Water area extraction method, device and product based on radar remote sensing image

Abstract

The invention relates to a water area extraction method and device based on a radar remote sensing image, a computer program product and electronic equipment, and belongs to the field of remote sensing monitoring. The water area extraction method comprises the following steps: acquiring a radar remote sensing image, an optical remote sensing image and topographic data of a water area to be extracted; preprocessing the radar remote sensing image and the optical remote sensing image; carrying out adaptive filtering processing on the preprocessed radar remote sensing image; calculating a water body index image of the to-be-extracted water area, and obtaining a preliminary extraction result of the to-be-extracted water area according to the water body index image; and according to a classifier constructed by adding a gradient condition and a vegetation coverage condition, carrying out reclassification on the preliminary extraction result to obtain a final extraction result of the water area to be extracted. According to the scheme, the radar image water area extraction precision can be improved, the influence of mountain shadow and low shrub vegetation noise on the water area boundary is reduced, and the intelligent level of water area information monitoring is improved.

Description

Technical Field

[0001] The present invention relates to the field of remote sensing monitoring technology, and in particular to a water area extraction method, extraction device, computer program product and electronic equipment based on radar remote sensing images. Background Art

[0002] The ecological environment is the foundation of human survival and development. Water ecological security is a crucial guarantee and requirement for achieving the national goal of "fundamental improvement in ecological environment quality and the development of an ecological civilization" by 2035, and a crucial foundation for building a "Beautiful China." Lakes are crucial vehicles for the water cycle in regional ecosystems, playing an indispensable role in the circulation of matter and energy in aquatic and terrestrial ecosystems. Lake area is the foundation of lake hydrological research. Accurately and rapidly monitoring the dynamic boundaries of lakes is not only a part of scientific research but also a prerequisite for effective policymaking and promoting sustainable ecological and environmental development.

[0003] Traditional hydrological information acquisition relies primarily on ground-based hydrological monitoring stations and manual field monitoring. While ground-based monitoring offers advantages in terms of high data accuracy and stable quality, it is limited by labor costs and the number of deployed sites, making it inadequate for the high-frequency, spatiotemporal monitoring of water bodies. Satellite remote sensing technology, with its short imaging cycle, wide coverage, and high spatiotemporal resolution, has played a significant role in water monitoring and has reduced its costs to a certain extent. However, due to the poor penetration of optical wavelengths and their susceptibility to cloud cover and water vapor, effective image acquisition in cloudy and rainy areas is challenging. Radar remote sensing, as an active radar system, is less affected by lighting and meteorological conditions and can penetrate clouds for all-day, all-weather imaging. This provides a richer and more effective data source for areas with low optical image quality. Furthermore, due to the smooth surface and low backscattering of water bodies, these areas appear predominantly dark in radar images, clearly distinguishable from land features. Currently, radar satellites in the United States and Japan primarily operate in the L-band, while those in Europe, Canada, and China primarily operate in the C- and X-bands. Applications of the S-band (frequency approximately 2-4 GHz, corresponding to wavelengths approximately 7.5-15 cm) to water body extraction remain largely unsuccessful. While the S-band is not as mature as the X-, C-, and L-bands, it exhibits less attenuation in rainy areas and can more reliably penetrate precipitation clouds compared to higher-frequency bands like the X-band, improving data reliability in complex weather conditions. Compared to lower-frequency bands like the L-band, its penetration is slightly lower, but its spatial resolution is improved, making the S-band a compromise in performance. The most commonly used method for water feature recognition in radar images is the threshold method, which uses statistical analysis of grayscale histogram features in the image to select the optimal threshold for efficient water body identification. However, when identifying water bodies over large areas, this method relies heavily on thresholds determined by the image intensity histogram. This, coupled with increased radar image noise, results in a high probability of error in automated water body identification. Summary of the Invention

[0004] Aiming at the shortcomings of the threshold method for extracting water areas from radar images, and considering the disadvantage that S-band radar remote sensing images are easily disturbed by shallows, wetlands, shrubs and other low vegetation, the present invention provides a water area extraction method based on radar remote sensing images. The method uses an adaptive speckle denoising anisotropic diffusion (ASRAD) filtering algorithm to suppress speckle noise in the image, calculates the water body index of the filtered radar remote sensing image, and constructs a classifier by adding slope and vegetation coverage conditions to reduce the influence of mountain shadows and low vegetation confusion, thereby achieving rapid and accurate extraction of the water area.

[0005] The basic principle of this method is that when optical remote sensing is obscured by cloud and fog, or when the optical image contains indistinguishable small water bodies or water bodies with similar colors to surrounding ground objects, the unique scattering properties of water bodies in Synthetic Aperture Radar (SAR) imagery make their boundaries easier to identify and extract. Water bodies have relatively smooth surfaces and are weak scatterers in SAR images, typically appearing as dark areas with low grayscale values, significantly different from surrounding ground objects. Using 5-meter high-resolution SAR imagery provides more detailed water information, but it also increases the impact of speckle noise and hill shadow on water boundary extraction. Therefore, a speckle noise suppression filter is used to remove noise interference. Then, by analyzing terrain and vegetation cover, low grayscale areas of hill shadow and low shrub vegetation are removed to extract the true water boundary. Therefore, this water area extraction method from S-band radar imagery based on speckle noise suppression can simultaneously reduce the interference of noise and hill shadow on the extraction results, achieving automated and high-precision water boundary extraction.

[0006] The solution proposed by the present invention overcomes the drawback of existing remote sensing water area information extraction, which is unable to accurately extract small ponds and small rivers. The advantage of this invention lies in that the water area extraction method based on radar remote sensing imagery not only compensates for the shortage of effective optical imagery during rainy weather, but also improves the accuracy of radar image water area extraction, reduces the impact of mountain shadows and low shrubby vegetation noise on water area boundaries, enhances the intelligent level of water area information monitoring, and realizes comprehensive, objective, efficient, and dynamic supervision of aquatic and terrestrial ecosystems.

[0007] In a first aspect, an embodiment of the present invention provides a water area extraction method based on radar remote sensing images, the method comprising: Obtain radar remote sensing images, optical remote sensing images and terrain data of the water area to be extracted; Preprocessing the radar remote sensing image and the optical remote sensing image; performing adaptive filtering processing on the pre-processed radar remote sensing image; Calculating a water index image of the water area to be extracted based on the radar remote sensing image after filtering, and obtaining a preliminary extraction result of the water area to be extracted based on the water index image; The preliminary extraction results are reclassified according to a classifier constructed by adding slope conditions and vegetation coverage conditions to obtain a final extraction result of the water area to be extracted.

[0008] In some embodiments, the radar remote sensing image is an S-band synthetic aperture radar remote sensing image with a resolution of 5 meters.

[0009] In some embodiments, the adaptive filtering process is an adaptive speckle denoising anisotropic diffusion filtering process.

[0010] In some embodiments, the filtering process includes: Calculating a backscatter coefficient image of the preprocessed radar remote sensing image; An adaptive speckle denoising anisotropic diffusion filtering process is performed on the backscatter coefficient image.

[0011] In some embodiments, obtaining a preliminary extraction result of the water area to be extracted based on the water body index image includes: Generating a water body index statistical histogram of the water area to be extracted according to the water body index image; According to the water body index statistical histogram and the water area type judgment threshold, a preliminary extraction result of the water area to be extracted is obtained.

[0012] In some embodiments, the classifier constructed by adding the slope condition and the vegetation coverage condition reclassifies the preliminary extraction result to obtain the final extraction result of the water area to be extracted, including: The terrain data is digital elevation model data, and the slope distribution image of the water area to be extracted is calculated based on the digital elevation model data; Calculating the vegetation coverage image of the water area to be extracted based on the preprocessed optical remote sensing image; The preliminary extraction results are reclassified according to the slope distribution image and the vegetation coverage image to obtain a final extraction result of the water area to be extracted.

[0013] In a second aspect, an embodiment of the present invention provides a water area extraction device based on radar remote sensing images, the device comprising: Image data acquisition module, used to obtain radar remote sensing images, optical remote sensing images and terrain data of the water area to be extracted; A preprocessing module, used for preprocessing the radar remote sensing image and the optical remote sensing image; A filtering processing module, used for performing adaptive filtering processing on the pre-processed radar remote sensing image; a preliminary extraction result acquisition module, configured to calculate a water index image of the water area to be extracted based on the radar remote sensing image after filtering, and obtain a preliminary extraction result of the water area to be extracted based on the water index image; The final extraction result acquisition module is used to reclassify the preliminary extraction results based on a classifier constructed by adding slope conditions and vegetation coverage conditions to obtain the final extraction results of the water area to be extracted.

[0014] In some embodiments, the radar remote sensing image is an S-band synthetic aperture radar remote sensing image with a resolution of 5 meters.

[0015] In a third aspect, some embodiments of the present invention provide a computer program product, comprising computer program instructions, which, when read and executed by a processor, execute the method described in any one of the embodiments of the first aspect.

[0016] In a fourth aspect, some embodiments of the present invention provide an electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the program, the method described in any one of the embodiments of the first aspect can be implemented.

[0017] The proposed water area extraction scheme based on radar remote sensing imagery leverages the advantages of satellite remote sensing technology, such as short imaging cycles, wide coverage, and high spatiotemporal resolution, thereby reducing the cost of water monitoring to a certain extent. Furthermore, by leveraging radar satellite remote sensing's cloud-penetrating capabilities for all-day, all-weather imaging, it further addresses the difficulty of optical remote sensing in effectively acquiring image data for monitoring in cloudy and rainy areas, thereby improving the effectiveness of water area monitoring data acquisition. This method has a clear theoretical and technical basis and employs an adaptive ASRAD algorithm to suppress noise in radar images. The algorithm automatically adjusts the number of iterations based on edge information, minimizing noise while ensuring the integrity of land and water boundaries. The radar image water index method employed enhances water characteristics while mitigating interference from vegetation and soil. By incorporating slope and vegetation cover into a decision tree classifier, the effects of hill shadows and the susceptibility of S-band radar images to interference from low vegetation are mitigated. This water area extraction method features clear model parameter relationships, a clear theoretical basis, a practical solution, and highly accurate monitoring results. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments of the present invention. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 A flowchart of a water area extraction method based on radar remote sensing images provided by an embodiment of the present invention; Figure 2 A flowchart of obtaining preliminary water area extraction results based on filtered radar remote sensing images provided by an embodiment of the present invention; Figure 3 A flow chart of reclassifying the preliminary water area extraction results to obtain the final water area extraction results provided by an embodiment of the present invention; Figure 4 A block diagram of a water area extraction device based on radar remote sensing images provided by an embodiment of the present invention; Figure 5 A schematic diagram of the composition of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the embodiments of the present invention will be further described below. It should be noted that the embodiments of the present invention and the features therein can be combined with each other without conflict.

[0021] The following description sets forth many specific details to facilitate a full understanding of the present invention, but the present invention can also be implemented in other ways than those described herein. Obviously, the embodiments in the specification are only some of the embodiments of the present invention, not all of them.

[0022] In the present invention, the water area extraction method and extraction device are described by way of example using lakes as an example. Those skilled in the art will appreciate that the water area extraction method and extraction device of the present invention can also be applied to extract water from other water areas or river basins, such as rivers, reservoirs, and ponds.

[0023] Figure 1 An embodiment of the present invention provides a method for extracting water areas based on radar remote sensing images, the method comprising the following steps: Step S1, obtaining radar remote sensing images, optical remote sensing images and terrain data of the water area to be extracted; Step S2, preprocessing the radar remote sensing image and the optical remote sensing image; Step S3, performing adaptive filtering processing on the pre-processed radar remote sensing image; Step S4, calculating a water index image of the water area to be extracted based on the radar remote sensing image after filtering, and obtaining a preliminary extraction result of the water area to be extracted based on the water index image; Step S5: reclassify the preliminary extraction result according to the classifier constructed by adding the slope condition and the vegetation coverage condition to obtain the final extraction result of the water area to be extracted.

[0024] In this embodiment, an adaptive filtering algorithm is used to suppress speckle noise in the image, and the water index of the filtered radar remote sensing image is calculated. By adding slope conditions and vegetation coverage conditions to construct a classifier, the influence of mountain shadows and low shrub vegetation noise is reduced, thereby achieving rapid and accurate extraction of the water area, improving the intelligent level of water area information monitoring, and realizing comprehensive, objective, efficient and dynamic supervision of aquatic and terrestrial ecosystems.

[0025] In some embodiments of the present invention, step S1 exemplarily includes steps S11 to S13: Step S11: obtaining a radar remote sensing image of the water area to be extracted.

[0026] For example, this embodiment downloads the spaceborne space radar remote sensing image of the water area to be extracted from the land observation satellite data service platform. The present invention intends to obtain a spaceborne S-band synthetic aperture radar remote sensing image P with a spatial resolution of 5 meters and radar signals of vertical polarization and cross polarization. The vertical polarization radar component image of the synthetic aperture radar remote sensing image P is recorded as P vv , the cross-polarization radar component image is recorded as P vh Radar remote sensing image P, vertical polarization radar component image P vv and cross-polarization radar component image P vh The image size is m rows and n columns.

[0027] Step S12: obtaining an optical remote sensing image of the water area to be extracted.

[0028] For example, in this embodiment, a multispectral optical remote sensing image with high spatial resolution is downloaded from the land observation satellite data service platform. The spectral band of the optical remote sensing image includes the visible light and near-red band range, and the optical remote sensing image is recorded as L. Among them, the red light band component image of the optical remote sensing image L is recorded as L red , the near-red band component image is recorded as L nir Optical remote sensing image L, red light band component image L of optical remote sensing image red and the near-red band component image L of the optical remote sensing image nirThe image size is m rows and n columns.

[0029] Step S13: obtaining the terrain data of the water area to be extracted.

[0030] Illustratively, in this embodiment, digital elevation model (DEM) data of the water area to be extracted with a spatial resolution of 30 meters is downloaded from the Google Earth Engine (GEE) platform, and the DEM data is recorded as D.

[0031] In this embodiment, the platforms for downloading radar remote sensing images, optical remote sensing images, and digital elevation model data are exemplarily implemented by the land observation satellite data service platform and the GEE platform, respectively. Those skilled in the art should understand that the platform can also be implemented by other public data platforms with corresponding functions.

[0032] It should be noted that, in this embodiment, there is no requirement for the execution order of steps S11 to S13, that is, these steps can be performed simultaneously or in any order.

[0033] In some embodiments of the present invention, step S2 exemplarily includes steps S21 to S22: In step S21, the radar remote sensing image of the water area to be extracted is preprocessed.

[0034] For example, this embodiment performs preprocessing operations such as multi-view, radiation correction, geometric correction, and orthorectification on the synthetic aperture radar remote sensing image P, and records the preprocessing result of the radar remote sensing image as P'. The preprocessed radar remote sensing image P' includes the vertical polarization radar component image P vv ' and cross-polarization radar component image P vh '. Preprocessed radar remote sensing image P', vertical polarization radar component image P vv ' and cross-polarization radar component image P vh 'The image size is m rows and n columns.

[0035] In step S22, the optical remote sensing image of the water area to be extracted is preprocessed.

[0036] For example, this embodiment preprocesses the optical remote sensing image, including operations such as radiation correction and geometric correction, and resamples it to the same spatial resolution as the radar remote sensing image. The preprocessed optical remote sensing image is recorded as L', and the preprocessed optical remote sensing image red light band component image is recorded as L red ', record the pre-processed near-red band component image of the optical remote sensing image as L nir'. Preprocessed optical remote sensing image L', red light band component image L of optical remote sensing image red ' and the near-red band component image L of the optical remote sensing image nir 'The image size is m rows and n columns.

[0037] It should be noted that, in this embodiment, there is no requirement for the execution order of steps S21 to S22, that is, these steps can be performed simultaneously or in any order.

[0038] In some embodiments of the present invention, step S3 exemplarily includes steps S31 to S32: In step S31, the backscatter coefficient image of the pre-processed radar remote sensing image is calculated.

[0039] For example, in this embodiment, the following formula is used to calculate the backscatter coefficient image P' of the pre-processed radar remote sensing image P': db :

[0040] Among them, (i, j) is the position of the pixel in the image, i is the row number of the image involved in the calculation, and the value range of i is between 1 and m, and j is the column number of the image involved in the calculation, and the value range of j is between 1 and n. is the pixel amplitude in the preprocessed radar remote sensing image P', is the maximum value of the radar remote sensing image P' before quantization, is the calibration constant of radar remote sensing image P', and All of these can be obtained through image metadata files.

[0041] Backscatter coefficient image P' db The vertical polarization backscatter coefficient component image is recorded as P' dbvv , the cross-polarization backscatter coefficient component image is recorded as P' dbvh The backscatter coefficient image P' of the preprocessed radar remote sensing image P' db , vertical polarization backscatter coefficient component image P' dbvv and the cross-polarization backscatter coefficient component image P' dbvh The image size is m rows and n columns.

[0042] In step S32 , an adaptive speckle denoising anisotropic diffusion filtering (ASRAD) process is performed on the backscatter coefficient image.

[0043] Specifically, it includes: Step S321, calculate the speckle reduction anisotropic diffusion (SRAD) filter numerical approximation of each pixel of the radar remote sensing image of the water area to be extracted .

[0044] in, is the normalized backscatter coefficient image, , express Normalized result of backscatter coefficient value corresponding to actual ground point, The normalized value ranges from 0 to 255.

[0045] is the time step, and its value is 0.05. is the spatial step length, which is set to 1.

[0046] For the The pixel at the iteration The diffusion coefficient,

[0047] in, and They are The gradient and Laplacian operator,

[0048] is the instantaneous diffusion coefficient,

[0049] is a smooth scaling function, which can be approximated as .in, is the noise reduction index, the value is 1; is the initial coherent speckle diffusion coefficient, , the weaker the spot correlation, The smaller the value. and At the edge center, the maximum and zero crossing are respectively experienced, so Can be used as an edge detector for images.

[0050] Step S322: Backscatter coefficient image P' db Perform SRAD filtering to obtain the filtered radar remote sensing image F db . Radar remote sensing image F db Including vertical polarization radar component image F dbvvand cross-polarization radar component image F dbvh SRAD filtered radar remote sensing image F db 、Radar remote sensing image F db The vertical polarization radar component image F dbvv and cross-polarization radar component image F dbvh The image size is m rows and n columns.

[0051] Step S323: Based on the SRAD filter image processing result, The index serves as a quantitative standard to adaptively control its iterative process, that is, to implement ASRAD filtering processing. The index represents the edge similarity between the original image and the filtered image to evaluate the image edge preservation ratio.

[0052]

[0053] in, is the normalized backscattering coefficient image The edge image, is the radar remote sensing image F after SRAD filtering db The edge image, and is the normalized backscattering coefficient image and the radar remote sensing image F after SRAD filtering db The average pixel value. The calculation formula is

[0054] when The closer the index is to 1, the better the image edge preservation is. When the iteration effect satisfies the formula The iteration terminates when is the number of iterations, is the threshold, The smaller it is, the smoother the image is and the more information is lost. A denoised image with high-intensity contrast edges can be obtained by increasing the value.

[0055] After the above-mentioned ASRAD filtering process, the radar remote sensing image F' is obtained. db , the radar remote sensing image F' db Including vertical polarization radar component image F' dbvv and cross-polarization radar component image F' dbvh Radar remote sensing image F' after ASRAD filtering db , vertical polarization radar component image F' dbvv and cross-polarization radar component image F' dbvhThe image size is m rows and n columns.

[0056] like Figure 2 As shown, in some embodiments of the present invention, step S4 exemplarily includes steps S41 to S43: In step S41, the water index image of the water area to be extracted is calculated based on the radar remote sensing image after ASRAD filtering.

[0057] For example, in this embodiment, the vertical polarization radar component image F' of the radar remote sensing image after ASRAD filtering is used. dbvv and cross-polarization radar component image F' dbvh , use the following formula to calculate the radar remote sensing image F' after ASRAD filtering db The water index of each pixel point in

[0058] And generate the water index image SDWI of the water area to be extracted. In the water index image SDWI of the water area to be extracted, the value of each pixel is the water index calculation result of the pixel Where (i, j) is the position of the pixel in the image, i is the row number of the image involved in the calculation, and its value range is between 1 and m, and j is the column number of the image involved in the calculation, and its value range is between 1 and n.

[0059] In step S42, a water index statistical histogram of the water area to be extracted is generated based on the water index image. In this histogram, the horizontal axis represents the calculated SDWI water index of the water area to be extracted, and the vertical axis represents the number of pixels corresponding to each SDWI statistical value in the SDWI water index image of the water area to be extracted.

[0060] In step S43, the threshold T is determined based on the water body index statistical histogram and the water area type. h , obtain the preliminary water body extraction result image of the water area to be extracted. Specifically, combined with the SDWI water body index histogram distribution result, the horizontal coordinate SDWI value corresponding to the lowest point of the curve shape between the two peaks in the histogram is determined as the water area type judgment threshold T for distinguishing water bodies from non-water body landform types. h . Further, generate the preliminary water body extraction result image W1, when ≥T h When , the pixel is extracted as water body, and the pixel of the preliminary water body extraction result image ,when <T h When the pixel is extracted as non-water body, the pixel of the preliminary water body extraction result image .

[0061] like Figure 3 As shown, in some embodiments of the present invention, step S5 exemplarily includes steps S51 to S53: In step S51 , a slope distribution image of the water area to be extracted is calculated based on the digital elevation model data of the water area to be extracted.

[0062] For example, in this embodiment, the DEM data D of the water area to be extracted is used to calculate the slope distribution image S of the water area to be extracted.

[0063] A 3*3 moving window is used on the DEM data D to calculate the slope distribution image S, where: f x is the rate of change of elevation in the north-south direction, f y is the rate of change of elevation in the east-west direction, (i, j) is the position of the pixel in the image, i is the row number of the image involved in the calculation, and the value range of i is between 1 and m, j is the column number of the image involved in the calculation, and the value range of j is between 1 and n.

[0064] In step S52, the vegetation coverage image of the water area to be extracted is calculated based on the pre-processed optical remote sensing image.

[0065] Specifically, it includes: Step S521, calculate the normalized vegetation index image NDVI of the water area to be extracted based on the pre-processed optical remote sensing image L'. In the normalized vegetation index image NDVI, the value of each pixel is the normalized vegetation index calculation result of the pixel. ,

[0066] Where (i, j) is the pixel's position in the image, i is the row number of the image involved in the calculation, ranging from 1 to m, and j is the column number of the image involved in the calculation, ranging from 1 to n. NDVI is used instead of vegetation information when calculating FVC using the pixel bisection method.

[0067] Step S522: Calculate the vegetation coverage image FVC. In the vegetation coverage image FVC, the value of each pixel is the calculated result of the vegetation coverage of the pixel. ,

[0068] Where (i, j) is the position of the pixel in the image, i is the row number of the image involved in the calculation, and the value range of i is between 1 and m, and j is the column number of the image involved in the calculation, and the value range of j is between 1 and n; and are the minimum and maximum values ​​of the Normalized Difference Vegetation Index (NDVI) image, respectively.

[0069] In step S53, the preliminary extraction results are reclassified according to the slope distribution image and the vegetation coverage image to obtain the final extraction results of the water area to be extracted.

[0070] In this embodiment's classifier, a decision tree uses thresholds for slope and vegetation cover features, starting from the root node and branching along the path until it reaches a leaf node, ultimately producing the final classification result. Using the decision tree's classification rules, the water extraction result (W1) is reclassified, yielding the reclassified water extraction result image (W2).

[0071] a. When the image hour, ; b. Image Slope classification of the area: and ≥P T When , the pixel is extracted as non-water body, and the pixel of the final water body extraction result image ; c. For images and <P T Vegetation coverage classification of the area: and <P T and ≥V T When , the pixel is extracted as non-water body, and the pixel of the final water body extraction result image ;when and <P T and <V T When , the pixel is extracted as water body, and the pixel of the final water body extraction result image is Among them, P T is the slope threshold, V T is the vegetation coverage threshold.

[0072] Figure 4 The embodiment of the present invention provides a water area extraction device 400 based on radar remote sensing images. Figure 1 The device 400 corresponds to the method embodiment and can execute the various steps involved in the above method embodiment. The specific functions of the device can be found in the description above. To avoid repetition, the detailed description is omitted here. The water area extraction device 400 based on radar remote sensing image includes the following modules: Image data acquisition module 410, used to acquire radar remote sensing images, optical remote sensing images and terrain data of the water area to be extracted; A preprocessing module 420 is used to preprocess the radar remote sensing image and the optical remote sensing image of the water area to be extracted; A filtering processing module 430 is used to perform adaptive filtering processing on the pre-processed radar remote sensing image; A preliminary extraction result acquisition module 440 is configured to calculate a water index image of the water area to be extracted based on the filtered radar remote sensing image, and obtain a preliminary extraction result of the water area to be extracted based on the water index image; The final extraction result acquisition module 450 is used to construct a classifier that adds slope conditions and vegetation coverage conditions, reclassify the preliminary extraction results, and obtain the final extraction results of the water area to be extracted.

[0073] Those skilled in the art should understand that the device provided in the embodiment of the present invention has the same implementation principle and technical effects as the aforementioned method embodiment. For the sake of brief description, any part not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned method embodiment. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the aforementioned devices and modules can all refer to the corresponding processes in the aforementioned method embodiment, and will not be repeated here.

[0074] The devices or modules described in the above embodiments may be implemented by computer chips or entities, or by products having certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, an in-vehicle human-computer interaction device, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0075] Some embodiments of the present invention provide a computer program product comprising computer program instructions, which, when read and executed by a processor, execute a method in any one of the embodiments of the water area extraction method based on radar remote sensing images.

[0076] like Figure 5 As shown, some embodiments of the present invention provide an electronic device 500, including a memory 510, a processor 520, and a computer program stored on the memory 510 and executable on the processor 520, wherein the processor 520 can implement a method in any one of the embodiments of the water area extraction method based on radar remote sensing images when reading and executing the program through a bus 530.

[0077] Processor 520 can process digital signals and can include various computing architectures, such as a complex instruction set computer architecture, a reduced instruction set computer architecture, or an architecture that implements a combination of multiple instruction sets. In some examples, processor 520 can be a microprocessor.

[0078] The memory 510 can be used to store instructions executed by the processor 520 or data related to the execution of instructions. These instructions and / or data may include code for implementing some or all of the functions of one or more modules described in the embodiments of the present invention. The processor 520 of the embodiments of the present invention can be used to execute the instructions in the memory 510 to implement the methods shown in the aforementioned embodiments. The memory 510 includes dynamic random access memory, static random access memory, flash memory, optical memory, or other memory known to those skilled in the art.

[0079] The proposed water area extraction scheme based on radar remote sensing imagery leverages the advantages of satellite remote sensing technology, such as short imaging cycles, wide coverage, and high spatiotemporal resolution, thereby reducing the cost of water monitoring to a certain extent. Furthermore, by leveraging radar satellite remote sensing's cloud-penetrating capabilities for all-day, all-weather imaging, it further addresses the difficulty of acquiring effective image data for monitoring in cloudy and rainy areas using optical remote sensing, thereby improving the effectiveness of water area monitoring data acquisition. The proposed method has a clear theoretical and technical basis and utilizes the ASRAD algorithm to suppress noise in radar images. The algorithm automatically adjusts the number of iterations based on edge information, minimizing noise while ensuring the integrity of the land-water boundary. The radar image water index method employed enhances water characteristics while mitigating interference from vegetation and soil. By incorporating slope and vegetation cover factors into a decision tree classifier, the impact of mountain shadows and low shrubby vegetation noise is mitigated. This water area extraction method features clear model parameter relationships, a clear theoretical basis, a practical solution, and highly accurate monitoring results.

[0080] In the several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions, and operations of the devices, methods, and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or part of the code, which contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the boxes can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, as well as the combination of boxes in the block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified functions or actions, or can be implemented using a combination of dedicated hardware and computer instructions.

[0081] In addition, the functional modules in the various embodiments of the present invention may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.

[0082] If the functions are implemented as software modules 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 the present invention, or the portion that contributes to the prior art, or the 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 instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include various media capable of storing program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0083] The above description is merely an embodiment of the present invention and is not intended to limit the scope of protection of the present invention. For those skilled in the art, the present invention may be subject to various modifications and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included in the scope of protection of the present invention. It should be noted that similar reference numerals and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined or explained in subsequent figures.

[0084] The above description is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

[0085] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

Claims

1. A water area extraction method based on radar remote sensing images, characterized in that: The method comprises: Obtain radar remote sensing images, optical remote sensing images and terrain data of the water area to be extracted; Preprocessing the radar remote sensing image and the optical remote sensing image; performing adaptive filtering processing on the pre-processed radar remote sensing image; Calculating a water index image of the water area to be extracted based on the radar remote sensing image after filtering, and obtaining a preliminary extraction result of the water area to be extracted based on the water index image; The preliminary extraction results are reclassified according to a classifier constructed by adding slope conditions and vegetation coverage conditions to obtain a final extraction result of the water area to be extracted.

2. The extraction method according to claim 1, wherein The radar remote sensing image is an S-band synthetic aperture radar remote sensing image with a resolution of 5 meters.

3. The extraction method according to claim 2, wherein The adaptive filtering process is an adaptive speckle denoising anisotropic diffusion filtering process.

4. The extraction method according to claim 3, wherein The filtering process includes: Calculating a backscatter coefficient image of the preprocessed radar remote sensing image; An adaptive speckle denoising anisotropic diffusion filtering process is performed on the backscatter coefficient image.

5. The extraction method according to claim 4, wherein Obtaining a preliminary extraction result of the water area to be extracted based on the water body index image includes: Generating a water body index statistical histogram of the water area to be extracted according to the water body index image; According to the water body index statistical histogram and the water area type judgment threshold, a preliminary extraction result of the water area to be extracted is obtained.

6. The extraction method according to claim 5, wherein The classifier constructed by adding the slope condition and the vegetation coverage condition reclassifies the preliminary extraction result to obtain the final extraction result of the water area to be extracted, including: The terrain data is digital elevation model data, and the slope distribution image of the water area to be extracted is calculated based on the digital elevation model data; Calculating the vegetation coverage image of the water area to be extracted based on the preprocessed optical remote sensing image; The preliminary extraction results are reclassified according to the slope distribution image and the vegetation coverage image to obtain a final extraction result of the water area to be extracted.

7. A water area extraction device based on radar remote sensing images, characterized in that: The device comprises: Image data acquisition module, used to obtain radar remote sensing images, optical remote sensing images and terrain data of the water area to be extracted; A preprocessing module, used for preprocessing the radar remote sensing image and the optical remote sensing image; A filtering processing module, used for performing adaptive filtering processing on the pre-processed radar remote sensing image; a preliminary extraction result acquisition module, configured to calculate a water index image of the water area to be extracted based on the radar remote sensing image after filtering, and obtain a preliminary extraction result of the water area to be extracted based on the water index image; The final extraction result acquisition module is used to reclassify the preliminary extraction results based on a classifier constructed by adding slope conditions and vegetation coverage conditions to obtain the final extraction results of the water area to be extracted.

8. The extraction device according to claim 7, characterized in that The radar remote sensing image is an S-band synthetic aperture radar remote sensing image with a resolution of 5 meters.

9. A computer program product, characterized in that The method comprises computer program instructions, which can implement the method according to any one of claims 1 to 6 when the computer program instructions are read and executed by a processor.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 6 can be implemented.

Images