Small and medium-sized reservoir water body area inversion correction method based on maximum similarity of SAR (Synthetic Aperture Radar) image contours
By generating the water surface profile set of reservoirs and determining the contour lines with the greatest similarity, the problem of misjudging water bodies in the mountain shadow in SAR images is solved, which significantly improves the accuracy and reliability of the area inversion of water bodies in small and medium-sized reservoirs, and improves the prediction accuracy of the hydrological model and the optimization effect of reservoir scheduling.
Patent Information
- Application Number
- CN202510218829.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-26
- Publication Date
- 2025-06-13
AI Technical Summary
When using SAR images to remotely sense water area, mountain shadows caused by rugged terrain are easily mistaken for water, resulting in mis-estimation of water area, reducing the accuracy and reliability of extraction results, especially on small and medium-sized reservoirs.
By generating the reservoir water surface profile set, using high-precision DEM data and SAR remote sensing images, the contour line with the greatest similarity to the reservoir water surface grid boundary is determined, and the water area is corrected to distinguish between mountain shadows and water bodies.
It significantly improves the accuracy and reliability of water area inversion in small and medium-sized reservoirs, reduces misjudgment, provides more accurate water area data, and improves the prediction accuracy of the hydrological model and the optimization effect of reservoir scheduling.
Smart Images
Figure CN120147885A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of hydrological remote sensing, and specifically relates to a method for inverting and correcting the water area of small and medium-sized reservoirs based on the maximum similarity of SAR image contours. Background Art
[0002] SAR (Synthetic Aperture Radar) images have won wide recognition in the field of water area inversion due to their ability to capture the unique radar reflection characteristics of water bodies and their all-weather and all-time operation advantages that are not restricted by weather changes and lighting conditions. Especially in application scenarios that require continuous and regular monitoring, such as reservoir management, the use of SAR technology is particularly common. However, when using SAR images for remote sensing of water areas, due to the unique side-looking slant range imaging mechanism of SAR images, complex phenomena such as perspective contraction, layover, and shadows will occur when encountering rugged terrain. Especially the shadow areas formed by mountains appear black or dark in SAR images, which is very similar to the appearance of water bodies because the backscattering coefficient of mountain shadows is close to that of water bodies. In this case, traditional threshold segmentation methods are prone to misidentifying mountain shadows as part of the water body when extracting water bodies, which will lead to incorrect estimation of the water area, especially having a greater impact on small and medium-sized reservoirs, thus reducing the accuracy and reliability of the water body extraction results.
[0003] To solve the problem of inaccurate water area inversion in SAR images, the quality of water body inversion can be improved by integrating multi-temporal or multi-source remote sensing data through image fusion technology. However, in practical applications, this method needs to ensure the matching of the time and space resolutions of different source remote sensing data, which often significantly reduces the amount of available data. In addition, image processing algorithms based on machine learning or deep learning can also be used to distinguish the differences between water bodies and shadows, but this requires a large amount of high-quality training data. For small and medium-sized reservoirs, due to their relatively small water surface area and relatively lack of relevant hydrological series data, this also limits the effective application of the above methods.
[0004] Therefore, how to make full use of remote sensing data to solve the problem of inaccurate water area inversion in SAR images in a simple and efficient manner is of great significance for improving the satellite's observation ability of the surface hydrological conditions, enhancing the accuracy of hydrological forecasting, and optimizing reservoir operation plans. At the same time, it can also provide more reliable data support for basin water resources management and flood control and drought relief decision-making. Summary of the Invention
[0005] The technical problem to be solved by the present invention is to provide a method for inverting and correcting the water area of small and medium-sized reservoirs based on the maximum similarity of SAR image contours, so as to accurately identify the water surface boundary of small and medium-sized reservoirs and improve the extraction accuracy of the water area.
[0006] To solve the above technical problems, the technical solution adopted by the present invention is: a method for inverting and correcting the water area of small and medium-sized reservoirs based on the maximum similarity of SAR image contours, including the following steps: S1. Data preparation and processing, including the following steps: S101. Data preparation: Prepare the initial data required for inverting and correcting the water area of small and medium-sized reservoirs; S102. Generate a set of water surface contour lines; S103. Extract the set of grid boundary points of the reservoir water surface; S2. Similarity analysis of the reservoir water surface contour to determine the contour line with the greatest similarity to the grid boundary of the reservoir water surface ; S3. Correction of the reservoir water area: Use the contour line determined in step S2 and the dam boundary line to segment the surface data converted from the initial reservoir water surface grid data. After removing the non-reservoir area water surface elements downstream of the dam boundary line, several surface elements are obtained. The area outside the contour line is considered as mountain shadow, and the area inside the contour line is considered as water body. Then, calculate the water area of the small and medium-sized reservoir corrected by the contour line.
[0007] In a preferred solution, in step S101, the initial data required for inverting and correcting the water area of small and medium-sized reservoirs includes two categories. One is the high-precision digital elevation data at the low water level of the reservoir, and the other is the initial reservoir water surface grid data obtained after preprocessing the SAR remote sensing image.
[0008] In a preferred solution, in step S102, using the high-precision digital elevation data, set the lower elevation limit h 1 and the upper limit h n according to the normal storage level and dead storage level of the small and medium-sized reservoir. Then, from the contour interval , generate a set of reservoir water surface contour lines, and the expression is: ; wherein, represents the reservoir water surface contour line with an elevation of , ; n is the number of contour lines in the set, .
[0009] In a preferred solution, in step S103, convert the initial reservoir water surface grid data into surface data, then convert the surface data into polyline elements, and then set the spacing of the polyline elements s to increase the density. After the polyline elements with increased density are converted into point elements, the set of points of the reservoir water surface grid boundary can be obtained, , mis the number of grid boundary points of the reservoir water surface.
[0010] In the preferred solution, in the step S2, the similarity between the water surface grid boundary and the contour line is characterized by the proportion of the number of points closer to the contour line in the water surface boundary point set, and iterative calculation is used to determine the contour line with the greatest similarity to the reservoir water surface grid boundary. .
[0011] In the preferred solution, the step S2 includes the following steps: S201. Calculate the similarity between the water surface grid boundary point set and the median contour line Calculate the similarity between the reservoir water surface grid boundary and the median contour line is the median elevation contour line among the potential contour lines of the reservoir water surface. , and at the initial calculation ; ; The similarity between the grid boundary and is calculated by the following formula: ; In the formula, , is the set of distances from each point in the water surface boundary point set P to the water surface contour line , where is the shortest straight-line distance from the water surface boundary point to the contour line ; is the similarity proximity threshold. When is less than this threshold, it can be considered that the water surface boundary point conforms to the contour line ; m is the number of grid boundary points of the reservoir water surface; S202. Calculate the similarity between the water surface grid boundary point set and the adjacent median contour line , and the expression is as follows: ; In the formula, , is the set of distances from each point in the water surface boundary point set P to the water surface contour line ; S203. Update and iterate When , the contour line with the greatest similarity to the reservoir water surface grid boundary is located in the potential contour line , and at this time: ; When , the contour line with the greatest similarity to the reservoir water surface grid boundary is located in the potential contour line within which: ; Repeat steps S201 and S202 until , determine the contour line with the greatest similarity to the reservoir water surface grid boundary .
[0012] In a preferred embodiment, step S3 includes the following steps: S301. Segmentation of reservoir water body surface data Use the contour line determined in step S2 and the dam boundary line to segment the surface data converted from the initial reservoir water surface grid data. After removing the non-reservoir water surface elements downstream of the dam boundary line, several surface element regions are obtained. The region outside the contour line is considered as mountain shadow, and the region inside the contour line is considered as water body; S302. Calculation of water body area.
[0013] In a preferred embodiment, in step S301, the set of segmented reservoir surfaces is: ; wherein, is the set of surface elements after removing the non-reservoir water surface elements downstream of the dam boundary line from the segmented water surface grid surface data.
[0014] In a preferred embodiment, in step S302, the area of the small and medium-sized reservoir water body corrected by the contour line is: ; wherein, k is the number of surface elements in the set , is the area of the th i surface element in the set.
[0015] The present invention also provides a computer-readable storage medium storing computer instructions for causing a computer to execute the above-described method for inverting and correcting the water body area of small and medium-sized reservoirs based on the maximum similarity of SAR image contours.
[0016] The method for inverting and correcting the water body area of small and medium-sized reservoirs based on the maximum similarity of SAR image contours provided by the present invention has the following beneficial effects: 1. The method of generating a reservoir water surface contour set using high-precision DEM data to correct the SAR image water surface inversion result can significantly improve the accuracy and reliability of the water body area inversion of small and medium-sized reservoirs, effectively distinguish mountain shadows and actual water bodies in SAR images, and reduce misjudgment.
[0017] 2. The present invention does not rely on the matching of multi-temporal or multi-source remote sensing data, reduces the dependence on training data, and is particularly suitable for small and medium-sized reservoir scenarios with relatively scarce data.
[0018] 3. By providing more accurate water area data, this method can improve the prediction accuracy of hydrological models, optimize reservoir operation plans, enhance water resource utilization efficiency and operation safety, and at the same time provide a scientific basis for basin water resource management and flood control and drought relief decision-making, enhancing the satellite's observation ability of surface hydrological conditions, thereby generating positive social and economic benefits in multiple fields such as water resource management, hydrological forecasting, reservoir operation, and flood control and drought relief. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] The present invention will be further described below in conjunction with the drawings and embodiments: Figure 1 is the overall flow chart of the method of the present invention; Figure 2 is the set of reservoir water surface boundary contours; Figure 3 is the initial reservoir water surface raster data; Figure 4 is the reservoir water surface boundary point feature; Figure 5 is the maximum similar contour line; Figure 6 is the set of segmented surface features. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0020] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0021] Embodiment 1: As Figure 1 shown, a method for inverting and correcting the water area of small and medium-sized reservoirs based on the maximum similarity of SAR image contours includes the following steps: S1. Data preparation and processing, including the following steps: S101. Data preparation: Prepare the initial data required for inverting and correcting the water area of small and medium-sized reservoirs. The initial data required for inverting and correcting the water area of small and medium-sized reservoirs includes two categories. One category is the high-precision digital elevation (DEM) data of the reservoir at low water level obtained by using the unmanned aerial vehicle laser point cloud generation technology or other accuracy improvement methods, and the other category is the initial reservoir water surface raster data obtained after a series of processing such as noise reduction and gray threshold extraction of the SAR remote sensing image.
[0022] S102. Generate a set of water surface contour lines: Using high-precision digital elevation data, set the lower elevation limit according to the normal storage level and dead storage level of small and medium-sized reservoirs h 1 and the upper limit h n . Then, from the contour interval , generate a set of reservoir water surface contour lines. The expression is: ; Among them, represents the reservoir water surface contour line with an elevation of , ; n is the number of contour lines in the set, .
[0023] In this embodiment, the lower elevation limit is set to 765m and the upper elevation limit is set to 785m according to the normal storage level and dead storage level of the reservoir, and the contour interval is set to 3m. The generated set of water surface contour lines is as shown in Figure 2 .
[0024] S103. Extract the set of grid boundary points of the reservoir water surface: Convert the initial grid data of the reservoir water surface into surface data, then convert the surface data into polyline features, and then set the spacing of the polyline features s for densification. After densification, the polyline features are converted into point features to obtain the set of points of the grid boundary of the reservoir water surface, , m is the number of grid boundary points of the reservoir water surface.
[0025] In this embodiment, the spacing s is set to 20m, Figure 3 is the initial grid data of the reservoir water surface, Figure 4 is the boundary point feature of the reservoir water surface after densification.
[0026] S2. Analyze the similarity of the reservoir water surface contour and determine the contour line with the greatest similarity to the grid boundary of the reservoir water surface : The similarity between the grid boundary of the water surface and the contour line is characterized by the proportion of the number of points closer to the contour line in the set of water surface boundary points. Use iterative calculation to determine the contour line with the greatest similarity to the grid boundary of the reservoir water surface .
[0027] Specifically, it includes the following steps: S201. Calculate the similarity between the set of grid boundary points of the water surface and the median contour line Calculate the similarity between the grid boundary of the reservoir water surface and the median contour line , is the median elevation contour line in the potential contour lines of the reservoir water surface, , and initially ; The similarity between the raster boundary and is calculated by the following formula: ; ; In the formula, , is the set of water surface boundary points P The distances from each point in to the water surface contour line where is the shortest straight-line distance from the water surface boundary point to the contour line is the similarity proximity threshold. When is less than this threshold, it can be considered that the water surface boundary point matches the contour line ; m is the number of reservoir water surface raster boundary point sets.
[0028] In this embodiment, the similarity proximity distance = 3m, and the number of reservoir water surface raster boundary point sets m = 230.
[0029] S202. Calculate the similarity between the water surface raster boundary point set and the adjacent median contour line , and the expression is as follows: ; In the formula, , is the set of water surface boundary points P The distances from each point in to the water surface contour line
[0030] S203. Update and iterate When , the contour line with the greatest similarity to the reservoir water surface raster boundary is located on the potential contour line . At this time: ; When , the contour line with the greatest similarity to the reservoir water surface raster boundary is located within the potential contour line . At this time: ; Repeat steps S201 and S202 until , and determine the contour line with the greatest similarity to the reservoir water surface raster boundary.
[0031] In this embodiment, the contour line with the greatest similarity to the reservoir water surface raster boundary determined by the reservoir water surface contour similarity analysis is as Figure 5 shown.
[0032] S3. Reservoir water body area correction: Use the contour line determined in step S2 and the dam boundary line to segment the surface data converted from the initial reservoir water surface raster data. After removing the non-reservoir water surface elements downstream of the dam boundary line, several surface elements are obtained.
[0033] Under the influence of mountain shadows, areas that are not actually water bodies are misidentified as water bodies, resulting in an overestimated water body area inversion result. Therefore, areas outside the contour line are considered mountain shadows, and areas inside the contour line are considered water bodies. Then, the water body area of small and medium-sized reservoirs corrected by the contour line is calculated.
[0034] Specifically, it includes the following steps: S301. Segmentation of reservoir water body surface data Use the contour line determined in step S2 and the dam boundary line to segment the surface data converted from the initial reservoir water surface raster data. After removing the non-reservoir water surface elements downstream of the dam boundary line, several surface elements are obtained. Areas outside the contour line are considered mountain shadows, and areas inside the contour line are considered water bodies; Then the set of segmented reservoir surfaces is: ; In the formula, is the set of surface elements after segmentation of the water surface raster surface data. In this embodiment, the surface elements after segmentation of the water surface raster surface data are as shown in Figure 6 .
[0035] S302. Calculation of water body area.
[0036] The water body area of small and medium-sized reservoirs corrected by the contour line is: ; In the formula, k is the number of surface elements in the set , is the area of the th surface element in the set i . The water body area of small and medium-sized reservoirs corrected by the contour line is 78864 m 2 , compared with 108675 m 2 before correction, it is more in line with the actual situation.
[0037] Embodiment 2: An embodiment of the present invention provides a computer-readable storage medium storing computer instructions for causing the computer to execute the method for inverting and correcting the water body area of small and medium-sized reservoirs based on the maximum similarity of SAR image contours described in Embodiment 1.
[0038] Among them, the storage medium can be a magnetic disk, an optical disc, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD), or a solid-state drive (SSD), etc.; the storage medium can also include a combination of the above-mentioned types of memories.
[0039] Those skilled in the art can easily understand that the above are only the preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.
Claims
1. A method for inverting and correcting the water area of small and medium-sized reservoirs based on the maximum similarity of SAR image contours, characterized in that: The following steps are involved: S1. Data preparation and processing, including the following steps: S101. Data preparation: prepare the initial data required for inversion and correction of water area of small and medium-sized reservoirs; S102, generating a water surface contour line set; S103, extracting a grid boundary point set of the reservoir water surface; S2. Similarity analysis of reservoir water surface contours to determine the contour line with the greatest similarity to the reservoir water surface grid boundary ; S3. Reservoir water area correction: Use the contour line determined in step S2 The initial reservoir water surface raster data is converted into surface data based on the dam boundary line, and the non-reservoir water surface elements downstream of the dam boundary line are removed to obtain several surface elements. The area outside the contour line is considered to be mountain shadow, and the area within the contour line is considered to be water body. Then the water area of small and medium-sized reservoirs corrected by the contour line is calculated.
2. According to claim 1, a method for inversion and correction of water area of small and medium-sized reservoirs based on maximum similarity of SAR image contours is characterized in that: In step S101, the initial data required for inversion and correction of the water area of small and medium-sized reservoirs include two types, one is the high-precision digital elevation data when the reservoir is at low water level, and the other is the initial reservoir water surface raster data obtained after SAR remote sensing image preprocessing.
3. The method for inversion and correction of water area of small and medium-sized reservoirs based on maximum similarity of SAR image contours according to claim 2 is characterized in that: In step S102, high-precision digital elevation data is used to set the lower limit of elevation according to the normal water level and dead water level of small and medium-sized reservoirs. h 1 and upper limit h n , and then by the equal height interval , generate the reservoir water surface contour set, the expression is: ； in, Indicates the elevation The water surface contour of the reservoir, ; n is the number of contour lines in the set, .
4. The method for inverting and correcting the water area of small and medium-sized reservoirs based on the maximum similarity of SAR image contours according to claim 2 is characterized in that: In step S103, the initial reservoir water surface grid data is converted into surface data, and then the surface data is converted into polyline elements, and then the polyline elements are set to have a spacing. s Densification is performed, and the densified polyline features are converted into point features to obtain the point set of the reservoir water surface grid boundary. , m is the number of grid boundary points on the reservoir surface.
5. The method for inversion and correction of water area of small and medium-sized reservoirs based on maximum similarity of SAR image contours according to claim 1 is characterized in that: In step S2, the similarity between the water surface grid boundary and the contour line is represented by the proportion of the number of points in the water surface boundary point set that are closer to the contour line, and an iterative calculation is used to determine the contour line with the greatest similarity to the reservoir water surface grid boundary. .
6. The method for inversion and correction of water area of small and medium-sized reservoirs based on maximum similarity of SAR image contours according to claim 5 is characterized in that: The step S2 includes the following steps: S201, calculate the similarity between the water surface grid boundary point set and the median contour line Calculate the grid boundary and median contour of the reservoir water surface The similarity of is the median elevation contour line in the reservoir water surface potential contour line, , when initially calculating ; Grid borders and Similarity Calculated by the following formula: ； In the formula, , is the water surface boundary point set P From each point to the water surface contour The distance set of The water surface boundary point To contour The shortest straight-line distance; is the similar proximity threshold, when When the water surface boundary point is smaller than this threshold, it can be considered that the water surface boundary point is close to the contour line. conform to; m is the number of grid boundary points of the reservoir surface; S202, calculate the similarity between the water surface grid boundary point set and the adjacent median contour line , the expression is as follows: ； In the formula, , is the water surface boundary point set P From each point to the water surface contour The distance set of ; S203, Update and Iteration when When the contour line with the greatest similarity to the reservoir water surface grid boundary is located on the potential contour line ,at this time: ； when When the contour line with the greatest similarity to the reservoir water surface grid boundary is located on the potential contour line Inside, at this time: ； Repeat steps S201 and S202 until , determine the contour line that is most similar to the reservoir water surface grid boundary .
7. The method for inversion and correction of water area of small and medium-sized reservoirs based on maximum similarity of SAR image contours according to claim 1 is characterized in that: The step S3 includes the following steps: S301, Reservoir water surface data segmentation Using the contour line determined in step S2 The initial reservoir water surface raster data is converted into surface data by segmenting the dam boundary line, and after removing the non-reservoir water surface elements downstream of the dam boundary line, several surface element areas are obtained. The areas outside the contour line are considered to be mountain shadows, and the areas within the contour line are considered to be water bodies. S302. Calculation of water body area.
8. The method for inverting and correcting the water area of small and medium-sized reservoirs based on maximum similarity of SAR image contours according to claim 7 is characterized in that: In step S301, the segmented reservoir surface set is: ； In the formula, It is a set of surface features after the water surface raster surface data is segmented and the non-reservoir water surface features downstream of the dam boundary line are removed.
9. The method for inversion and correction of water area of small and medium-sized reservoirs based on maximum similarity of SAR image contours according to claim 7 is characterized in that: In step S302, the water area of the small and medium-sized reservoirs after contour line correction is: ； In the formula, k For collection The number of polygon features within For collection Middle i The area of a polygon feature.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the method for inverting and correcting the water area of small and medium-sized reservoirs based on maximum similarity of SAR image contours as described in any one of claims 1 to 9.