Satellite remote sensing image processing method, device and equipment and storage medium

Abstract

This application discloses a method, apparatus, device, and storage medium for processing satellite remote sensing images, relating to the field of remote sensing image processing technology. The method includes: acquiring satellite remote sensing images; determining a gradient compensation operator for each pixel position in the panchromatic image; determining the local texture complexity based on each band of the multispectral image, and determining an adjustment coefficient for each pixel position in the multispectral image based on the local texture complexity; fusing each band of the multispectral image using the gradient compensation operator and the adjustment coefficient, and determining the processed satellite remote sensing image based on the pixels of each fused band. Because an edge enhancement term, weighted by the gradient compensation operator and the adjustment coefficient, is superimposed while maintaining the original pixel values ​​of the multispectral image, the proportional relationships between the original color channels are fully preserved, enabling the fused image to accurately reflect the original colors of ground features and avoiding spectral distortion.

Description

Technical Field

[0001] This application relates to the field of remote sensing image processing technology, and in particular to methods, apparatus, equipment and storage media for processing satellite remote sensing images. Background Technology

[0002] With the rapid development of high-resolution Earth observation technology, the domestically produced Gaofen-2 (GF-2) satellite has been widely used in fields such as land resource surveys, agricultural monitoring, urban planning, and disaster assessment. The GF-2 satellite carries both a panchromatic camera and a multispectral camera, enabling it to acquire high-resolution panchromatic images (spatial resolution better than 1 meter) and lower-resolution multispectral images (spatial resolution approximately 4 meters) of the same area. To fully utilize the complementary information of the two types of images, it is usually necessary to fuse the high-resolution panchromatic image with the low-resolution multispectral image to obtain a high-resolution color image for subsequent interpretation and analysis.

[0003] However, when performing image fusion, commonly used fusion algorithms (such as IHS transform) directly replace the high-frequency details of panchromatic images with the brightness components of multispectral images. Since the brightness distribution of panchromatic and multispectral images is inconsistent, the fusion result will have obvious color distortion. Wavelet fusion directly superimposes high-frequency details, which can easily lead to edge blurring. Summary of the Invention

[0004] The main purpose of this application is to provide a method, apparatus, device and storage medium for processing satellite remote sensing images, which aims to solve the technical problem that traditional algorithms simply add high-frequency details of panchromatic images to multispectral images, which easily leads to color (spectral) distortion and blurred edge textures at high resolution.

[0005] To achieve the above objectives, this application proposes a method for processing satellite remote sensing images, the method comprising: Acquire satellite remote sensing images, including panchromatic images and multispectral images; A gradient compensation operator is determined for each pixel position of the panchromatic image, and the gradient compensation operator is used to reflect the edge and texture positions in the panchromatic image; The local texture complexity is determined based on each band of the multispectral image, and the adjustment coefficient for each pixel position of the multispectral image is determined based on the local texture complexity. For each band of the multispectral image, the edge enhancement amount is obtained by fusing the gradient compensation operator and the adjustment coefficient and then superimposed on the original pixel value of the band to obtain the fused pixels of each band. The satellite remote sensing image is determined based on the pixels of each band obtained through fusion and after processing.

[0006] In one embodiment, the step of determining the gradient compensation operator for the panchromatic image includes: Determine the horizontal and vertical gradients at each pixel location in the panchromatic image; The gradient magnitude of each pixel in the panchromatic image is determined based on the horizontal and vertical gradients, and the gradient magnitude is used as a gradient compensation operator.

[0007] In one embodiment, the step of determining the local texture complexity based on each band of the multispectral image includes: For the target band of the multispectral image, a local window of the pixel is determined based on the position of each pixel; the target band is any band in the multispectral image. The average absolute deviation of the local window of the pixel is used as the local texture complexity of the pixel position.

[0008] In one embodiment, the step of determining the adjustment coefficient for each pixel position of the multispectral image based on the local texture complexity includes: Obtain the first and second thresholds of the preset piecewise function; The local texture complexity is compared with the first threshold and the second threshold, and the adjustment coefficient for each pixel position of the multispectral image is determined based on the comparison result.

[0009] In one embodiment, the step of fusing the edge enhancement amount obtained based on the gradient compensation operator and the adjustment coefficient and then superimposing it onto the original pixel value of the band to obtain the fused pixels of each band includes: For each band of the multispectral image, the original pixel value of each pixel location and the weight coefficient corresponding to the band are obtained; The weighting coefficient, the gradient compensation operator, and the adjustment coefficient are multiplied together, and the product is used as the edge enhancement amount. The edge enhancement amount is superimposed on the original pixel value to obtain the fused pixels of each band.

[0010] In one embodiment, the step of acquiring satellite remote sensing imagery includes: Obtain metadata files of raw satellite remote sensing images and determine key parameters based on the metadata files; The original pixel values ​​of the satellite remote sensing image are converted into entrance pupil radiance based on the entrance pupil radiance model. The prior atmospheric model is matched based on the key parameters, and the entrance pupil radiance is corrected based on the prior atmospheric model to obtain the surface reflectance image. The satellite remote sensing image is obtained by performing fine correction on the regional digital elevation model corresponding to the surface reflectance image.

[0011] In one embodiment, after the step of determining the processed satellite remote sensing image based on the pixels of each band obtained through fusion, the method further includes: Obtain the affine transformation matrix corresponding to the processed satellite remote sensing image, and construct a mapping from geographic coordinates to image pixels based on the affine transformation matrix; Obtain a preset vector boundary file, and obtain the geographic coordinates of all points on the vector boundary based on the preset vector boundary file; The geographic coordinates are substituted into the mapping to obtain the corresponding geographic pixel coordinate sequence, and the pixel polygon is determined based on the geographic pixel coordinate sequence; Create a mask image with the same size as the processed satellite remote sensing image, and fill the mask image with polygons based on the pixel polygons to obtain a binary mask; The processed satellite remote sensing image is cropped based on the binary mask to obtain the satellite remote sensing image after removing black borders.

[0012] Furthermore, to achieve the above objectives, this application also proposes a satellite remote sensing image processing apparatus, which includes: The image acquisition module is used to acquire satellite remote sensing images, including panchromatic images and multispectral images; The gradient compensation module is used to determine the gradient compensation operator for each pixel position of the panchromatic image, and the gradient compensation operator is used to reflect the edge and texture positions in the panchromatic image. The texture adjustment module is used to determine the local texture complexity based on each band of the multispectral image, and to determine the adjustment coefficient of each pixel position of the multispectral image according to the local texture complexity. The fusion module is used to fuse each band of the multispectral image using the gradient compensation operator and the adjustment coefficient, and to determine the processed satellite remote sensing image based on the pixels of each fused band.

[0013] In addition, to achieve the above objectives, this application also proposes a satellite remote sensing image processing apparatus, the apparatus comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the satellite remote sensing image processing method described above.

[0014] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the satellite remote sensing image processing method described above.

[0015] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the satellite remote sensing image processing method described above.

[0016] One or more technical solutions proposed in this application have at least the following technical effects: This application acquires satellite remote sensing imagery, including panchromatic and multispectral images; determines a gradient compensation operator for each pixel location in the panchromatic image, which reflects the edge and texture locations within the panchromatic image; determines the local texture complexity based on each band of the multispectral image, and determines an adjustment coefficient for each pixel location in the multispectral image based on the local texture complexity; for each band of the multispectral image, the edge enhancement amount is fused based on the gradient compensation operator and the adjustment coefficient, and then superimposed onto the original pixel value of the band to obtain the fused pixels for each band; and determines the processed satellite remote sensing image based on the fused pixels for each band. Because it only identifies whether each pixel location belongs to a real edge or texture structure through the gradient compensation operator, and does not simply extract all high-frequency components, the structural information that truly reflects the geometric boundaries of ground features is used for subsequent injection, avoiding the misinterpretation of fluctuations in flat areas or sensor noise as "details" for injection, thus ensuring the targeted nature of edge enhancement. By adjusting coefficients to limit the edge enhancement requirements of different land cover types, the injection intensity of the panchromatic image is adaptively controlled, ensuring that the fused image remains smooth and natural in flat areas and sharp and clear in detailed areas, effectively solving the "edge blurring" problem in traditional algorithms. Simultaneously, since the original brightness information of the multispectral image is not replaced, but rather an edge enhancement term weighted by gradient compensation operators and adjustment coefficients is superimposed while maintaining the original pixel values ​​of the multispectral image, the proportional relationships between the original color channels are fully preserved, enabling the fused image to truly reflect the original colors of the land cover and fundamentally avoiding spectral distortion. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating an embodiment of the satellite remote sensing image processing method of this application. Figure 2 This is a flowchart illustrating Embodiment 2 of the satellite remote sensing image processing method of this application; Figure 3 This is a flowchart illustrating Embodiment 3 of the satellite remote sensing image processing method of this application; Figure 4 This is a schematic diagram of a single-band grayscale, high-resolution panchromatic image in one implementation of this application; Figure 5 This is a schematic diagram of the corresponding multi-band color, low-resolution multispectral image. Figure 6 This is a high-resolution, true-color satellite remote sensing image after removing black borders. Figure 7 This is a schematic diagram of the module structure of the satellite remote sensing image processing device according to an embodiment of this application; Figure 8 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the satellite remote sensing image processing method in the embodiments of this application.

[0020] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0021] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0022] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0023] The main solution of this application embodiment is as follows: acquiring satellite remote sensing images, including panchromatic images and multispectral images; determining a gradient compensation operator for each pixel position in the panchromatic image, the gradient compensation operator being used to reflect the edge and texture positions in the panchromatic image; determining the local texture complexity based on each band of the multispectral image, and determining the adjustment coefficient for each pixel position in the multispectral image based on the local texture complexity; fusing each band of the multispectral image through the gradient compensation operator and the adjustment coefficient, and determining the processed satellite remote sensing image based on the pixels of each fused band.

[0024] This application provides a solution that uses a gradient compensation operator to calculate a panchromatic image, thereby enabling edge intensity recognition at pixel locations. Regions with large gradient values ​​correspond to true geometric boundaries (such as road or building edges), while regions with small gradient values ​​correspond to flat areas or noise. By using this gradient compensation operator extraction method, only the true edge structure of the panchromatic image is used for fusion, avoiding the injection of noise or irrelevant signals into the multispectral image, thus preventing edge blurring and spectral distortion.

[0025] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, server, etc., or an electronic device or virtual device capable of performing the above functions. The following description uses a satellite remote sensing image processing device (hereinafter referred to as the processing device) as an example to illustrate this embodiment and the subsequent embodiments.

[0026] Based on this, embodiments of this application provide a method for processing satellite remote sensing images, referring to... Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the satellite remote sensing image processing method of this application.

[0027] In this embodiment, the satellite remote sensing image processing method includes steps S10 to S40: Step S10: Acquire satellite remote sensing imagery; The satellite remote sensing images include panchromatic images and multispectral images.

[0028] It is understandable that satellite remote sensing imagery refers to surface image data acquired by sensors carried by Earth observation satellites. Examples include remote sensing images from the GF-2 satellite (Gaofen-2), the ZY-3 satellite (Ziyuan-3), and the HJ-1A satellite (Huanjing-1A).

[0029] It should be understood that the aforementioned panchromatic imagery can refer to grayscale images acquired by satellite-borne sensors within a single wavelength range (typically covering the visible and near-infrared regions). Such panchromatic images possess high spatial resolution but lack color information. For example, the panchromatic images from the Gaofen-2 satellite have a spatial resolution better than 1 meter.

[0030] It is understandable that the aforementioned multispectral images refer to images acquired by sensors onboard a satellite in multiple bands (such as red, green, blue, and near-infrared bands). These images contain color information, but their spatial resolution is typically lower than that of panchromatic images. For example, the spatial resolution of multispectral images from the Gaofen-2 satellite is approximately 4 meters.

[0031] It should be noted that in satellite remote sensing imagery corresponding to the same environment, panchromatic and multispectral images cover the same environmental area. In practical use, the processing device can read data packets of satellite remote sensing imagery from local storage media, network servers, or satellite data distribution systems. This data packet contains at least two image files: one is the panchromatic image file corresponding to the panchromatic image, and the other is the multispectral image file corresponding to the multispectral image. The processing device loads the panchromatic image file into a panchromatic image data matrix and the multispectral image file into a multispectral image data matrix, respectively. Each element of the panchromatic image data matrix represents the grayscale value of a pixel location, and each element of the multispectral image data matrix represents the pixel value of a pixel location in a specific band. After acquiring these two types of images, the processing device records the spatial correspondence between them (e.g., the number of rows and columns of the images, georeferenced information, etc.) to prepare input data for subsequent fusion processing.

[0032] In some applications, since the original panchromatic and multispectral images may not be aligned, the original satellite remote sensing images can be aligned and geometrically coarsely corrected beforehand to obtain preliminary satellite remote sensing images. Specifically, the steps for acquiring satellite remote sensing images include: acquiring the metadata file of the original satellite remote sensing images and determining key parameters based on the metadata file; converting the original pixel values ​​of the original satellite remote sensing images into entrance pupil radiance according to the entrance pupil radiance model; matching the key parameters to a priori atmospheric model and correcting the entrance pupil radiance according to the prior atmospheric model to obtain a surface reflectance image; and performing fine correction according to the regional digital elevation model corresponding to the surface reflectance image to obtain the satellite remote sensing image.

[0033] It is understandable that the aforementioned raw satellite remote sensing imagery refers to the unprocessed image data directly acquired by the satellite's onboard sensors, without undergoing radiometric calibration or atmospheric correction. The pixel values ​​of this imagery are digital numbers (DN). Raw satellite remote sensing imagery may correspond to metadata files used to record auxiliary information, such as XML format data files, which can record key auxiliary parameters such as imaging time, sensor gain, bias, solar altitude angle, satellite attitude and observation angle, RPC coefficients, and sensor latitude and longitude.

[0034] It should be noted that the entrance pupil radiance model can refer to a mathematical model used to convert raw pixel values ​​into radiance values ​​at the sensor's entrance pupil. In one example of this application, the entrance pupil radiance model is: ; in, Indicates the brightness of the entrance pupil. This represents the sensor gain coefficient of the sensors carried by the satellite. This represents the sensor bias coefficient of the sensors carried by the satellite. Represents the pixel value (raw pixel value).

[0035] It is understandable that the aforementioned raw pixel values ​​can refer to the digitally quantized values ​​recorded at each pixel location in the original satellite remote sensing image. By converting the raw pixel values ​​based on key parameters using the entrance pupil radiance model, the physical radiance when light enters the aperture (entrance pupil) of the satellite-borne sensor can be determined, i.e., the entrance pupil radiance, thus achieving radiometric calibration of the original satellite remote sensing image.

[0036] It should be understood that while converting raw pixel values ​​to entrance pupil radiance eliminates differences in sensor response, it still incorporates interference from atmospheric transport paths. The aforementioned prior atmospheric model can refer to a pre-built model combining atmospheric state parameters. This model library pre-calculates multiple sets of parameters required for atmospheric correction (such as aerosol optical thickness, water vapor content, ozone content, etc.) according to aerosol type (e.g., marine, terrestrial, urban) and visibility level (e.g., 5km, 10km, 23km, 50km). Atmospheric correction using prior atmospheric models can eliminate or reduce interference from atmospheric transport paths.

[0037] It is understandable that surface reflectance imagery is the image obtained after radiometric calibration and atmospheric correction of the original satellite remote sensing imagery. Each pixel value in this surface reflectance imagery represents the proportion of solar radiation reflected by the corresponding ground target. The aforementioned regional digital elevation model can refer to a digital elevation model (DEM) covering the same geographical area as the surface reflectance imagery. This regional DEM records the elevation of each ground point in raster form. By using the regional DEM as a vertical constraint to perform least-squares adjustment correction on the original RPC coefficients, geometric distortions caused by terrain can be eliminated, resulting in a satellite remote sensing image with accurate geometric location. This satellite remote sensing imagery is a high-quality imagery after radiometric calibration, atmospheric correction, and fine-tuning correction. This satellite remote sensing imagery contains both panchromatic and multispectral imagery data. The panchromatic and multispectral images correspond to the same geographical area and can both have accurate geographical coordinates.

[0038] Step S20: Determine the gradient compensation operator for each pixel position of the panchromatic image; The gradient compensation operator is used to reflect the edge and texture positions in the panchromatic image. Step S30: Determine the local texture complexity based on each band of the multispectral image, and determine the adjustment coefficient of each pixel position in the multispectral image according to the local texture complexity.

[0039] It should be noted that the gradient compensation operator matrix is ​​a matrix of the same size as the panchromatic image. Each element value (i.e., the gradient compensation operator) in this matrix characterizes the edge strength or texture saliency of the corresponding pixel location in the panchromatic image. A larger gradient compensation operator value indicates a higher probability that the pixel location is in an edge or texture-rich region; a smaller value indicates a higher probability that the pixel location is in a flat or noisy region. For example, the gradient compensation operator value is larger at building edges or road boundaries; and smaller in flat areas such as water surfaces or uniform farmland.

[0040] It is understandable that the aforementioned local texture complexity can refer to a metric used to quantitatively describe the degree of drastic change in pixel values ​​within the region surrounding a pixel location in a multispectral image. The higher the local texture complexity, the richer the texture of the region surrounding the pixel (e.g., densely built-up areas, forest edges); the lower the local texture complexity, the flatter the region surrounding the pixel (e.g., water surfaces, bare land, uniform farmland).

[0041] In some embodiments of this application, the local texture complexity can be determined based on a single band of a multispectral image, or it can be determined by combining information from multiple bands.

[0042] It should be noted that the above adjustment coefficient can be a weighting factor used to control the intensity of edge texture injection, and this adjustment coefficient is positively correlated with the local texture complexity.

[0043] In practice, for regions with higher local texture complexity, the adjustment coefficient is set to a larger value, resulting in more pronounced edge enhancement during subsequent fusion; conversely, for regions with lower local texture complexity, the adjustment coefficient is set to a smaller value, preserving the original spectral characteristics during subsequent fusion. The adjustment coefficient can range from 0 to 1. By determining the adjustment coefficient for each pixel location in each band, a coefficient matrix of the same size as the multispectral image can be obtained. Each element in this coefficient matrix (i.e., the adjustment coefficient) controls the edge injection intensity at the corresponding pixel location during subsequent fusion.

[0044] Step S40: For each band of the multispectral image, the edge enhancement amount obtained by fusing the gradient compensation operator and the adjustment coefficient is superimposed on the original pixel value of the band to obtain the fused pixels of each band. Step S50: Determine the processed satellite remote sensing image based on the pixels of each band obtained by fusion.

[0045] It should be noted that by fusing, edge texture information extracted from panchromatic images can be injected into multispectral images to generate processed satellite remote sensing images with both high spatial resolution and spectral fidelity.

[0046] It is understandable that, since the aligned multispectral and panchromatic images correspond to the same geographic area and can both have precise geographic coordinates, each gradient compensation operator in the gradient compensation operator matrix can correspond to each adjustment coefficient in the adjustment coefficient matrix.

[0047] Specifically, for any band in a multispectral image, the processing device can acquire the original pixel value of each pixel location in that band. Based on the original pixel value corresponding to that pixel location, the gradient compensation operator, and the adjustment coefficient, fusion can be performed to obtain the edge enhancement amount. This edge enhancement amount is then superimposed on the original pixel value to obtain the fused pixels for each band. Based on the fused pixel values ​​corresponding to each pixel location, the fused image for that band can be determined. By combining the fused images of different bands by band, a final multi-band image can be obtained. This multi-band image can then be output as a processed satellite remote sensing image. The following explanation uses examples, but does not impose specific limitations on this embodiment.

[0048] For example, the processing device is processing the red band of a multispectral image. The original pixel value at pixel location (row 100, column 150) in the original image is 120. The processing device queries the gradient compensation operator value corresponding to this pixel location, which is 80, the adjustment coefficient is 0.6, and the weight coefficient for the red band is 0.3. The processing device calculates the new pixel value according to the fusion formula: New pixel value = 120 + 0.6 × 80 × 0.3 = 120 + 14.4 = 134.4, rounded to 134. At another pixel location (row 200, column 300), located at the edge of a building, the original pixel value is 90, the gradient compensation operator value is 90, the adjustment coefficient is 0.9, and the weight coefficient remains 0.3. Therefore, the new pixel value = 90 + 0.9 × 90 × 0.3 = 90 + 24.3 = 114.3, rounded to 114. After the processing equipment completes calculations for all 2500×2500 pixel positions in the red band, it obtains the fused image for the red band. Next, the processing equipment performs the same calculations for the green and blue bands respectively, obtaining fused images for the green and blue bands. Finally, the processing equipment combines the fused images of the red, green, and blue bands into a single three-band processed satellite remote sensing image. This image has the same resolution as the panchromatic image, while preserving the color information of the original multispectral image and enhancing edge sharpness.

[0049] This application embodiment acquires satellite remote sensing images, including panchromatic and multispectral images; determines a gradient compensation operator for each pixel location in the panchromatic image, which reflects the edge and texture locations in the panchromatic image; determines the local texture complexity based on each band of the multispectral image, and determines the adjustment coefficient for each pixel location in the multispectral image based on the local texture complexity; for each band of the multispectral image, the edge enhancement amount is fused based on the gradient compensation operator and the adjustment coefficient and superimposed on the original pixel value of the band to obtain the fused pixels of each band; and determines the processed satellite remote sensing image based on the fused pixels of each band. Since only the gradient compensation operator is used to identify whether each pixel location belongs to a real edge or texture structure, and not simply extracting all high-frequency components, the structural information that truly reflects the geometric boundaries of ground features is used for subsequent injection, avoiding the misinterpretation of fluctuations in flat areas or sensor noise as "details" for injection, thus ensuring the targeted nature of edge enhancement. By adjusting coefficients to limit the edge enhancement requirements of different land cover types, the injection intensity of the panchromatic image is adaptively controlled, ensuring that the fused image remains smooth and natural in flat areas and sharp and clear in detailed areas, effectively solving the "edge blurring" problem in traditional algorithms. Simultaneously, since the original brightness information of the multispectral image is not replaced, but rather an edge enhancement term weighted by gradient compensation operators and adjustment coefficients is superimposed while maintaining the original pixel values ​​of the multispectral image, the proportional relationships between the original color channels are fully preserved, enabling the fused image to truly reflect the original colors of the land cover and fundamentally avoiding spectral distortion.

[0050] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 2 , Figure 2 This is a flowchart illustrating Embodiment 2 of the satellite remote sensing image processing method of this application.

[0051] like Figure 2 As shown in the embodiments of this application, the step of determining the gradient compensation operator of the panchromatic image includes: Step S21: Determine the horizontal and vertical gradients at each pixel location in the panchromatic image; Step S22: Determine the gradient magnitude of each pixel position in the panchromatic image based on the horizontal gradient and the vertical gradient, and use the gradient magnitude as a gradient compensation operator.

[0052] It should be noted that the aforementioned horizontal gradient can refer to the rate of change of pixel value at a certain pixel location along the horizontal direction in a panchromatic image. The larger the absolute value of the horizontal gradient, the more significant the difference in pixel values ​​on both sides of that pixel location in the horizontal direction, indicating the presence of a horizontal edge or texture. Similarly, the aforementioned vertical gradient can refer to the rate of change of pixel value at a certain pixel location along the vertical direction in a panchromatic image. The larger the absolute value of the vertical gradient, the more significant the difference in pixel values ​​on both sides of that pixel location in the vertical direction, indicating the presence of a vertical edge or texture. In actual calculations, the center difference method can be used, that is, using the difference between the right and left pixels of the current pixel location to calculate the horizontal gradient, and using the difference between the pixels below and above the current pixel location to calculate the vertical gradient. Alternatively, convolutional kernels such as the Sobel operator or the Prewitt operator can be used for calculation; the specific calculation method is not limited in this embodiment.

[0053] It should be explained that the gradient magnitude mentioned above can refer to a scalar value calculated based on the horizontal and vertical gradients, which is used to comprehensively reflect the overall edge strength of the pixel position in both the horizontal and vertical directions.

[0054] In some embodiments of this application, the gradient compensation operator can be calculated as follows: ; For example, for any pixel location in a panchromatic image ,in, Indicates the line number. Indicates the column number. Indicates the pixel position The horizontal gradient at that location Indicates the pixel position Vertical gradient at that location This indicates the gradient magnitude at that pixel location.

[0055] In some embodiments of this application, in order to determine the local texture complexity, the step of determining the local texture complexity based on each band of the multispectral image includes: for the target band of the multispectral image, determining a local window of a pixel based on each pixel position; and using the average absolute deviation of the local window of the pixel as the local texture complexity of the pixel position.

[0056] It is understood that the target band mentioned above can be any one of the multiple bands contained in the multispectral image. For example, multispectral images typically contain blue, green, red, and near-infrared bands, and the processing device can process each band as the target band in sequence. The aforementioned pixel local window can refer to a local area centered on a pixel location in the multispectral image, consisting of that pixel and its surrounding neighboring pixels. In the embodiments of this application, the size of the pixel local window is not limited, and it can be 3×3 pixels, 5×5 pixels, or 7×7 pixels, etc.

[0057] It should be noted that mean absolute deviation (MAD) is a statistic used to measure the dispersion of a dataset. When a local window covers a textured region (such as urban building clusters, woodlands, or areas where roads and buildings intersect), the pixel values ​​at each pixel location within the window change drastically, and the data is relatively scattered. Therefore, the calculated MAD is large, indicating higher local texture complexity. By using MAD as the local texture complexity of a pixel location, a basis is provided for dynamically adjusting the edge injection intensity during the subsequent fusion process.

[0058] The exemplary calculation expression for the local texture complexity in this embodiment is as follows: ; in, Indicates the position of pixels in the current band of the multispectral image. The texture complexity value within a local window centered on the pixel. This indicates the total number of pixels contained within the local pixel window. Represented by pixel position A local window centered on a pixel. Represents the coordinates of any pixel position within a local window of a pixel. This indicates the pixel position within the local window of that pixel. Pixel value at that location, It is the arithmetic mean of the pixel values ​​at all pixel locations within the local window of this pixel.

[0059] In some embodiments of this application, in order to determine the adjustment coefficient of each pixel position of a multispectral image, the step of determining the adjustment coefficient of each pixel position of the multispectral image based on the local texture complexity includes: obtaining a first threshold and a second threshold of a preset piecewise function; comparing the local texture complexity with the first threshold and the second threshold, and determining the adjustment coefficient of each pixel position of the multispectral image based on the comparison result.

[0060] It should be noted that the aforementioned preset piecewise function refers to a pre-defined piecewise function used to map local texture complexity to adjustment coefficients. This piecewise function outputs different adjustment coefficient values ​​based on the different ranges of local texture complexity. The aforementioned first threshold and second threshold refer to two critical values ​​in the preset piecewise function used to divide different intervals. The first threshold is less than the second threshold, and together they divide the range of local texture complexity into three intervals: the interval less than the first threshold (corresponding to flat regions), the interval greater than or equal to the first threshold and less than the second threshold (corresponding to medium texture regions), and the interval greater than or equal to the second threshold (corresponding to rich texture regions). The adjustment coefficient is positively correlated with the local texture complexity; that is, flat regions correspond to smaller adjustment coefficients, and rich texture regions correspond to larger adjustment coefficients. For example, the first threshold can be set to T1, and the second threshold to T2. When the local texture complexity is less than T1, the adjustment coefficient is 0.3; when the local texture complexity is between T1 and T2, the adjustment coefficient is 0.6; and when the local texture complexity is greater than or equal to T2, the adjustment coefficient is 0.9. The corresponding example expression is shown below: ; in, Indicates pixel position The adjustment coefficient at the location.

[0061] This application's embodiments determine the horizontal and vertical gradients at each pixel location in a panchromatic image; based on these gradients, the gradient magnitude at each pixel location is determined, and this gradient magnitude is used as a gradient compensation operator. By accurately extracting structural information truly representing edges and textures from the panchromatic image, interference from noise and fluctuations in flat areas is eliminated, providing a clean and quantitative edge intensity reference for subsequent fusion, thus effectively solving the technical problem of "blurred edge textures at high resolution." Furthermore, since the gradient compensation operator is only used to control the injection intensity and does not replace any channels of the multispectral image, it lays the foundation for maintaining spectral fidelity in subsequent operations.

[0062] Based on the first and / or second embodiments of this application, in the third embodiment of this application, the content that is the same as or similar to the first and / or second embodiments described above can be referred to the above description and will not be repeated hereafter. Based on this, please refer to... Figure 3 , Figure 3 This is a flowchart illustrating Embodiment 3 of the satellite remote sensing image processing method of this application.

[0063] In this embodiment of the application, the step of fusing each band of the multispectral image using the gradient compensation operator and the adjustment coefficient, and determining the processed satellite remote sensing image based on the pixels of each fused band, includes: Step S41: For each band of the multispectral image, obtain the original pixel value of each pixel position and the weight coefficient corresponding to the band. Step S42: Multiply the weight coefficient, the gradient compensation operator, and the adjustment coefficient, and use the product as the edge enhancement amount; Step S43: The edge enhancement amount is superimposed on the original pixel value to obtain the fused pixels of each band.

[0064] It should be explained that the aforementioned weighting coefficients can be weighting factors corresponding to each band of the multispectral image, used to represent the proportion of the panchromatic image's spectral contribution to that band. The weighting factors for each band can be set according to actual applications, and this application does not impose any restrictions on this. For example, for the red, green, and blue bands, the weighting coefficients can be set to 0.3, 0.3, and 0.4, respectively.

[0065] Understandably, the aforementioned fusion process involves calculating the gradient compensation operator and adjustment coefficient values ​​extracted from the panchromatic image according to a preset fusion rule to obtain an edge enhancement increment. This edge enhancement increment is then added to the original pixel value to obtain the fused new pixel value. This process is repeated for each band of the multispectral image to obtain the fused image for all bands. Finally, the processing equipment combines the fused images of all bands into a single multiband image according to their original band order; this multiband image is the processed satellite remote sensing image.

[0066] In some embodiments of this application, exemplary calculation formulas for fusion may be as follows: ; in, Represents the original pixel value. Used to represent the weighting coefficient corresponding to band k This represents the merged pixel values.

[0067] In some embodiments of this application, the processing device can also perform spectral fidelity verification on the fusion result. Specifically, the root mean square error before and after fusion of the k-th band can be calculated first: ; in, This represents the root mean square error before and after the fusion of the k-th band. Indicates height and width. This represents the original pixel value of the k-th band before fusion. This represents the pixel value of the k-th band after fusion.

[0068] Then, the spectral fidelity index is obtained by averaging across all bands: ; in, Indicates the number of bands, such as 4 bands (red, green, blue, near-infrared). .

[0069] It should be noted that the spectral fidelity index was achieved through this application. This means that the average difference before and after fusion is less than 15, meeting industry standards. Compared to traditional wavelet fusion (D-24), this represents an improvement of approximately 37%.

[0070] In some embodiments of this application, after the step of determining the processed satellite remote sensing image based on the pixels of each band obtained by fusion, the method further includes: obtaining the affine transformation matrix corresponding to the processed satellite remote sensing image, and constructing a mapping from geographic coordinates to image pixels based on the affine transformation matrix; obtaining a preset vector boundary file, and obtaining the geographic coordinates of all points on the vector boundary based on the preset vector boundary file; substituting the geographic coordinates into the mapping to obtain the corresponding geographic pixel coordinate sequence, and determining the pixel polygon based on the geographic pixel coordinate sequence; Create a mask image with the same size as the processed satellite remote sensing image, and fill the mask image with polygons based on the pixel polygons to obtain a binary mask; crop the processed satellite remote sensing image based on the binary mask to obtain the satellite remote sensing image after removing black borders.

[0071] It should be understood that an affine transformation matrix is ​​a matrix that can be used to describe the linear mapping relationship between image pixel coordinates (i.e., image cells) and geographic coordinates. It can define the image's position, size, rotation, and tilt in space. The aforementioned vector boundary refers to the outline of the geographic region corresponding to the target geographic area, as specified by the user, such as the administrative boundary of a province, the administrative boundary of a city, or the scope of an economic zone. A vector boundary file is a file that stores vector boundaries. Vector boundaries can be formed by connecting a series of ordered geographic coordinate points, creating a closed polygon.

[0072] It should be noted that, in order to eliminate the black borders that may result from coordinate system transformation, this application can perform vector-driven adaptive mask clipping operations. Specifically, this can be achieved using an affine transformation matrix. High-precision mapping, in which That is, the specific pixel location (pixel coordinates) in the image. Coordinates in a geographic coordinate system can be represented by latitude and longitude. It is an affine transformation matrix used to map geographic coordinates to image pixel coordinates. This matrix can contain image rotation, translation and scaling information, enabling high-precision alignment between vector boundaries (geographic coordinates) and image pixels (pixel coordinates).

[0073] It should be noted that the aforementioned polygon filling involves filling the pixel polygons to the corresponding positions in the mask image. For the mask within the vector boundary of the pixel polygon, the value can be 1; for the mask outside the vector boundary of the pixel polygon, the value can be 0. This yields a binary mask for the satellite remote sensing image. This binary mask can be used to perform cropping processing on the processed satellite remote sensing image. By multiplying the pixel value of the pixel coordinates with the mask, the final satellite remote sensing image, after removing black borders and retaining the target geographical area, can be obtained. Specifically, it can be done as follows... Figure 4-6 ,in Figure 4 This is a schematic diagram of a single-band grayscale, high-resolution panchromatic image in one implementation of this application. Figure 5 This is a schematic diagram of the corresponding multi-band color, low-resolution multispectral image (where the legend bands are: red Band_1, green Band_2, and blue Band_3). Figure 6 The image is a high-resolution, true-color satellite remote sensing image after removing black borders (the legend bands are: red Layer_1, green Layer_2, and blue Layer_3).

[0074] This application embodiment obtains the original pixel values ​​and corresponding weight coefficients for each pixel location in each band of a multispectral image. It then fuses the images based on the weight coefficients, original pixel values, gradient compensation operators, and adjustment coefficients, and determines the processed satellite remote sensing image based on the fused pixels in each band. Since the original pixel values ​​are fully preserved, the original proportional relationships between the bands of the multispectral image remain unchanged. Therefore, the fused image accurately reflects the original spectral characteristics of ground objects, fundamentally avoiding color distortion. By introducing band-specific weight coefficients, edge enhancement information can be allocated according to the actual spectral contribution ratio of each band, making the fusion result more physically reasonable and avoiding spectral distortion caused by equal enhancement amounts in each band.

[0075] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the processing method of satellite remote sensing images in this application. Any simple transformations based on this technical concept are within the protection scope of this application.

[0076] This application also provides a satellite remote sensing image processing apparatus, please refer to... Figure 7 , Figure 7 This is a schematic diagram of the module structure of a satellite remote sensing image processing device according to an embodiment of the application. The satellite remote sensing image processing device includes: Image acquisition module 10 is used to acquire satellite remote sensing images, including panchromatic images and multispectral images; The gradient compensation module 20 is used to determine the gradient compensation operator for each pixel position of the panchromatic image, and the gradient compensation operator is used to reflect the edge and texture positions in the panchromatic image. The texture adjustment module 30 is used to determine the local texture complexity based on each band of the multispectral image, and to determine the adjustment coefficient of each pixel position of the multispectral image according to the local texture complexity. The fusion module 40 is used to fuse the edge enhancement amount obtained by fusing the gradient compensation operator and the adjustment coefficient for each band of the multispectral image and then superimpose it on the original pixel value of the band to obtain the fused pixels of each band. The fusion module 40 is also used to determine the processed satellite remote sensing image based on the pixels of each band obtained by fusion.

[0077] The satellite remote sensing image processing apparatus provided in this application employs the satellite remote sensing image processing method described in the above embodiments. This method effectively solves the technical problem that traditional algorithms, which simply add high-frequency details from panchromatic images to multispectral images, easily lead to color (spectral) distortion and blurred edge textures at high resolutions. Compared with the prior art, the beneficial effects of the satellite remote sensing image processing apparatus provided in this application are the same as those of the satellite remote sensing image processing method provided in the above embodiments. Furthermore, other technical features in the satellite remote sensing image processing apparatus are the same as those disclosed in the methods of the above embodiments, and will not be elaborated upon here.

[0078] This application provides a satellite remote sensing image processing apparatus, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the satellite remote sensing image processing method in Embodiment 1 above.

[0079] The following is for reference. Figure 8This document illustrates a schematic diagram of a processing device suitable for implementing the embodiments of this application for satellite remote sensing images. The processing device for satellite remote sensing images in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 8 The satellite remote sensing image processing device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0080] like Figure 8 As shown, the satellite remote sensing image processing device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 1002 or a program loaded from storage device 1003 into random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the satellite remote sensing image processing device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the satellite remote sensing image processing equipment to communicate wirelessly or wiredly with other devices to exchange data. Although satellite remote sensing image processing equipment with various systems is shown in the figure, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems may be implemented alternatively.

[0081] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0082] The satellite remote sensing image processing device provided in this application, employing the satellite remote sensing image processing method described in the above embodiments, can solve the technical problem that traditional algorithms, which simply add high-frequency details from panchromatic images to multispectral images, easily lead to color (spectral) distortion and blurred edge textures at high resolutions. Compared with the prior art, the beneficial effects of the satellite remote sensing image processing device provided in this application are the same as those of the satellite remote sensing image processing method provided in the above embodiments, and other technical features in this satellite remote sensing image processing device are the same as those disclosed in the method of the previous embodiment, and will not be repeated here.

[0083] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0084] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0085] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the satellite remote sensing image processing method in the above embodiments.

[0086] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or flash memory, optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0087] The aforementioned computer-readable storage medium may be included in the satellite remote sensing image processing equipment; or it may exist independently and not be assembled into the satellite remote sensing image processing equipment.

[0088] The aforementioned computer-readable storage medium carries one or more programs that, when executed by a satellite remote sensing image processing device, cause the satellite remote sensing image processing device to: Acquire satellite remote sensing images, including panchromatic images and multispectral images; A gradient compensation operator is determined for each pixel position of the panchromatic image, and the gradient compensation operator is used to reflect the edge and texture positions in the panchromatic image; The local texture complexity is determined based on each band of the multispectral image, and the adjustment coefficient for each pixel position of the multispectral image is determined based on the local texture complexity. For each band of the multispectral image, the edge enhancement amount is obtained by fusing the gradient compensation operator and the adjustment coefficient and then superimposed on the original pixel value of the band to obtain the fused pixels of each band. The satellite remote sensing image is determined based on the pixels of each band obtained through fusion and after processing.

[0089] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0090] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation that may be implemented in systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0091] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0092] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described satellite remote sensing image processing method. This solves the technical problem that traditional algorithms, which simply add high-frequency details from panchromatic images to multispectral images, easily lead to color (spectral) distortion and blurred edge textures at high resolutions. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the satellite remote sensing image processing method provided in the above embodiments, and will not be repeated here.

[0093] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the satellite remote sensing image processing method described above.

[0094] The computer program product provided in this application can solve the technical problem that traditional algorithms, which simply add high-frequency details from panchromatic images to multispectral images, easily lead to color (spectral) distortion and blurred edge textures at high resolutions. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the satellite remote sensing image processing method provided in the above embodiments, and will not be repeated here.

[0095] The above description is only a part of the embodiments of this application and does not limit the scope of protection of this application. All equivalent structural transformations made under the technical concept of this application and using the content of this application specification and drawings, or direct / indirect applications in other related technical fields, are included in the scope of protection of this application.

Claims

1. A method for processing satellite remote sensing images, characterized in that, The method includes: Acquire satellite remote sensing images, including panchromatic images and multispectral images; A gradient compensation operator is determined for each pixel position of the panchromatic image, and the gradient compensation operator is used to reflect the edge and texture positions in the panchromatic image; The local texture complexity is determined based on each band of the multispectral image, and the adjustment coefficient for each pixel position of the multispectral image is determined based on the local texture complexity. For each band of the multispectral image, the edge enhancement amount is obtained by fusing the gradient compensation operator and the adjustment coefficient and then superimposed on the original pixel value of the band to obtain the fused pixels of each band. The satellite remote sensing image is determined based on the pixels of each band obtained through fusion and after processing.

2. The satellite remote sensing image processing method as described in claim 1, characterized in that, The step of determining the gradient compensation operator for the panchromatic image includes: Determine the horizontal and vertical gradients at each pixel location in the panchromatic image; The gradient magnitude of each pixel in the panchromatic image is determined based on the horizontal and vertical gradients, and the gradient magnitude is used as a gradient compensation operator.

3. The satellite remote sensing image processing method as described in claim 1, characterized in that, The step of determining the local texture complexity based on each band of the multispectral image includes: For the target band of the multispectral image, a local window of the pixel is determined based on the position of each pixel; the target band is any band in the multispectral image. The average absolute deviation of the local window of the pixel is used as the local texture complexity of the pixel position.

4. The satellite remote sensing image processing method as described in claim 1, characterized in that, The step of determining the adjustment coefficient for each pixel position of the multispectral image based on the local texture complexity includes: Obtain the first and second thresholds of the preset piecewise function; The local texture complexity is compared with the first threshold and the second threshold, and the adjustment coefficient for each pixel position of the multispectral image is determined based on the comparison result.

5. The satellite remote sensing image processing method as described in claim 1, characterized in that, The step of fusing the edge enhancement amount obtained based on the gradient compensation operator and the adjustment coefficient and then superimposing it onto the original pixel value of the band to obtain the fused pixels of each band includes: For each band of the multispectral image, the original pixel value of each pixel location and the weight coefficient corresponding to the band are obtained; The weighting coefficient, the gradient compensation operator, and the adjustment coefficient are multiplied together, and the product is used as the edge enhancement amount. The edge enhancement amount is superimposed on the original pixel value to obtain the fused pixels of each band.

6. The satellite remote sensing image processing method as described in claim 1, characterized in that, The steps for acquiring satellite remote sensing images include: Obtain metadata files of raw satellite remote sensing images and determine key parameters based on the metadata files; The original pixel values ​​of the satellite remote sensing image are converted into entrance pupil radiance based on the entrance pupil radiance model. The prior atmospheric model is matched based on the key parameters, and the entrance pupil radiance is corrected based on the prior atmospheric model to obtain the surface reflectance image. The satellite remote sensing image is obtained by performing fine correction on the regional digital elevation model corresponding to the surface reflectance image.

7. The satellite remote sensing image processing method as described in claim 1, characterized in that, Following the step of determining the processed satellite remote sensing image based on the pixels of each band obtained through fusion, the method further includes: Obtain the affine transformation matrix corresponding to the processed satellite remote sensing image, and construct a mapping from geographic coordinates to image pixels based on the affine transformation matrix; Obtain a preset vector boundary file, and obtain the geographic coordinates of all points on the vector boundary based on the preset vector boundary file; The geographic coordinates are substituted into the mapping to obtain the corresponding geographic pixel coordinate sequence, and the pixel polygon is determined based on the geographic pixel coordinate sequence; Create a mask image with the same size as the processed satellite remote sensing image, and fill the mask image with polygons based on the pixel polygons to obtain a binary mask; The processed satellite remote sensing image is cropped based on the binary mask to obtain the satellite remote sensing image after removing black borders.

8. A satellite remote sensing image processing apparatus, characterized in that, The satellite remote sensing image processing device includes: The image acquisition module is used to acquire satellite remote sensing images, including panchromatic images and multispectral images; The gradient compensation module is used to determine the gradient compensation operator for each pixel position of the panchromatic image, and the gradient compensation operator is used to reflect the edge and texture positions in the panchromatic image. The texture adjustment module is used to determine the local texture complexity based on each band of the multispectral image, and to determine the adjustment coefficient of each pixel position of the multispectral image according to the local texture complexity. The fusion module is used to fuse each band of the multispectral image based on the gradient compensation operator and the adjustment coefficient to obtain an edge enhancement amount, which is then superimposed on the original pixel value of the band to obtain the fused pixels of each band. The fusion module is also used to determine the processed satellite remote sensing image based on the pixels of each band obtained by fusion.

9. A satellite remote sensing image processing device, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the method for processing satellite remote sensing imagery as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the satellite remote sensing image processing method as described in any one of claims 1 to 7.

Images