(B C ) and the area outside the contour outside(B C ), complete the marking, 1 for the inside and 0 for the outside, and calculate the area penalty term inside the contour. Calculate B c1 and B c2 which are the average gray values of the images inside and outside the evolution curve Bc0 of the contour respectively. Use the previously estimated foreground and background means Bc1 and Bc2 to adjust (evolve) each point of the level set, and judge whether the gray value of the current point increases or decreases after introducing the energy functional, so as to judge whether it is a foreground or background pixel.

[0146] Exemplarily, the present invention can determine the gray value of each pixel point based on the following formula.

[0147]

[0148] Among them, F1(B c1 ,B c2 ,B C ) represents the gray value of the pixel point (x, y) after generalization, B C represents the contour edge, B c0 represents the initial value of the contour edge, that is, the evolution curve of the contour edge, B c1 represents the average gray value inside the contour, B c2 represents the average gray value outside the contour, μ, ν, λ1 and λ2 are positive constants, for example, λ1 = λ2 = 1, Length(B C ) represents the length of the contour edge, Area(inside(B C)) represents the area inside the contour edge, inside(B C ) represents the inside of the contour edge, outside(B C ) represents the outside of the contour edge, and u0(x, y) represents the gray value of the pixel point (x, y) before generalization.

[0149] S1044. Based on the edge contours, foreground regions, and background regions of each sub-block image, convert each sub-block image into each binary image.

[0150] Exemplarily, the present invention can perform segmentation and marking on the foreground region and the background region based on the edge contour, and fill the image into a binary image.

[0151] S1045. Perform inter-block merging based on each binary image to obtain an edge-optimized image.

[0152] Exemplarily, the present invention can perform an opening operation within the binary image, erode first and then dilate, to smooth the burrs and sharp corners and eliminate the external isolated points.

[0153] Another exemplarily, the present invention can perform a closing operation within the binary image, dilate first and then erode to remove small holes in the foreground.

[0154] S105. Based on the edge-optimized image, determine the coverage area of the submerged plants.

[0155] As a possible implementation manner, the present invention can calculate the area of the foreground part in the edge-optimized image, and determine the area of the foreground part as the coverage area of the submerged plants.

[0156] The present invention provides an intelligent monitoring method for the coverage area of submerged plants. While obtaining the projection image of the submerged plants, the present invention obtains the turbidity data corresponding to the projection image; and based on the turbidity data, performs image stretching on the projection image by using different image stretching methods to enhance the contrast of the projection image and obtain an enhanced image. Then, perform edge detection and optimization on the enhanced image to obtain an edge-optimized image, and finally obtain the coverage area of the sleeping plants according to the edge-optimized image. Since in the process of monitoring the coverage area of submerged plants, the present invention enhances the contrast before edge detection by means of image stretching based on turbidity, improves the clarity of the captured image of the submerged plants, and thus improves the accuracy of detecting the coverage area of the submerged plants.

[0157] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The order of execution of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0158] The following is an apparatus embodiment of the present invention. For the details not described in detail therein, reference may be made to the corresponding method embodiments above.

[0159] Figure 2 The structural schematic diagram of an intelligent monitoring device for the coverage area of submerged plants provided by an embodiment of the present invention is shown. The intelligent monitoring device 200 includes a communication module 201 and a processing module 202.

[0160] The communication module 201 is used to obtain the projection image of the submerged plants and the turbidity data corresponding to the projection image.

[0161] The processing module 202 is used to determine the turbidity level corresponding to the projection image based on the turbidity data corresponding to the projection image; perform image stretching on the projection image based on the turbidity level corresponding to the projection image to obtain an enhanced image; wherein, different image stretching methods correspond to different turbidity levels; perform edge detection and optimization on the enhanced image to obtain an edge-optimized image; determine the coverage area of the submerged plants based on the edge-optimized image.

[0162] In a possible implementation manner, the processing module 202 is further used to extract the image blur feature and the image structure feature of the projection image; divide the projection image into multiple regions based on the image blur feature and the image structure feature; wherein, the clarity of each region of the projection image is different; perform attention analysis on the clarity of the projection image to obtain an attention matrix; the weight of the clear pixels in the attention matrix is higher than the weight of the blurred pixels; reconstruct the projection image based on the multiple regions of the projection image and the attention matrix to implement preprocessing of the projection image.

[0163] In a possible implementation manner, the processing module 202 is specifically used to determine the stretching scheme of the projection image based on the turbidity level corresponding to the projection image; the stretching scheme includes one of the following: histogram stretching, Retinex stretching, logarithmic transformation stretching, exponential transformation stretching, maximum-minimum stretching, and piecewise linear stretching; perform image stretching on the projection image based on the stretching scheme of the projection image to obtain an enhanced image.

[0164] In a possible implementation, the processing module 202 is specifically configured to, if the turbidity level corresponding to the projected image is low turbidity, the stretching scheme of the projected image is histogram stretching or Retinex stretching; correspondingly, the processing module is specifically configured to perform a spatial conversion on the projected image to convert the RGB space into the HSV space, obtaining the H component, the S component, and the V component of the HSV space; for any component of the HSV space, calculate the total number of pixels of the component and the gray values of each pixel in the component; based on the total number of pixels of the component and the gray values of each pixel in the component, perform a level division to obtain multiple gray levels; calculate the cumulative distribution values corresponding to each gray level; based on the cumulative distribution values corresponding to each gray level, perform histogram equalization on the component to obtain the stretched component corresponding to the component; based on the stretched components corresponding to each component in the HSV space, perform a spatial conversion to convert the HSV space into the RGB space, obtaining the enhanced image.

[0165] In a possible implementation, the processing module 202 is specifically configured to perform a blocking operation on the enhanced image to obtain multiple blocked images; for any blocked image, use the Canny operator to perform edge detection on the submerged plants in the blocked image to determine the blocked image with edge contours; perform foreground and background differentiation on each blocked image with edge contours to determine the foreground area and the background area in each blocked image; based on the edge contours, the foreground area, and the background area of each blocked image, convert each blocked image into each binary image; based on each binary image, perform inter-block merging to obtain the edge-optimized image.

[0166] In a possible implementation, the processing module 202 is specifically configured to perform Gaussian smoothing processing on the any blocked image to remove the high-frequency noise in the any blocked image, obtaining the Gaussian smoothed image; use the soble horizontal operator and the soble vertical operator to perform convolution processing on the Gaussian smoothed image to obtain the edge intensity and direction of each pixel point in the Gaussian smoothed image; based on the edge intensity and direction of each pixel point in the Gaussian smoothed image, adopt the double-threshold algorithm to determine the first edge and the second edge in the Gaussian smoothed image, where the first edge is the edge determined by using the low threshold in the double threshold, and the second edge is the edge determined by using the high threshold in the double threshold; in the Gaussian smoothed image, based on the second edge, select new edge points between the first edge and the second edge to complete the breakpoints in the second edge, forming a closed edge contour, obtaining the blocked image with edge contours.

[0167] In a possible implementation, the processing module 202 is specifically configured to, for any block image with an edge contour, calculate the average gray value of the area inside the contour and the area outside the contour respectively based on the edge contour; perform gray generalization on each pixel point in the block image based on the average gray values of the area inside the contour and the area outside the contour to obtain a generalized block image; and determine the foreground area and the background area of the block image based on the gray values of the pixel points in the generalized block image.

[0168] Figure 3 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. As Figure 3 shown, the electronic device 300 of this embodiment includes: a processor 301, a memory 302, and a computer program 303 stored in the memory 302 and executable on the processor 301. When the processor 301 executes the computer program 303, the steps in the above method embodiments are implemented, such as Figure 1 the steps 101 to 105 shown. Alternatively, when the processor 301 executes the computer program 303, the functions of each module / unit in the above device embodiments are implemented. For example, Figure 2 the functions of the communication module 201 and the processing module 202 shown.

[0169] Exemplarily, the computer program 303 can be divided into one or more modules / units. The one or more modules / units are stored in the memory 302 and executed by the processor 301 to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program 303 in the electronic device 300. For example, the computer program 303 can be divided into Figure 2 the communication module 201 and the processing module 202 shown.

[0170] The so-called processor 301 may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor, or the processor may also be any conventional processor, etc.

[0171] The memory 302 may be an internal storage unit of the electronic device 300, such as a hard disk or memory of the electronic device 300. The memory 302 may also be an external storage device of the electronic device 300, such as a plug-in hard disk equipped on the electronic device 300, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. Further, the memory 302 may also include both the internal storage unit of the electronic device 300 and an external storage device. The memory 302 is used to store the computer program and other programs and data required by the terminal. The memory 302 may also be used to temporarily store data that has been output or is to be output.

[0172] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiments can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of the present application. The specific working processes of the units and modules in the above system can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0173] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0174] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0175] In the embodiments provided by the present invention, it should be understood that the disclosed device / terminal and method can be implemented in other ways. For example, the device / terminal embodiments described above are merely illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical or other forms.

[0176] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0177] In addition, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0178] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above method embodiments of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0179] The above-described embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. An intelligent monitoring method for the coverage area of submerged plants, characterized in that, Including: Obtaining a projection image of submerged plants and turbidity data corresponding to the projection image; Determining a turbidity level corresponding to the projection image based on the turbidity data corresponding to the projection image; Performing image stretching on the projection image based on the turbidity level corresponding to the projection image to obtain an enhanced image; wherein, different image stretching methods are used for different turbidity levels; Performing edge detection and optimization on the enhanced image to obtain an edge-optimized image; Determining the coverage area of the submerged plants based on the edge-optimized image; Before performing image stretching on the projection image based on the turbidity level corresponding to the projection image to obtain an enhanced image, it further includes: extracting image blur features and image structure features of the projection image; dividing the projection image into multiple regions based on the image blur features and image structure features; wherein, the clarity of each region of the projection image is different; performing attention analysis on the clarity of the projection image to obtain an attention matrix; the weight of clear pixels in the attention matrix is higher than that of blurred pixels; reconstructing the projection image based on the multiple regions of the projection image and the attention matrix to implement preprocessing of the projection image.

2. The intelligent monitoring method for the coverage area of submerged plants according to claim 1, characterized in that Performing image stretching on the projection image based on the turbidity level corresponding to the projection image to obtain an enhanced image includes: Determining a stretching scheme for the projection image based on the turbidity level corresponding to the projection image; the stretching scheme includes one of the following: histogram stretching, Retinex stretching, logarithmic transformation stretching, exponential transformation stretching, maximum-minimum stretching, and piecewise linear stretching; Performing image stretching on the projection image based on the stretching scheme of the projection image to obtain an enhanced image.

3. The intelligent monitoring method for the coverage area of submerged plants according to claim 1, characterized in that Determining the stretching scheme for the projection image based on the turbidity level corresponding to the projection image includes: If the turbidity level corresponding to the projection image is low turbidity, determining the stretching scheme for the projection image as histogram stretching or Retinex stretching; Correspondingly, performing image stretching on the projection image based on the stretching scheme of the projection image to obtain an enhanced image includes: Performing a spatial transformation on the projection image to convert the RGB space to the HSV space, obtaining the H component, S component, and V component of the HSV space; For any component of the HSV space, calculating the total number of pixels of the component and the gray values of each pixel in the component; Performing level division based on the total number of pixels of the component and the gray values of each pixel in the component to obtain multiple gray levels; Calculating the cumulative distribution values corresponding to each gray level; Performing histogram equalization on the component based on the cumulative distribution values corresponding to each gray level to obtain a stretched component corresponding to the component; Performing a spatial transformation based on the stretched components corresponding to each component in the HSV space to convert the HSV space to the RGB space to obtain the enhanced image.

4. The intelligent monitoring method for the submerged plant coverage area according to claim 1, characterized in that Performing edge detection and optimization on the enhanced image to obtain an edge-optimized image includes: Performing a block operation on the enhanced image to obtain multiple block images; For any sub-block image, use the Canny operator to perform edge detection on the submerged plants in the sub-block image to determine the sub-block image with an edge contour; Distinguish the foreground and background of each sub-block image with an edge contour to determine the foreground area and background area in each sub-block image; Based on the edge contour, foreground area and background area of each sub-block image, convert each sub-block image into each binary image; Based on the binary images, perform inter-block merging to obtain an edge-optimized image.

5. The intelligent monitoring method for the submerged plant coverage area according to claim 4, characterized in that, The step of, for any sub-block image, using the Canny operator to perform edge detection on the submerged plants in the sub-block image to determine the sub-block image with an edge contour includes: Perform Gaussian smoothing on the sub-block image to remove high-frequency noise in the sub-block image and obtain a Gaussian-smoothed image; Use the soble horizontal operator and the soble vertical operator to perform convolution processing on the Gaussian-smoothed image to obtain the edge intensity and direction of each pixel point in the Gaussian-smoothed image; Based on the edge intensity and direction of each pixel point in the Gaussian-smoothed image, use the double-threshold algorithm to determine the first edge and the second edge in the Gaussian-smoothed image. The first edge is the edge determined by the low threshold in the double threshold, and the second edge is the edge determined by the high threshold in the double threshold; In the Gaussian-smoothed image, based on the second edge, select new edge points between the first edge and the second edge to complete the breakpoints in the second edge and form a closed edge contour to obtain the sub-block image with an edge contour.

6. The intelligent monitoring method for the submerged plant coverage area according to claim 4, wherein The step of distinguishing the foreground and background of each sub-block image with an edge contour to determine the foreground area and background area in each sub-block image includes: For any sub-block image with an edge contour, based on the edge contour, calculate the average gray value of the area inside the contour and the area outside the contour respectively; Based on the average gray values of the area inside the contour and the area outside the contour, perform gray generalization on each pixel point in the sub-block image to obtain a generalized sub-block image; Based on the gray value of each pixel point in the generalized sub-block image, determine the foreground area and background area of the sub-block image.

7. An intelligent monitoring device for the coverage area of submerged plants, characterized in that, It includes: A communication module for acquiring a projection image of submerged plants and turbidity data corresponding to the projection image; A processing module for determining the turbidity level corresponding to the projection image based on the turbidity data corresponding to the projection image; performing image stretching on the projection image based on the turbidity level corresponding to the projection image to obtain an enhanced image, where different image stretching methods correspond to different turbidity levels; performing edge detection and optimization on the enhanced image to obtain an edge-optimized image; and determining the coverage area of the submerged plants based on the edge-optimized image. The processing module is further configured to extract the image blur feature and the image structure feature of the projection image; divide the projection image into multiple regions based on the image blur feature and the image structure feature; wherein, the clarity of each region of the projection image is different; perform attention analysis on the clarity of the projection image to obtain an attention matrix; the weight of the clear pixels in the attention matrix is higher than the weight of the blurred pixels; reconstruct the projection image based on the multiple regions of the projection image and the attention matrix to implement preprocessing of the projection image.

8. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory stores a computer program, and the processor is configured to call and run the computer program stored in the memory to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 6 above are implemented.

Images