Plateau lake area and water storage capacity dynamic monitoring method based on remote sensing image analysis

Abstract

The invention discloses a plateau lake area and water storage capacity dynamic monitoring method based on remote sensing image analysis, and relates to the technical field of lake monitoring, and the method comprises the following steps: carrying out the binarization processing of a lake gray-scale image based on a first segmentation threshold value and a second segmentation threshold value, obtaining a lake binary image, and obtaining a main analysis region based on the lake binary image; performing expansion optimization on the initial contour to obtain a final analysis area; acquiring a real-time region area based on the final analysis region; based on the second number of historical lake maps and the corresponding water storage capacity, obtaining a relation function between the region area and the water storage capacity, and marking the relation function as a water storage relation function; substituting the real-time area into a water storage relation function to obtain real-time water storage capacity; the method is used for solving the problem that the obtained area of the lake is inaccurate due to errors when image processing is carried out on the clear and shallow lake water on the shore in the existing lake monitoring technology.

Description

Technical Field

[0001] This invention relates to the field of lake monitoring technology, specifically a method for dynamic monitoring of the area and water storage of plateau lakes based on remote sensing image analysis. Background Technology

[0002] Due to their unique geographical location, plateau lakes are a key component of water resources and the ecological environment. Changes in their water volume have a significant impact on the surrounding areas, so it is necessary to monitor the area and water storage of plateau lakes. With the development of image processing technology, it is possible to analyze the area of ​​plateau lakes based on images. However, plateau lakes often have shallower areas near the shore. If the water is clear and the shoreline is too smooth, it is difficult to distinguish between the lake area and the shoreline area from the image. For example, patent application CN110276797A discloses a lake area extraction method. This method fails to perform image enhancement processing on the shallower areas, resulting in discontinuous pixels at the lake area edges. This leads to errors in obtaining the lake area and makes the subsequently obtained lake area inaccurate. Existing lake monitoring technologies have errors when processing images of clear and shallow lake water near the shore, resulting in inaccurate lake area acquisition. Summary of the Invention

[0003] This invention aims to at least partially solve one of the technical problems in the prior art. It involves binarizing a lake grayscale image based on a first and a second segmentation threshold to obtain a binarized lake image, obtaining the main analysis region based on the binarized lake image, performing dilation optimization on the initial contour to obtain the final analysis region, and obtaining the real-time area based on the final analysis region. Furthermore, it obtains the relationship function between the area and the water storage based on a second number of historical lake images and their corresponding water storage volumes, and labels this function as the water storage relationship function. This addresses the problem in existing lake monitoring technologies where errors exist when processing images of clear, shallow lake water near the shore, leading to inaccurate lake area measurements.

[0004] To achieve the above objectives, this application provides a method for dynamic monitoring of the area and water storage of plateau lakes based on remote sensing image analysis, comprising the following steps: Obtain RGB category images from remote sensing images of plateau lakes and label them as lake maps; The RGB values ​​of each pixel in the lake image are converted into grayscale values ​​using a weighted average method to obtain a grayscale image of the lake. A first range threshold and a second range threshold are obtained based on a first number of historical lake maps; The first segmentation threshold and the second segmentation threshold are obtained based on the first range threshold, the second range threshold, and the lake grayscale image; The lake grayscale image is binarized based on the first segmentation threshold and the second segmentation threshold to obtain the lake binarized image; The main analysis region was obtained based on the lake binarization map; The initial contour is dilated and optimized to obtain the final analysis region; Obtain the real-time area of ​​the region based on the final analysis region; Based on the second number of historical lake maps and the corresponding water storage, the relationship function between the area and water storage is obtained and marked as the water storage relationship function; Substitute the real-time area into the water storage relationship function to obtain the real-time water storage volume; The real-time area and real-time water storage are used to display the area and water storage of the lake.

[0005] Furthermore, obtaining the first range threshold and the second range threshold based on the first number of historical lake maps includes the following sub-steps: Obtain the first number of historical lake images, and calibrate the pixels of the lake portion in the historical lake images to obtain the calibrated pixels; The RGB values ​​of each pixel in the historical lake image are converted into grayscale values ​​using a weighted average method to obtain a historical grayscale image.

[0006] Furthermore, obtaining the first range threshold and the second range threshold based on the first number of historical lake maps also includes the following sub-steps: Obtain the minimum and maximum gray values ​​of the labeled pixel after deleting the minimum and maximum gray values ​​of a labeled pixel, respectively, and mark them as the first range threshold and the second range threshold.

[0007] Furthermore, obtaining the first segmentation threshold and the second segmentation threshold based on the first range threshold, the second range threshold, and the lake grayscale image includes the following sub-steps: Obtain the grayscale value of each pixel in the lake grayscale image and label it as the lake grayscale value; Divide the gray values ​​from 0 to 255 into p equal intervals, and mark them as gray value ranges; Count the frequency of lake grayscale values ​​within each grayscale division range and label it as the grayscale frequency of the lake division; A histogram is drawn with the lake's grayscale value on the X-axis, the frequency of the lake's grayscale values ​​on the Y-axis, and the grayscale range as the histogram interval. This histogram is then labeled as the lake's grayscale histogram.

[0008] Furthermore, obtaining the first segmentation threshold and the second segmentation threshold based on the first range threshold, the second range threshold, and the lake grayscale image also includes the following sub-steps: The interval with the highest gray-level frequency in the lake gray-level histogram is obtained and marked as the peak interval. The median abscissa of the interval with the minimum gray frequency of the lake is obtained on the left and right sides of the peak interval, and marked as the first image threshold and the second image threshold. Calculate the average of the first range threshold and the first graphic threshold, and mark it as the first segmentation threshold; Calculate the average of the second range threshold and the second graphic threshold, and mark it as the second segmentation threshold.

[0009] Furthermore, the process of binarizing the lake grayscale image based on the first and second segmentation thresholds to obtain the lake binarized image includes the following sub-steps: In the lake grayscale image, lake grayscale values ​​that are greater than or equal to the first segmentation threshold and less than or equal to the second segmentation threshold are set to 0, and lake grayscale values ​​that are less than the first segmentation threshold or greater than the second segmentation threshold are set to 255 to obtain a lake binarized image.

[0010] Furthermore, obtaining the main analysis region based on the lake binarization map includes the following sub-steps: Mark the pixels with a grayscale value of 0 in the binarized lake image as the initial pixels; Starting from any initial pixel, search for a pixel in each of the eight directions and mark the found pixel as the search pixel; the eight directions are up, down, left, right, upper left, lower left, upper right, and lower right. If the search pixel contains a new initial pixel, new search pixels are obtained in eight directions from the new initial pixel. If the new search pixel contains a new initial pixel, the search is repeated with the new initial pixel until the new search pixel does not contain a new initial pixel. Then, it is considered that one search is completed, and the area formed by the initial pixels obtained after one search is marked as the area to be analyzed. The region with the most pixels is identified and marked as the main analysis region.

[0011] Furthermore, the process of dilating and optimizing the initial contour to obtain the final analysis region includes the following sub-steps: Initial pixels that are not within the main analysis area are marked as discrete pixels; Obtain the edge pixels of the main analysis region and mark them as main edge pixels. Construct a rectangle with a side length j times the pixel width, centered on each main edge pixel, and mark it as an expansion rectangle. Determine if the expansion rectangle contains discrete pixels. If it does, mark the discrete pixels within the expansion rectangle as expansion pixels. Connect the expansion pixels to the nearest main edge pixel to obtain the first line segment. Then, construct a new expansion rectangle centered on the expansion pixel and determine if the new expansion rectangle contains new discrete pixels. Connect the new discrete pixels to the expansion pixels to obtain the second line segment. Repeat the construction of the expansion matrix and connecting line segments until the new expansion rectangle does not contain any new discrete pixels. For consecutive line segments up to the nth line segment, j is an odd number greater than 3. Mark all pixels on line segments 1 to n as incremental pixels; The final analysis area is formed by combining all incremental pixels and the main analysis area.

[0012] Furthermore, obtaining the real-time region area based on the final analysis region includes the following sub-steps: Obtain the number of pixels in the final analysis area and mark it as the final analysis count; obtain the area of ​​each pixel. Obtain the scaling ratio of the lake image compared to the actual size of the objects, and label it as the image scaling ratio; The real-time region area is calculated as: Ss = Z * Mj / x; where Ss is the real-time region area, Z is the final number of analyses, Mj is the area of ​​each pixel, and x is the image scaling ratio.

[0013] Furthermore, based on a second set of historical lake maps and their corresponding water storage, the relationship function between the area and water storage is obtained and labeled as the water storage relationship function, which includes the following sub-steps: Obtain the area of ​​the final analysis region of the second number of historical lake maps, and mark it as the historical region area; A plane rectangular coordinate system is established with the historical area as the horizontal axis and the water storage volume as the vertical axis, and it is marked as the water storage relationship coordinate system. By using the historical area and corresponding water storage volume as the x and y coordinates of the water storage coordinate points, and fitting all water storage coordinate points together, a water storage relationship function is obtained.

[0014] The beneficial effects of this invention are as follows: This invention obtains the main analysis region based on the binarized lake map, performs dilation optimization on the initial contour to obtain the final analysis region, and obtains the real-time area based on the final analysis region; it obtains the relationship function between the area and the water storage based on a second number of historical lake maps and the corresponding water storage, and marks it as the water storage relationship function. The advantage is that it can optimize the lake obtained from the image to obtain a more accurate lake area and improve the accuracy of the lake area in the lake region. This invention obtains the final analysis area by dilating and optimizing the initial contour. Its advantage is that the contour is unclear due to the shallow water on the shore, resulting in discontinuous contour pixels. This method can obtain a continuous lake contour, making the subsequent calculation of the lake area more accurate. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the steps of the method of the present invention; Figure 2 This is a schematic diagram of the lake grayscale histogram of the present invention; Figure 3 This is a schematic diagram of the expansion matrix of the present invention; Figure 4 This is a schematic diagram of the incremental pixel points of the present invention; Figure 5 This is a schematic diagram of the water storage relationship function of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Example 1, please refer to Figure 1 As shown, this application provides a method for dynamic monitoring of the area and water storage of plateau lakes based on remote sensing image analysis, including the following steps: Step S1: Obtain RGB category images from remote sensing images of plateau lakes and label them as lake maps; the lake maps here are RGB format images.

[0018] Step S2: Convert the RGB values ​​of each pixel in the lake image into grayscale values ​​using a weighted average method to obtain a grayscale image of the lake. In practical applications, the weighted average method involves summing the R, G, and B values ​​with weights of 0.299, 0.578, and 0.114, respectively. These weights take into account the human eye's sensitivity to different colors and can more realistically reflect visual effects. For example, when the RGB value is (88, 195, 224), the grayscale value is 0.299*88 + 0.578*195 + 0.114*224 = 166, and the calculation result is rounded to the nearest integer.

[0019] Step S3 involves obtaining a first range threshold and a second range threshold based on a first number of historical lake maps; Step S3 also includes the following sub-steps: Step S301: Obtain a first number of historical lake images, and calibrate the lake part pixels in the historical lake images to obtain calibrated pixels; Step S302: Convert the RGB values ​​of each pixel in the historical lake image into grayscale values ​​using a weighted average method to obtain a historical grayscale image; Step S303: Obtain the minimum and maximum gray values ​​of the labeled pixel after deleting the minimum and maximum gray values ​​of a labeled pixel respectively, and mark them as the first range threshold and the second range threshold respectively; In practical applications, historical lake images are obtained as historical images of the lake to be detected. The distribution range of the lake's grayscale values ​​is obtained based on the historical images, for example, the first range threshold and the second range threshold are 51 to 165.

[0020] Step S4 involves obtaining a first segmentation threshold and a second segmentation threshold based on the first range threshold, the second range threshold, and the lake grayscale image; Step S4 also includes the following sub-steps: Step S401: Obtain the grayscale value of each pixel in the lake grayscale image and mark it as the lake grayscale value; Step S402: Divide the gray values ​​from 0 to 255 into p equally spaced ranges, and mark them as gray value division ranges; Step S403: Count the frequency of lake gray values ​​within each gray level division range and mark them as the gray level frequency of the divided lakes; Step S404: Using the lake's grayscale value as the X-axis, the lake's grayscale frequency as the Y-axis, and the grayscale range as the histogram interval, draw a histogram and mark it as the lake's grayscale histogram. Step S405: Obtain the interval with the highest gray frequency in the lake gray histogram and mark it as the peak interval; Step S406: Obtain the median horizontal coordinate of the interval with the smallest gray frequency of the lake on the left and right sides of the peak interval respectively, and mark it as the first image threshold and the second image threshold. Step S407: Calculate the average of the first range threshold and the first graphic threshold, and mark it as the first segmentation threshold; Step S408: Calculate the average of the second range threshold and the second image threshold, and mark it as the second segmentation threshold; combine the historical image and the real-time image to obtain the binarization segmentation threshold, so that the subsequent acquisition of the lake binarization image is more accurate; For practical applications, please refer to Figure 2 As shown, the gray values ​​from 0 to 255 are divided into 8 equal intervals. In the obtained lake image, the lake occupies the largest area, so the peak interval can be identified as the gray value distribution range of the lake. It also contains the soil and rock parts and the shaded parts. The gray values ​​of the soil and rock parts and the shaded parts are different from those of the lake, so a multi-peak histogram will appear. The gray values ​​of the shaded parts are distributed in the area of ​​smaller gray values, while the soil and rock parts of the plateau are distributed in the area of ​​larger gray values. The minimum value between them is the segmentation threshold. Therefore, the interval with the smallest gray frequency of the lake can be divided into the left and right sides of the peak interval as the threshold range of the lake area. The first image threshold and the second image threshold are 47 and 175, respectively. The average of the first range threshold 51 and the first image threshold 47 is 49, and 49 is the first segmentation threshold. The average of the second range threshold 165 and the second image threshold 175 is 170, and 170 is the second segmentation threshold.

[0021] Step S5 involves binarizing the lake grayscale image based on the first and second segmentation thresholds to obtain a lake binarized image; Step S5 also includes the following sub-steps: Step S501: Set the lake grayscale values ​​in the lake grayscale image that are greater than or equal to the first segmentation threshold and less than or equal to the second segmentation threshold to 0, and set the lake grayscale values ​​in the lake grayscale image that are less than the first segmentation threshold or greater than the second segmentation threshold to 255 to obtain a lake binarized image. In practical applications, lake grayscale values ​​greater than or equal to 49 and less than or equal to 170 in the lake grayscale image are set to 0, and lake grayscale values ​​less than 49 or greater than 170 in the lake grayscale image are set to 255 to obtain a lake binarized image.

[0022] Step S6: Obtain the main analysis region based on the lake binarization map; Step S6 also includes the following sub-steps: Step S601: Mark the pixels with a gray value of 0 in the lake binarized image as the initial pixels; Step S602: Starting from any initial pixel, search for a pixel in each of the eight directions and mark the found pixel as the search pixel; where the eight directions are up, down, left, right, upper left, lower left, upper right, and lower right. Step S603: If the search pixel contains a new initial pixel, obtain new search pixels in eight directions from the new initial pixel. If the new search pixel contains a new initial pixel, repeat the search with the new initial pixel until the new search pixel does not contain a new initial pixel. Then, it is considered that one search is completed, and the area formed by the initial pixels obtained after one search is marked as the area to be analyzed. The lake area is a whole area, so this method can be used to search all continuous pixels in the image. Step S604: Obtain the region to be analyzed with the most pixels and mark it as the main analysis region. Since the region with a gray value of 0 is identified as a lake pixel, and the lake is the largest continuous region in the lake image, the lake region can be initially obtained. At the same time, the binarized region contains noisy pixels, which can be filtered out by this method.

[0023] Step S7 involves performing dilation optimization on the initial contour to obtain the final analysis region; Step S7 also includes the following sub-steps: Step S701: Mark the initial pixels that are not in the main analysis area as discrete pixels; Step S702: Obtain the edge pixels of the main analysis region and mark them as main body edge pixels. Construct a rectangle with a side length j times the pixel width, centered on each main body edge pixel, and mark it as an expansion rectangle. Determine whether the expansion rectangle contains discrete pixels. If it does, mark the discrete pixels within the expansion rectangle as expansion pixels. Connect the expansion pixels with the nearest main body edge pixel to obtain the first line segment. Then, construct a new expansion rectangle centered on the expansion pixels and determine whether the new expansion rectangle contains new discrete pixels. Connect the new discrete pixels with the expansion pixels to obtain the first line segment. Take the second line segment, and repeat the establishment of the dilation matrix and connection of line segments until the new dilation rectangle does not contain any new discrete pixels; among them, for continuous line segments up to the nth line segment, j is an odd number greater than 3; because the water is clear and shallow in the edge area of ​​the lake region, the edge area of ​​the lake region in the obtained binarized image separates the main area, resulting in isolated lake pixels in the main analysis area. Therefore, the surrounding discrete pixels can be integrated to obtain the correct lake area. Since it is necessary to select the discrete pixels around the main analysis area, j is an odd number greater than 3. For example, j is set to 5. If the shallow water area is large, j can be appropriately increased. Step S703: Mark all pixels on line segments 1 to n as incremental pixels; Step S704: Combine all incremental pixels and the main analysis region to form the final analysis region; In practical applications, taking the acquisition of an incremental pixel as an example, j is set to 5. Please participate... Figure 3 and Figure 4 The search increment pixels are obtained, and then all the search increment pixels are obtained. All the increment pixels and the main analysis region are combined to form the final analysis region.

[0024] Step S8: Obtain the real-time area based on the final analysis region; Step S8 also includes the following sub-steps: Step S801: Obtain the number of pixels in the final analysis area and mark it as the final analysis number; obtain the area of ​​each pixel. Step S802: Obtain the scaling ratio of the lake image compared to the actual size of the object, and mark it as the image scaling ratio; Step S803, calculate the real-time region area as: Ss=Z*Mj / x; where Ss is the real-time region area, Z is the final number of analyses, Mj is the area of ​​each pixel, and x is the image reduction ratio; In practical applications, the final number of analyzed pixels was 2,132,442, the area of ​​each pixel was 0.0079 m², the image reduction ratio was 1 / 142,470, and the calculated real-time region area was: Ss = 2401 (km²). 2 The calculation result is rounded to the nearest integer.

[0025] Step S9 involves obtaining the relationship function between the area and water storage of a region based on a second number of historical lake maps and their corresponding water storage volumes, and labeling it as the water storage relationship function. Step S9 also includes the following sub-steps: Step S901: Obtain the area of ​​the final analysis region of the second number of historical lake maps and mark it as the historical region area; Step S902: Establish a plane rectangular coordinate system with the area of ​​the historical region as the horizontal axis and the water storage volume as the vertical axis, and mark it as the water storage relationship coordinate system; Step S903: Use the area of ​​the historical region and the corresponding water storage volume as the abscissa and ordinate of the water storage coordinate points, and fit all the water storage coordinate points to obtain the water storage relationship function. For practical applications, please refer to Figure 5 As shown, obtain the water storage relationship function.

[0026] Step S10: Substitute the real-time area into the water storage relationship function to obtain the real-time water storage volume.

[0027] Step S11: Display the real-time area and real-time water storage as the area and water storage of the lake. For practical applications, please refer to Figure 5 As shown, when the real-time region area is 2401 (km²) 2 When the storage relationship function is 2401 (km), the x-axis is 2401. 2 When the vertical axis is 78.61 (m), the vertical axis is 78.61 (m). 3 The real-time water storage capacity is 78.61 (m³). 3 ), will 2401 (km) 2 ) and 78.61 (m 3 This indicates the area and water storage capacity of the lake.

[0028] Example 2: This application also provides an electronic device, which may include: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus. The memory stores computer-readable instructions, and the processor can call the instructions in the memory. When the computer-readable instructions are executed by the processor, the steps in the method for dynamic monitoring of the area and water storage of plateau lakes based on remote sensing image analysis are performed to achieve the following functions: acquiring RGB category images of plateau lakes from remote sensing images and labeling them as lake maps; converting the RGB values ​​of each pixel in the lake map into grayscale values ​​using a weighted average method to obtain a lake grayscale map; obtaining a first range threshold and a second range threshold based on a first number of historical lake maps; and based on the first range threshold, the second range threshold, and the lake grayscale map... The image obtains a first segmentation threshold and a second segmentation threshold; based on the first and second segmentation thresholds, the lake grayscale image is binarized to obtain a lake binarized image; the main analysis region is obtained based on the lake binarized image; the initial contour is dilated and optimized to obtain the final analysis region; the real-time region area is obtained based on the final analysis region; the relationship function between the region area and the water storage is obtained based on a second number of historical lake images and the corresponding water storage, and is marked as the water storage relationship function; the real-time region area is substituted into the water storage relationship function to obtain the real-time water storage; the real-time region area and the real-time water storage are displayed as the lake area and water storage.

[0029] Furthermore, when the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0030] Example 3: This application also provides a computer program product, which includes a computer program stored on a computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the dynamic monitoring method for the area and water storage of plateau lakes based on remote sensing image analysis provided by the above methods. The method includes: acquiring RGB category images from remote sensing images of plateau lakes and labeling them as lake maps; converting the RGB values ​​of each pixel in the lake map into grayscale values ​​using a weighted average method to obtain a lake grayscale map; and obtaining a first range threshold and a second range threshold based on a first number of historical lake maps. First and second segmentation thresholds are obtained based on a first range threshold, a second range threshold, and a lake grayscale image. The lake grayscale image is then binarized using these thresholds to obtain a binarized lake image. The main analysis region is obtained from the binarized lake image. The initial contour is expanded to obtain the final analysis region. The real-time region area is obtained from the final analysis region. A relationship function between the region area and water storage is obtained based on a second number of historical lake images and their corresponding water storage volumes, and this function is labeled as the water storage relationship function. The real-time region area is substituted into the water storage relationship function to obtain the real-time water storage volume. The real-time region area and real-time water storage volume are then displayed as the lake's area and water storage volume.

[0031] Example 4: This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program performs the steps described above in the method for dynamic monitoring of the area and water storage of plateau lakes based on remote sensing image analysis, to achieve the following functions: acquiring RGB category images from remote sensing images of plateau lakes and labeling them as lake maps; converting the RGB values ​​of each pixel in the lake map into grayscale values ​​using a weighted average method to obtain a lake grayscale map; obtaining a first range threshold and a second range threshold based on a first number of historical lake maps; and based on the first range threshold and the second range threshold... The process involves: obtaining a first and second segmentation threshold from the lake's grayscale image; binarizing the lake's grayscale image based on the first and second segmentation thresholds to obtain a binarized lake image; obtaining the main analysis region based on the binarized lake image; performing dilation optimization on the initial contour to obtain the final analysis region; obtaining the real-time region area based on the final analysis region; obtaining the relationship function between the region area and the water storage based on a second number of historical lake images and their corresponding water storage, and labeling it as the water storage relationship function; substituting the real-time region area into the water storage relationship function to obtain the real-time water storage; and displaying the real-time region area and real-time water storage as the lake's area and water storage.

[0032] Based on the above description of the embodiments, the embodiments of the present invention can be provided as methods, systems, or computer program products. Based on this understanding, the above technical solutions, in essence or in terms of their contribution to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or certain parts of the embodiments.

[0033] In the embodiments provided in this application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces. The indirect coupling or communication connection between systems, modules, and units may be electrical, mechanical, or other forms.

[0034] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions 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 this application.

Claims

1. A method for dynamic monitoring of the area and water storage of plateau lakes based on remote sensing image analysis, characterized in that, Includes the following steps: Obtain RGB category images from remote sensing images of plateau lakes and label them as lake maps; The RGB values ​​of each pixel in the lake image are converted into grayscale values ​​using a weighted average method to obtain a grayscale image of the lake. A first range threshold and a second range threshold are obtained based on a first number of historical lake maps; The first segmentation threshold and the second segmentation threshold are obtained based on the first range threshold, the second range threshold, and the lake grayscale image; The lake grayscale image is binarized based on the first segmentation threshold and the second segmentation threshold to obtain the lake binarized image; The main analysis region is obtained based on the lake binarization map; The initial contour is dilated and optimized to obtain the final analysis region; Obtain the real-time area of ​​the region based on the final analysis region; Based on the second number of historical lake maps and the corresponding water storage, the relationship function between the area and water storage is obtained and marked as the water storage relationship function; Substitute the real-time area into the water storage relationship function to obtain the real-time water storage volume; The real-time area and real-time water storage are used to display the area and water storage of the lake.

2. The method for dynamic monitoring of plateau lake area and water storage based on remote sensing image analysis according to claim 1, characterized in that, Obtaining the first range threshold and the second range threshold based on a first number of historical lake maps includes the following sub-steps: Obtain the first number of historical lake images, and then calibrate the pixels of the lake portion in the historical lake images to obtain the calibrated pixels; The RGB values ​​of each pixel in the historical lake image are converted into grayscale values ​​using a weighted average method to obtain a historical grayscale image.

3. The method for dynamic monitoring of plateau lake area and water storage based on remote sensing image analysis according to claim 2, characterized in that, Obtaining the first range threshold and the second range threshold based on a first number of historical lake maps also includes the following sub-steps: Obtain the minimum and maximum gray values ​​of the labeled pixel after deleting the minimum and maximum gray values ​​of a labeled pixel, respectively, and mark them as the first range threshold and the second range threshold.

4. The method for dynamic monitoring of plateau lake area and water storage based on remote sensing image analysis according to claim 3, characterized in that, Obtaining the first segmentation threshold and the second segmentation threshold based on the first range threshold, the second range threshold, and the lake grayscale image includes the following sub-steps: Obtain the grayscale value of each pixel in the grayscale image of the lake and label it as the lake's grayscale value; Divide the gray values ​​from 0 to 255 into p equal intervals, and mark them as gray value ranges; Count the frequency of lake grayscale values ​​within each grayscale division range and label it as the grayscale frequency of the lake division; A histogram is drawn with the lake's grayscale value on the X-axis, the frequency of the lake's grayscale values ​​on the Y-axis, and the grayscale range as the histogram interval. This histogram is then labeled as the lake's grayscale histogram.

5. The method for dynamic monitoring of plateau lake area and water storage based on remote sensing image analysis according to claim 4, characterized in that, Obtaining the first segmentation threshold and the second segmentation threshold based on the first range threshold, the second range threshold, and the lake grayscale image further includes the following sub-steps: The interval with the highest gray-level frequency in the lake gray-level histogram is obtained and marked as the peak interval. The median abscissa of the interval with the minimum gray frequency of the lake is obtained on the left and right sides of the peak interval, and marked as the first image threshold and the second image threshold. Calculate the average of the first range threshold and the first graphic threshold, and mark it as the first segmentation threshold; Calculate the average of the second range threshold and the second graphic threshold, and mark it as the second segmentation threshold.

6. The method for dynamic monitoring of plateau lake area and water storage based on remote sensing image analysis according to claim 5, characterized in that, Binarizing the lake grayscale image based on the first and second segmentation thresholds to obtain the lake binarized image includes the following sub-steps: In the lake grayscale image, lake grayscale values ​​that are greater than or equal to the first segmentation threshold and less than or equal to the second segmentation threshold are set to 0, and lake grayscale values ​​that are less than the first segmentation threshold or greater than the second segmentation threshold are set to 255 to obtain a lake binarized image.

7. The method for dynamic monitoring of plateau lake area and water storage based on remote sensing image analysis according to claim 6, characterized in that, Obtaining the main analysis region based on the lake binarization map includes the following sub-steps: Mark the pixels with a grayscale value of 0 in the binarized lake image as the initial pixels; Starting from any initial pixel, search for a pixel in each of the eight directions and mark the found pixel as the search pixel; the eight directions are up, down, left, right, upper left, lower left, upper right, and lower right. If the search pixel contains a new initial pixel, new search pixels are obtained in eight directions from the new initial pixel. If the new search pixel contains a new initial pixel, the search is repeated with the new initial pixel until the new search pixel does not contain a new initial pixel. Then, it is considered that one search is completed, and the area formed by the initial pixels obtained after one search is marked as the area to be analyzed. The region with the most pixels is identified and marked as the main analysis region.

8. The method for dynamic monitoring of plateau lake area and water storage based on remote sensing image analysis according to claim 7, characterized in that, Obtaining the final analysis region by dilating the initial contour includes the following sub-steps: Initial pixels that are not within the main analysis area are marked as discrete pixels; Obtain the edge pixels of the main analysis region and mark them as main edge pixels. Construct a rectangle with a side length j times the pixel width, centered on each main edge pixel, and mark it as an expansion rectangle. Determine if the expansion rectangle contains discrete pixels. If it does, mark the discrete pixels within the expansion rectangle as expansion pixels. Connect the expansion pixels to the nearest main edge pixel to obtain the first line segment. Then, construct a new expansion rectangle centered on the expansion pixel and determine if the new expansion rectangle contains new discrete pixels. Connect the new discrete pixels to the expansion pixels to obtain the second line segment. Repeat the construction of the expansion matrix and connecting line segments until the new expansion rectangle does not contain any new discrete pixels. For consecutive line segments up to the nth line segment, j is an odd number greater than 3. Mark all pixels on line segments 1 to n as incremental pixels; The final analysis area is formed by combining all incremental pixels and the main analysis area.

9. The method for dynamic monitoring of plateau lake area and water storage based on remote sensing image analysis according to claim 8, characterized in that, Obtaining the real-time region area based on the final analysis region includes the following sub-steps: Obtain the number of pixels in the final analysis area and mark it as the final analysis count; obtain the area of ​​each pixel. Obtain the scaling ratio of the lake image compared to the actual size of the objects, and label it as the image scaling ratio; The real-time region area is calculated as: Ss = Z * Mj / x; where Ss is the real-time region area, Z is the final number of analyses, Mj is the area of ​​each pixel, and x is the image reduction ratio.

10. The method for dynamic monitoring of plateau lake area and water storage based on remote sensing image analysis according to claim 9, characterized in that, Based on the second set of historical lake maps and their corresponding water storage, the relationship function between the area and water storage of the region is obtained and labeled as the water storage relationship function. This includes the following sub-steps: Obtain the area of ​​the final analysis region of the second number of historical lake maps, and mark it as the historical region area; A plane rectangular coordinate system is established with the historical area as the horizontal axis and the water storage volume as the vertical axis, and it is marked as the water storage relationship coordinate system. By using the historical area and corresponding water storage volume as the x and y coordinates of the water storage coordinate points, and fitting all water storage coordinate points together, a water storage relationship function is obtained.

Images