A remote sensing image connection point matching method and a computer readable medium

Abstract

The application provides a remote sensing image connection point matching method and a computer readable medium. The application solves an intersection polygon involved in image matching, and constructs a minimum circumscribed rectangle of the intersection polygon according to a geographical range of the intersection polygon. After grid division of the minimum circumscribed rectangle is performed to obtain a plurality of subblocks, the subblocks are screened based on a subblock overlap rate, and a plurality of screened minimum circumscribed rectangle subblocks are obtained. Subblock division is performed on the plurality of screened minimum circumscribed rectangle subblocks, and a screened subblock is reserved when a subblock overlap rate meets a threshold. After local geometric correction is performed on the screened subblock according to longitude and latitude ranges of four corners of the screened subblock, a plurality of local geometric correction subblocks are obtained, the local geometric correction subblocks are matched respectively, a random sample consensus algorithm is used for outlier elimination on a matching result, and finally, connection point coordinates are obtained. The remote sensing image connection point matching threshold is effectively reduced. In addition, the subblock idea is easy to be optimized in parallel, and the efficiency can be further improved.

Description

Technical Field

[0001] This invention belongs to the field of photogrammetry and remote sensing technology, specifically relating to a method for matching tie points in remote sensing images and a computer-readable medium. Background Technology

[0002] Image matching refers to the process of identifying corresponding points between two or more images using a specific matching algorithm. As a preliminary step in image fusion and target change detection, image matching has been widely applied in remote sensing, digital photogrammetry, cartography, and military applications. However, unlike ordinary visual images, remote sensing images are characterized by their wide image width and large data volume, making conventional image matching algorithms difficult to apply effectively. Further optimization and improvement of matching algorithms are needed. Existing remote sensing image matching algorithms lack a standard procedure, requiring practitioners to possess both image processing and photogrammetry knowledge, thus limiting the further application of conventional image matching algorithms in remote sensing image matching. Therefore, there is an urgent need for a remote sensing image tie-point matching framework that can fully consider the characteristics of remote sensing images, perform specific processing on them, and enable conventional image matching algorithms to be directly applied to remote sensing image matching. Summary of the Invention

[0003] This invention addresses the lack of a unified framework in existing remote sensing image matching algorithms by designing a new framework for matching remote sensing image tie points. This framework divides the image into a grid based on object-space geographic constraints and eliminates terrain influences through local correction of sub-blocks. Then, the points matched from the sub-blocks are converted into pixel coordinates on the entire image using a localization model. Finally, the image-space coordinates corresponding to the sub-blocks are merged to generate tie points for the entire image. This framework, combined with a block-based parallel approach, improves efficiency and effectively lowers the barrier to entry for remote sensing image tie point matching.

[0004] To achieve the above objectives, the present invention provides a method for matching tie points in remote sensing images, comprising the following steps:

[0005] Step 1: Acquire the first image and its corresponding RPC positioning model. Obtain the pixel width and pixel height of the first image to construct the image-side four corner points of the first image. Use the RPC positioning model of the first image to calculate the object-side geographic coordinates corresponding to the image-side four corner points of the first image. Acquire the second image and its corresponding RPC positioning model. Obtain the pixel width and pixel height of the second image to construct the image-side four corner points of the second image. Use the RPC positioning model of the second image to calculate the object-side geographic coordinates corresponding to the image-side four corner points of the second image.

[0006] Step 2: Construct a vector polygon of the first image using the object-side geographic coordinates corresponding to the four corner points of the image. Construct a vector polygon of the second image using the object-side geographic coordinates corresponding to the four corner points of the image. Calculate the intersecting polygon between the vector polygons of the first and second images using a spatial intersection calculation method and obtain the geographic range of the intersecting polygon. Construct the minimum bounding rectangle of the intersecting polygon based on the geographic range of the intersecting polygon.

[0007] Step 3: Combine the DEM auxiliary data with the RPC positioning model of the first image to calculate the image resolution of the first image. Combine the DEM auxiliary data with the RPC positioning model of the second image to calculate the image resolution of the second image. Further obtain the resampling resolution. Calculate the pixel width and pixel height corresponding to the resampling resolution based on the minimum bounding rectangle of the intersecting polygon.

[0008] Step 4: Divide the minimum bounding rectangle of the intersecting polygons into multiple minimum bounding rectangle blocks by grid division. Calculate the overlap rate between each minimum bounding rectangle block and the intersecting polygons in turn. Further, select the minimum bounding rectangle blocks with an overlap rate greater than the overlap rate threshold from the multiple minimum bounding rectangle blocks as multiple filtered minimum bounding rectangle blocks.

[0009] Step 5: Perform grid division on multiple filtered minimum bounding rectangle blocks in sequence to obtain multiple sub-blocks of each filtered minimum bounding rectangle block. Calculate the overlap rate between each sub-block of each filtered minimum bounding rectangle block and the intersecting polygon. Further, select the sub-blocks of each filtered minimum bounding rectangle block whose overlap rate with the intersecting polygon is greater than the fine overlap rate threshold from the multiple sub-blocks of each filtered minimum bounding rectangle block as multiple filtered sub-blocks of each filtered minimum bounding rectangle block.

[0010] Step 6: Based on each filtered rectangular sub-block of the minimum bounding rectangle block after each filtering, and combined with DEM auxiliary data, perform local geometric correction on the first image to obtain multiple geometrically corrected sub-blocks of the first image. Based on the latitude and longitude of the four corners of each filtered sub-block of the minimum bounding rectangle block after each filtering, and combined with DEM auxiliary data, perform local geometric correction on the second image to obtain multiple geometrically corrected sub-blocks of the second image.

[0011] Step 7: Using each geometrically corrected sub-block of the first image as the reference image and the geometrically corrected sub-block of the second image as the registration image, a matching algorithm is used to obtain multiple sets of corresponding point pairs located at the geographic coordinates of the geometrically corrected sub-blocks of the first and second images; the geographic coordinates of each geometrically corrected sub-block of the first image are mapped to the corresponding pixel coordinates on the first image, and the geographic coordinates of each geometrically corrected sub-block of the second image are mapped to the corresponding pixel coordinates on the second image.

[0012] Step 8: Use a random sampling consensus algorithm to remove outliers from the pixel coordinates corresponding to the geographic coordinates of each geometrically corrected sub-block in the first image and the pixel coordinates corresponding to the geographic coordinates of each geometrically corrected sub-block in the second image, to obtain the pixel coordinates of the connection points in the first image and the pixel coordinates of the connection points in the second image.

[0013] Preferably, step 2, obtaining the geographical extent of the intersecting polygons, is as follows:

[0014] Obtain the longitude and latitude of each point in the intersecting polygon in sequence;

[0015] To construct the longitude range of an intersecting polygon, we select the minimum and maximum longitude values ​​of each point within the polygon from among the longitudes of multiple points in the polygon. This range is defined as follows:

[0016] [long_min, long_max]

[0017] Where long_min represents the minimum longitude of each point in the intersecting polygon, and long_max represents the maximum longitude of each point in the intersecting polygon;

[0018] In an intersecting polygon, the minimum and maximum latitude values ​​of each point are selected from the latitude values ​​of multiple points within the polygon. These are then used as the latitude range of the intersecting polygon, defined as follows:

[0019] [lat_min,lat_max]

[0020] Where lat_min represents the minimum latitude of each point in the intersecting polygon, and lat_max represents the maximum latitude of each point in the intersecting polygon;

[0021] The geographical extent of the intersecting polygons is constructed by the longitude range and the latitude range of the intersecting polygons.

[0022] Step 2 involves constructing the minimum bounding rectangle of the intersecting polygons based on their geographical extent, as detailed below:

[0023] Use (long_min, lat_min), (long_min, lat_max), (long_max, lat_max), (long_min, lat_max) as the four vertices of the minimum bounding rectangle of the intersecting polygon to construct the minimum bounding rectangle of the intersecting polygon.

[0024] Preferably, step 3 involves calculating the image resolution of the first image using the RPC positioning model of the first image. The specific process is as follows:

[0025] The first image is defined as: ImgA;

[0026] Obtain the width of the first image in pixels ImgA, and define it as: ImgA_w;

[0027] Obtain the pixel height of the first image, ImgA, and define it as: ImgA_h;

[0028] The pixel coordinates of the first pixel in the first image are obtained and defined as follows:

[0029]

[0030] Where N represents the pixel step size, ( ) represents the column coordinate of the first pixel in the first image. () represents the row coordinate of the first pixel in the first image. This represents the first pixel of the first image.

[0031] Combining DEM-assisted data with the RPC positioning model of the first image, the 3D geographic coordinates corresponding to the pixel coordinates of the first pixel in the first image are calculated using ray tracing, and are defined as follows:

[0032]

[0033] in, This represents the longitude of the three-dimensional geographic coordinates of the first pixel in the first image. This represents the latitude of the three-dimensional geographic coordinates of the first pixel in the first image. This indicates the elevation of the three-dimensional geographic coordinates of the first pixel in the first image.

[0034] The pixel coordinates of the second pixel in the first image are obtained and defined as follows:

[0035]

[0036] Where N represents the pixel step size, ( ) represents the column coordinate of the second pixel in the first image. () represents the column coordinate of the second pixel in the first image. This represents the second pixel of the second image.

[0037] Combining DEM-assisted data with the RPC positioning model of the first image, the 3D geographic coordinates corresponding to the pixel coordinates of the second pixel point in the first image are calculated using the ray tracing method and defined as follows:

[0038]

[0039] in, This indicates the longitude of the three-dimensional geographic coordinates of the second pixel in the first image. This represents the latitude of the three-dimensional geographic coordinates of the second pixel in the first image. This indicates the elevation of the three-dimensional geographic coordinates of the first pixel in the second image;

[0040] The resolution of the first image is calculated as follows:

[0041]

[0042] in, The resolution of the first image is given by N, where N represents the pixel step size.

[0043] Step 3 involves calculating the image resolution of the second image using the RPC positioning model. The specific process is as follows:

[0044] The second image is defined as: ImgB;

[0045] Obtain the width of the second image in pixels ImgB, defined as: ImgB_w;

[0046] The pixel height of the second image, ImgB, is obtained and defined as: ImgB_h;

[0047] The pixel coordinates of the first pixel in the second image are obtained and defined as follows:

[0048]

[0049] Where N represents the pixel step size, ( ) represents the column coordinate of the first pixel in the second image. () represents the row coordinate of the first pixel in the second image. This represents the first pixel of the second image.

[0050] Combining DEM-assisted data with the RPC positioning model of the second image, the ray tracing method is used to calculate the 3D geographic coordinates corresponding to the pixel coordinates of the first pixel in the second image, defined as:

[0051]

[0052] in, This indicates the longitude of the three-dimensional geographic coordinates of the first pixel in the second image. This represents the latitude and three-dimensional geographic coordinates of the first pixel in the second image. This indicates the elevation of the three-dimensional geographic coordinates of the first pixel in the second image;

[0053] The pixel coordinates of the second pixel in the second image are obtained and defined as follows:

[0054]

[0055] Where N represents the pixel step size, ( ) represents the column coordinate of the second pixel in the second image. () represents the column coordinate of the second pixel in the second image. This represents the second pixel of the second image.

[0056] Combining DEM-assisted data with the RPC positioning model of the second image, the 3D geographic coordinates corresponding to the pixel coordinates of the second pixel point in the second image are calculated using the ray tracing method and defined as follows:

[0057]

[0058] in, This indicates the longitude of the three-dimensional geographic coordinates of the second pixel point in the second image. This represents the latitude and three-dimensional geographic coordinates of the second pixel in the second image. This indicates the elevation of the three-dimensional geographic coordinates of the second pixel in the second image;

[0059] The resolution of the second image is calculated as follows:

[0060]

[0061] in, The resolution of the second image is given by N, which represents the pixel step size.

[0062] Step 3 further obtains the resampling resolution, and the specific process is as follows:

[0063] The larger value between the resolution of the first image and the resolution of the second image is used as the resampling resolution, which is defined as follows: ;

[0064] Step 3 involves calculating the pixel width m_W and pixel height m_H at the resampling resolution based on the minimum bounding rectangle of the intersecting polygons, using the following formula:

[0065]

[0066] Where m_W represents the pixel width of the minimum bounding rectangle of the intersecting polygons at the resampling resolution, and m_H represents the pixel height of the minimum bounding rectangle of the intersecting polygons at the resampling resolution. The values ​​represent the resampling resolution, long_min represents the minimum longitude of each point in the intersecting polygon, long_max represents the maximum longitude of each point in the intersecting polygon, lat_min represents the minimum latitude of each point in the intersecting polygon, and lat_max represents the maximum latitude of each point in the intersecting polygon.

[0067] Preferably, step 4 involves dividing the intersecting polygons into multiple minimum bounding rectangle blocks using a grid. The specific process is as follows:

[0068] Set the number of grid divisions in the column direction to be as follows Set the number of grid divisions in the row direction to be [number]. ;

[0069] Calculate the pixel width and pixel height of each minimum bounding rectangle block, as follows:

[0070]

[0071] in, The pixel width of each block within the minimum bounding rectangle. The pixel height of each block of the minimum bounding rectangle. The number of grid divisions in the column direction, To set the number of grid divisions in the row direction, m_W represents the pixel width corresponding to the minimum bounding rectangle of the intersecting polygons at the resampling resolution, and m_H represents the pixel height corresponding to the minimum bounding rectangle of the intersecting polygons at the resampling resolution.

[0072] The smallest bounding rectangle block in row j and column i is defined as:

[0073]

[0074] ,

[0075] Where i represents the block number in the column direction, and j represents the block number in the row direction. The number of grid divisions in the column direction, To set the number of grid divisions in the row direction;

[0076] The top-left pixel coordinates of the smallest bounding rectangle block in row j and column i are calculated as follows:

[0077]

[0078]

[0079]

[0080] in, The pixel width of each block within the minimum bounding rectangle. The pixel height of each block of the minimum bounding rectangle. Let represent the top-left pixel coordinates of the smallest bounding rectangle block in row j and column i. The row coordinate represents the top-left pixel coordinate of the smallest bounding rectangle block in row j and column i. The column coordinate represents the top-left pixel coordinate of the smallest bounding rectangle block in row j and column i;

[0081] The bottom-right pixel coordinates of the smallest bounding rectangle block in row j and column i are calculated as follows:

[0082]

[0083]

[0084]

[0085] in, This represents the bottom-right pixel coordinates of the smallest bounding rectangle block in row j and column i. The row coordinate represents the bottom-right pixel coordinate of the smallest bounding rectangle block in row j and column i. The column coordinate represents the bottom right pixel coordinate of the smallest bounding rectangle block in row j and column i;

[0086] Calculate the minimum longitude, maximum longitude, minimum latitude, and maximum latitude of the minimum bounding rectangle of the minimum bounding rectangle block in row j and column i, as follows:

[0087]

[0088] in, Let represent the minimum longitude of the smallest bounding rectangle block in row j and column i. This represents the maximum longitude of the smallest bounding rectangular block in row j and column i. This represents the minimum latitude of the smallest bounding rectangle block in the j-th row and i-th column. represents the maximum latitude of the smallest bounding rectangle block in row j and column i, long_min represents the minimum longitude of each point in the intersecting polygon, long_max represents the maximum longitude of each point in the intersecting polygon, lat_min represents the minimum latitude of each point in the intersecting polygon, and lat_max represents the maximum latitude of each point in the intersecting polygon.

[0089] Will( , ), ( , ), ( , ), ( , ) as four vertices to construct the geographic rectangle corresponding to the minimum bounding rectangle of the j-th row and i-th column;

[0090] Step 4 involves sequentially calculating the overlap rate between each minimum bounding rectangle block and the intersecting polygon, as detailed below:

[0091] The overlap rate of the smallest bounding rectangle block in row j and column i is calculated as follows:

[0092]

[0093] ,

[0094] Where i represents the block number in the column direction, and j represents the block number in the row direction. The number of grid divisions in the column direction, To set the number of grid divisions in the row direction, Let represent the overlap rate of the smallest bounding rectangle block in the j-th row and i-th column. Represents the space intersection operation. This indicates how to calculate the area of ​​a polygon. Let represent the smallest bounding rectangle block in row j and column i. Represents intersecting polygons;

[0095] like ,but ;

[0096] like ,but 0;

[0097] in, Indicates the overlap rate threshold. The attribute representing the smallest bounding rectangle block in the j-th row and i-th column;

[0098] The method for obtaining the minimum bounding rectangle block after multiple filtering steps in step 4 is as follows:

[0099] Select from multiple minimum bounding rectangle blocks The minimum bounding rectangle is used to divide the data into multiple filtered minimum bounding rectangles, as defined below:

[0100]

[0101] g∈[1,K], ,

[0102] Where i represents the block number in the column direction, and j represents the block number in the row direction. The number of grid divisions in the column direction, To set the number of grid divisions in the row direction, This represents the smallest bounding rectangle block after the g-th filtering, i.e., the smallest bounding rectangle block after the g-th filtering. Line number The minimum bounding rectangle of the column, where K represents the number of minimum bounding rectangles after filtering;

[0103] Preferably, step 5 involves sequentially dividing the multiple filtered minimum bounding rectangle blocks into grids to obtain multiple sub-blocks for each filtered minimum bounding rectangle block, as detailed below:

[0104] Set the number of grid divisions in the column direction to be as follows: Set the number of grid cells to be divided in the row direction. ;

[0105] Calculate the pixel width and pixel height of each sub-block within the smallest bounding rectangle after filtering, as follows:

[0106]

[0107] in, The pixel width of each sub-block of the smallest bounding rectangle after filtering. The pixel height of each sub-block of the smallest bounding rectangle after filtering; The number of grid divisions in the column direction is determined by the number of grids. The number of grid divisions in the row direction is specified. The pixel width of each block within the minimum bounding rectangle. The pixel height of each block of the smallest bounding rectangle;

[0108] The sub-block in row q and column p of the minimum bounding rectangle after the g-th filtering is defined as:

[0109]

[0110] , ,

[0111] Where g represents the sub-block of the smallest bounding rectangle after filtering, K represents the number of the smallest bounding rectangles after filtering, q represents the sub-block number in the column direction, and p represents the sub-block number in the row direction. The number of grid divisions in the column direction is determined by the number of grids. The number of grid divisions in the row direction of the block.

[0112] The top-left pixel coordinates of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering are calculated as follows:

[0113]

[0114]

[0115]

[0116] in, The pixel width of each sub-block of the smallest bounding rectangle after filtering. The pixel height of each sub-block of the smallest bounding rectangle after filtering. This represents the top-left pixel coordinates of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the row coordinate of the top-left pixel coordinate of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the column coordinate of the top-left pixel coordinate of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering. This represents the row coordinate of the top-left pixel of the smallest bounding rectangle after the g-th filtering, located in the j-th row and i-th column. This represents the column coordinate of the top-left pixel coordinate of the smallest bounding rectangle after the g-th filtering, located in the j-th row and i-th column.

[0117] Calculate the bottom-right pixel coordinates of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering, as follows:

[0118]

[0119]

[0120]

[0121] in, The pixel width of each sub-block of the smallest bounding rectangle after filtering. The pixel height of each sub-block of the smallest bounding rectangle after filtering. This represents the bottom-right pixel coordinates of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the row coordinate of the bottom right pixel of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the column coordinate of the bottom right pixel coordinate of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering. This represents the row coordinate of the top-left pixel of the smallest bounding rectangle after the g-th filtering, located in the j-th row and i-th column. This represents the column coordinate of the top-left pixel coordinate of the smallest bounding rectangle after the g-th filtering, located in the j-th row and i-th column;

[0122] Calculate the minimum, maximum, minimum, and maximum longitude of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering, as follows:

[0123]

[0124] in, This represents the minimum longitude of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the maximum longitude of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the minimum dimension of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. represents the maximum latitude of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering, long_min represents the minimum longitude of each point in the intersecting polygon, long_max represents the maximum longitude of each point in the intersecting polygon, lat_min represents the minimum latitude of each point in the intersecting polygon, and lat_max represents the maximum latitude of each point in the intersecting polygon.

[0125] Will( , ), ( , ), ( , ), ( , ) are used as four vertices to construct the geographic rectangle block corresponding to the sub-block in the q-th row and p-th column of the minimum bounding rectangle block after the g-th filtering;

[0126] Step 5 involves calculating the overlap rate between each sub-block of the smallest bounding rectangle after each filtering and the intersecting polygon, as detailed below:

[0127]

[0128] , ,

[0129] Where g represents the sub-block of the smallest bounding rectangle after filtering, K represents the number of the smallest bounding rectangles after filtering, q represents the sub-block number in the column direction, and p represents the sub-block number in the row direction. The number of grid divisions in the column direction is determined by the number of grids. The number of grid divisions in the row direction of the block. This represents the overlap rate of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. Represents the space intersection operation. This indicates how to calculate the area of ​​a polygon. This represents the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering. Represents intersecting polygons;

[0130] like ,but ;

[0131] like ,but 0;

[0132] in, Indicates the overlap rate threshold. This represents the attribute of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering;

[0133] The method for obtaining multiple filtered sub-blocks of the minimum bounding rectangle block after filtering in step 5 is as follows:

[0134] Filter from multiple sub-blocks of the minimum bounding rectangle after filtering. The sub-blocks are obtained by dividing the smallest bounding rectangle into multiple filtered sub-blocks, as defined below:

[0135]

[0136] , , ,

[0137] Where g represents the sub-block of the smallest bounding rectangle after filtering, and K represents the number of the smallest bounding rectangle blocks after filtering. Indicates the sub-block number in the column direction. The sub-block number indicating the row direction. The number of grid divisions in the column direction is determined by the number of grids. The number of grid divisions in the row direction is specified. This represents the h-th sub-block selected from the smallest bounding rectangle block after the g-th filtering, i.e., the h-th sub-block selected from the smallest bounding rectangle block after the g-th filtering. Line number Sub-blocks of the column, This represents the number of sub-blocks filtered within the smallest bounding rectangle block after the g-th filtering.

[0138] Preferably, step 6 involves performing local geometric correction on the first image to obtain multiple geometrically corrected sub-blocks of the first image, as detailed below:

[0139] Based on the smallest bounding rectangle block after the g-th filtering, the first... Line number Sub-blocks of a column The coordinates of the four corresponding vertices ( , ), ( , ), ( , ), ( , ), and resample the region according to the resampling resolution By combining the RPC positioning model of the first image with auxiliary DEM data, an indirect method is used for local geometric correction, thereby obtaining the geometrically corrected sub-block of the first image, defined as follows: , , .

[0140] in, This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number Sub-blocks of a column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number Minimum longitude of the sub-block in the column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number The maximum longitude of the sub-blocks in the column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number The minimum dimension of a sub-block in a column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number The maximum dimension of the sub-blocks in the column. K represents the minimum number of bounding rectangle blocks after filtering. This represents the number of sub-blocks filtered within the smallest bounding rectangle after the g-th filtering.

[0141] Step 6 involves performing local geometric correction on the second image to obtain multiple geometrically corrected sub-blocks of the second image, as detailed below:

[0142] Based on the smallest bounding rectangle block after the g-th filtering, the first... Line number Sub-blocks of a column The coordinates of the four corresponding vertices ( , ), ( , ), ( , ), ( , ), and resample the region according to the resampling resolution By combining the RPC positioning model of the second image with auxiliary DEM data, an indirect method is used for local geometric correction, thereby obtaining the geometrically corrected sub-block of the second image, defined as follows: , , ;

[0143] in, This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number Sub-blocks of a column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number Minimum longitude of the sub-block in the column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number The maximum longitude of the sub-blocks in the column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number The minimum dimension of a sub-block in a column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number The maximum dimension of the sub-blocks in the column. K represents the minimum number of bounding rectangle blocks after filtering. This represents the number of sub-blocks filtered within the smallest bounding rectangle block after the g-th filtering.

[0144] Preferably, step 7 involves using a matching algorithm to obtain multiple sets of corresponding point pairs located at the geographic coordinates of the geometrically corrected sub-block of the first image and the geographic coordinates of the geometrically corrected sub-block of the second image, as follows:

[0145] Match each corrected sub-block of the first image geometry with each corrected sub-block of the second image geometry to obtain the pixel coordinates of the corresponding points on each corrected sub-block of the first image geometry and the pixel coordinates of the corresponding points on each corrected sub-block of the second image geometry.

[0146] For each geometrically corrected sub-block of the first image, the pixel coordinates of the corresponding points are obtained based on the latitude and longitude of the top left corner of the sub-block, and the minimum longitude and minimum latitude values ​​of the sub-block are obtained. The pixel coordinates of the corresponding points are mapped to geographic coordinates by summing the product of the pixel coordinates and the resampling resolution with the minimum value.

[0147] For each geometrically corrected sub-block of the second image, the pixel coordinates of the corresponding points are obtained based on the latitude and longitude of the top left corner of the sub-block, and the minimum longitude and minimum latitude values ​​of the sub-block are obtained. The pixel coordinates of the corresponding points are mapped to geographic coordinates by summing the product of the pixel coordinates and the resampling resolution with the minimum value.

[0148] Step 7 describes mapping the geographic coordinates of each geometrically corrected sub-block of the first image to the corresponding pixel coordinates on the first image:

[0149] The elevation of each geometrically corrected sub-block of the first image is obtained from the DEM auxiliary data by bilinear interpolation based on the geographic coordinates of each geometrically corrected sub-block at the corresponding longitude and latitude.

[0150] The geographic 3D coordinates of each geometrically corrected sub-block of the first image are constructed by combining the geographic coordinates of each geometrically corrected sub-block of the first image and the elevation of each geometrically corrected sub-block of the first image.

[0151] By combining the geographic 3D coordinates of each geometrically corrected sub-block of the first image, the pixel coordinates corresponding to the geographic coordinates of each geometrically corrected sub-block on the first image are obtained by forward calculation using the RPC positioning model of the first image based on the RPC model.

[0152] Step 7, which maps the geographic coordinates of each geometrically corrected sub-block of the second image to the pixel coordinates on the second image, is as follows:

[0153] The elevation of each geometrically corrected sub-block of the second image is obtained from the DEM auxiliary data by bilinear interpolation based on the geographic coordinates of each geometrically corrected sub-block at the corresponding longitude and latitude position.

[0154] The geographic 3D coordinates of each geometrically corrected sub-block of the second image are constructed by combining the geographic coordinates of each geometrically corrected sub-block of the second image with the elevation of each geometrically corrected sub-block of the second image.

[0155] By combining the geographic 3D coordinates of each geometrically corrected sub-block in the second image, the pixel coordinates corresponding to the geographic coordinates of each geometrically corrected sub-block in the second image are obtained by forward calculation using the RPC positioning model of the second image based on the RPC model.

[0156] Preferably, the pixel coordinates of the first image connection point in step 8 are:

[0157] The pixel coordinates of each geometrically corrected sub-block in the first image after gross error removal are matched and then mapped to geographic coordinates.

[0158] Preferably, the pixel coordinates of the second image connection point in step 8 are:

[0159] The pixel coordinates of each geometrically corrected sub-block in the second image after gross error removal are mapped to geographic coordinates.

[0160] The present invention also provides a computer-readable medium storing a computer program executed by an electronic device, which, when run on the electronic device, causes the electronic device to perform the steps of the method for assessing the state of the transformer combination.

[0161] The advantages of this invention lie in its ability to simplify the problem of remote sensing image tie point matching to that of ordinary visual images by dividing the image into a grid based on object-side geographic constraints, effectively eliminating invalid regions by combining overlap rate, and eliminating terrain influences through local geometric correction of sub-blocks. This significantly lowers the barrier to entry for remote sensing image tie point matching. Furthermore, the block-based approach of this method is easy to design in parallel structures, which can further improve tie point matching efficiency. Attached Figure Description

[0162] Figure 1 : Flowchart of the method according to an embodiment of the present invention. Detailed Implementation

[0163] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0164] In specific implementation, the method proposed in the technical solution of this invention can be automatically executed by those skilled in the art using computer software technology. System devices for implementing the method, such as computer-readable storage media storing the corresponding computer program of the technical solution of this invention and computer equipment including the computer program running the corresponding computer program, should also be within the protection scope of this invention.

[0165] The following is combined with Figure 1 The technical solution of this invention is a method for matching tie points in remote sensing images, as described in the following details:

[0166] Step 1: Acquire the first image and its corresponding RPC positioning model. Obtain the pixel width and pixel height of the first image to construct the image-side four corner points of the first image. Use the RPC positioning model of the first image to calculate the object-side geographic coordinates corresponding to the image-side four corner points of the first image. Acquire the second image and its corresponding RPC positioning model. Obtain the pixel width and pixel height of the second image to construct the image-side four corner points of the second image. Use the RPC positioning model of the second image to calculate the object-side geographic coordinates corresponding to the image-side four corner points of the second image.

[0167] Step 2: Construct a vector polygon of the first image using the object-side geographic coordinates corresponding to the four corner points of the image. Construct a vector polygon of the second image using the object-side geographic coordinates corresponding to the four corner points of the image. Calculate the intersecting polygon between the vector polygons of the first and second images using a spatial intersection calculation method and obtain the geographic range of the intersecting polygon. Construct the minimum bounding rectangle of the intersecting polygon based on the geographic range of the intersecting polygon.

[0168] Step 2, obtaining the geographical extent of the intersecting polygons, is detailed below:

[0169] Obtain the longitude and latitude of each point in the intersecting polygon in sequence;

[0170] To construct the longitude range of an intersecting polygon, we select the minimum and maximum longitude values ​​of each point within the polygon from among the longitudes of multiple points in the polygon. This range is defined as follows:

[0171] [long_min, long_max]

[0172] Where long_min represents the minimum longitude of each point in the intersecting polygon, and long_max represents the maximum longitude of each point in the intersecting polygon;

[0173] Among multiple points in an intersecting polygon, the minimum and maximum latitude values ​​of each point in the intersecting polygon are selected as the latitude range of the intersecting polygon, defined as follows:

[0174] [lat_min,lat_max]

[0175] Where lat_min represents the minimum latitude of each point in the intersecting polygon, and lat_max represents the maximum latitude of each point in the intersecting polygon;

[0176] The geographical extent of the intersecting polygons is constructed by the longitude range and the latitude range of the intersecting polygons.

[0177] Step 2 involves constructing the minimum bounding rectangle of the intersecting polygons based on their geographical extent, as detailed below:

[0178] Use (long_min, lat_min), (long_min, lat_max), (long_max, lat_max), (long_min, lat_max) as the four vertices of the minimum bounding rectangle of the intersecting polygon to construct the minimum bounding rectangle of the intersecting polygon.

[0179] Step 3: Combine the DEM auxiliary data with the RPC positioning model of the first image to calculate the image resolution of the first image. Combine the DEM auxiliary data with the RPC positioning model of the second image to calculate the image resolution of the second image. Further obtain the resampling resolution. Calculate the pixel width and pixel height corresponding to the resampling resolution based on the minimum bounding rectangle of the intersecting polygon.

[0180] Step 3 involves calculating the image resolution of the first image using the RPC positioning model. The specific process is as follows:

[0181] The first image is defined as: ImgA;

[0182] Obtain the width of the first image in pixels ImgA, and define it as: ImgA_w;

[0183] Obtain the pixel height of the first image, ImgA, and define it as: ImgA_h;

[0184] The pixel coordinates of the first pixel in the first image are obtained and defined as follows:

[0185]

[0186] Where N represents the pixel step size, ( ) represents the column coordinate of the first pixel in the first image. () represents the row coordinate of the first pixel in the first image. This represents the first pixel of the first image.

[0187] Combining DEM-assisted data with the RPC positioning model of the first image, the 3D geographic coordinates corresponding to the pixel coordinates of the first pixel in the first image are calculated using ray tracing, and are defined as follows:

[0188]

[0189] in, This represents the longitude of the three-dimensional geographic coordinates of the first pixel in the first image. This represents the latitude of the three-dimensional geographic coordinates of the first pixel in the first image. This indicates the elevation of the three-dimensional geographic coordinates of the first pixel in the first image.

[0190] The pixel coordinates of the second pixel in the first image are obtained and defined as follows:

[0191]

[0192] Where N represents the pixel step size, ( ) represents the column coordinate of the second pixel in the first image. () represents the column coordinate of the second pixel in the first image. This represents the second pixel of the second image.

[0193] Combining DEM-assisted data with the RPC positioning model of the first image, the 3D geographic coordinates corresponding to the pixel coordinates of the second pixel point in the first image are calculated using the ray tracing method and defined as follows:

[0194]

[0195] in, This indicates the longitude of the three-dimensional geographic coordinates of the second pixel in the first image. This represents the latitude of the three-dimensional geographic coordinates of the second pixel in the first image. This indicates the elevation of the three-dimensional geographic coordinates of the first pixel in the second image;

[0196] The resolution of the first image is calculated as follows:

[0197]

[0198] in, The resolution of the first image is given by N, where N represents the pixel step size.

[0199] Step 3 involves calculating the image resolution of the second image using the RPC positioning model. The specific process is as follows:

[0200] The second image is defined as: ImgB;

[0201] Obtain the width of the second image in pixels ImgB, defined as: ImgB_w;

[0202] The pixel height of the second image, ImgB, is obtained and defined as: ImgB_h;

[0203] The pixel coordinates of the first pixel in the second image are obtained and defined as follows:

[0204]

[0205] Where N represents the pixel step size, ( ) represents the column coordinate of the first pixel in the second image. () represents the row coordinate of the first pixel in the second image. This represents the first pixel of the second image.

[0206] Combining DEM-assisted data with the RPC positioning model of the second image, the ray tracing method is used to calculate the 3D geographic coordinates corresponding to the pixel coordinates of the first pixel in the second image, defined as:

[0207]

[0208] in, This indicates the longitude of the three-dimensional geographic coordinates of the first pixel in the second image. This represents the latitude of the three-dimensional geographic coordinates of the first pixel in the second image. This indicates the elevation of the three-dimensional geographic coordinates of the first pixel in the second image;

[0209] The pixel coordinates of the second pixel in the second image are obtained, defined as:

[0210]

[0211] Where N represents the pixel step size, ( ) represents the column coordinate of the second pixel in the second image. () represents the column coordinate of the second pixel in the second image. This represents the second pixel of the second image.

[0212] Combining DEM-assisted data with the RPC positioning model of the second image, the 3D geographic coordinates corresponding to the pixel coordinates of the second pixel point in the second image are calculated using the ray tracing method and defined as follows:

[0213]

[0214] in, This indicates the longitude of the three-dimensional geographic coordinates of the second pixel point in the second image. This represents the latitude and three-dimensional geographic coordinates of the second pixel in the second image. This indicates the elevation of the three-dimensional geographic coordinates of the second pixel in the second image;

[0215] The resolution of the second image is calculated as follows:

[0216]

[0217] in, The resolution of the second image is given by N, which represents the pixel step size.

[0218] Step 3 further obtains the resampling resolution, and the specific process is as follows:

[0219] The larger value between the resolution of the first image and the resolution of the second image is used as the resampling resolution, which is defined as follows: ;

[0220] Step 3 involves calculating the pixel width m_W and pixel height m_H at the resampling resolution based on the minimum bounding rectangle of the intersecting polygons, using the following formula:

[0221]

[0222] Where m_W represents the pixel width of the minimum bounding rectangle of the intersecting polygons at the resampling resolution, and m_H represents the pixel height of the minimum bounding rectangle of the intersecting polygons at the resampling resolution. The values ​​represent the resampling resolution, long_min represents the minimum longitude of each point in the intersecting polygon, long_max represents the maximum longitude of each point in the intersecting polygon, lat_min represents the minimum latitude of each point in the intersecting polygon, and lat_max represents the maximum latitude of each point in the intersecting polygon.

[0223] Step 4: Divide the minimum bounding rectangle of the intersecting polygons into multiple minimum bounding rectangle blocks by grid division. Calculate the overlap rate between each minimum bounding rectangle block and the intersecting polygons in turn. Further, select the minimum bounding rectangle blocks with an overlap rate greater than the overlap rate threshold from the multiple minimum bounding rectangle blocks as multiple filtered minimum bounding rectangle blocks.

[0224] Step 4 involves dividing the minimum bounding rectangle of the intersecting polygons into multiple minimum bounding rectangle blocks using a grid. The specific process is as follows:

[0225] Set the number of grid cells in the column direction to be [number]. Set the number of grid divisions in the row direction to be [number]. ;

[0226] Calculate the pixel width and pixel height of each minimum bounding rectangle block, as follows:

[0227]

[0228] in, The pixel width of each block within the minimum bounding rectangle. The pixel height of each block within the minimum bounding rectangle. The number of grid divisions in the column direction, To set the number of grid divisions in the row direction, m_W represents the pixel width corresponding to the minimum bounding rectangle of the intersecting polygons at the resampling resolution, and m_H represents the pixel height corresponding to the minimum bounding rectangle of the intersecting polygons at the resampling resolution.

[0229] The smallest bounding rectangle block in row j and column i is defined as:

[0230]

[0231] ,

[0232] Where i represents the block number in the column direction, and j represents the block number in the row direction. The number of grid divisions in the column direction, To set the number of grid divisions in the row direction;

[0233] The top-left pixel coordinates of the smallest bounding rectangle block in row j and column i are calculated as follows:

[0234]

[0235]

[0236]

[0237] in, The pixel width of each block within the minimum bounding rectangle. The pixel height of each block within the minimum bounding rectangle. Let represent the top-left pixel coordinates of the smallest bounding rectangle block in row j and column i. The row coordinate represents the top-left pixel coordinate of the smallest bounding rectangle block in row j and column i. The column coordinate represents the top-left pixel coordinate of the smallest bounding rectangle block in row j and column i;

[0238] The bottom-right pixel coordinates of the smallest bounding rectangle block in row j and column i are calculated as follows:

[0239]

[0240]

[0241]

[0242] in, This represents the bottom-right pixel coordinates of the smallest bounding rectangle block in row j and column i. The row coordinate represents the bottom-right pixel coordinate of the smallest bounding rectangle block in row j and column i. The column coordinate represents the bottom right pixel coordinate of the smallest bounding rectangle block in row j and column i;

[0243] Calculate the minimum longitude, maximum longitude, minimum latitude, and maximum latitude of the minimum bounding rectangle of the minimum bounding rectangle block in row j and column i, as follows:

[0244]

[0245] in, Let represent the minimum longitude of the smallest bounding rectangle block in row j and column i. This represents the maximum longitude of the smallest bounding rectangular block in row j and column i. This represents the minimum latitude of the smallest bounding rectangle block in the j-th row and i-th column. represents the maximum latitude of the smallest bounding rectangle block in row j and column i, long_min represents the minimum longitude of each point in the intersecting polygon, long_max represents the maximum longitude of each point in the intersecting polygon, lat_min represents the minimum latitude of each point in the intersecting polygon, and lat_max represents the maximum latitude of each point in the intersecting polygon.

[0246] Will( , ), ( , ), ( , ), ( , ) as four vertices to construct the geographic rectangle corresponding to the minimum bounding rectangle of the j-th row and i-th column;

[0247] Step 4 involves sequentially calculating the overlap rate between each minimum bounding rectangle block and the intersecting polygon, as detailed below:

[0248] The overlap rate of the smallest bounding rectangle block in row j and column i is calculated as follows:

[0249]

[0250] ,

[0251] Where i represents the block number in the column direction, and j represents the block number in the row direction. The number of grid divisions in the column direction, To set the number of grid divisions in the row direction, Let represent the overlap rate of the smallest bounding rectangle block in the j-th row and i-th column. Represents the space intersection operation. This indicates the calculation of the area of ​​a polygon. Let represent the smallest bounding rectangle block in row j and column i. Represents intersecting polygons;

[0252] like ,but ;

[0253] like ,but 0;

[0254] in, Indicates the overlap rate threshold. The attribute representing the smallest bounding rectangle block in the j-th row and i-th column;

[0255] The method for obtaining the minimum bounding rectangle block after multiple filtering steps in step 4 is as follows:

[0256] Select from multiple minimum bounding rectangle blocks The minimum bounding rectangle is used to divide the data into multiple filtered minimum bounding rectangles, as defined below:

[0257]

[0258] g∈[1,K], ,

[0259] Where i represents the block number in the column direction, and j represents the block number in the row direction. The number of grid divisions in the column direction, To set the number of grid divisions in the row direction, This represents the smallest bounding rectangle block after the g-th filtering, i.e., the smallest bounding rectangle block after the g-th filtering. Line number The minimum bounding rectangle of the column, where K represents the number of minimum bounding rectangles after filtering;

[0260] Step 5: Perform grid division on multiple filtered minimum bounding rectangle blocks in sequence to obtain multiple sub-blocks of each filtered minimum bounding rectangle block. Calculate the overlap rate between each sub-block of each filtered minimum bounding rectangle block and the intersecting polygon. Further, select the sub-blocks of each filtered minimum bounding rectangle block whose overlap rate with the intersecting polygon is greater than the fine overlap rate threshold from the multiple sub-blocks of each filtered minimum bounding rectangle block as multiple filtered sub-blocks of each filtered minimum bounding rectangle block.

[0261] Step 5 involves sequentially dividing the multiple filtered minimum bounding rectangle blocks into a grid, resulting in multiple sub-blocks for each filtered minimum bounding rectangle block, as detailed below:

[0262] Set the number of grid divisions in the column direction to be as follows: Set the number of grid cells to be divided in the row direction. ;

[0263] Calculate the pixel width and pixel height of each sub-block within the smallest bounding rectangle after filtering, as follows:

[0264]

[0265] in, The pixel width of each sub-block of the smallest bounding rectangle after filtering. The pixel height of each sub-block of the smallest bounding rectangle after filtering; The number of grid divisions in the column direction is determined by the number of grids. The number of grid divisions in the row direction is specified. The pixel width of each block within the minimum bounding rectangle. The pixel height of each block of the smallest bounding rectangle;

[0266] The sub-block in row q and column p of the minimum bounding rectangle after the g-th filtering is defined as:

[0267]

[0268] , ,

[0269] Where g represents the sub-block of the smallest bounding rectangle after filtering, K represents the number of the smallest bounding rectangles after filtering, q represents the sub-block number in the column direction, and p represents the sub-block number in the row direction. The number of grid divisions in the column direction is determined by the number of grids. The number of grid divisions in the row direction of the block.

[0270] The top-left pixel coordinates of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering are calculated as follows:

[0271]

[0272]

[0273]

[0274] in, The pixel width of each sub-block of the smallest bounding rectangle after filtering. The pixel height of each sub-block of the smallest bounding rectangle after filtering. This represents the top-left pixel coordinates of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the row coordinate of the top-left pixel coordinate of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the column coordinate of the top-left pixel coordinate of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering. This represents the row coordinate of the top-left pixel of the smallest bounding rectangle after the g-th filtering, located in the j-th row and i-th column. This represents the column coordinate of the top-left pixel coordinate of the smallest bounding rectangle after the g-th filtering, located in the j-th row and i-th column.

[0275] Calculate the bottom-right pixel coordinates of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering, as follows:

[0276]

[0277]

[0278]

[0279] in, The pixel width of each sub-block of the smallest bounding rectangle after filtering. The pixel height of each sub-block of the smallest bounding rectangle after filtering. This represents the bottom-right pixel coordinates of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the row coordinate of the bottom right pixel of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the column coordinate of the bottom right pixel coordinate of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering. This represents the row coordinate of the top-left pixel of the smallest bounding rectangle after the g-th filtering, located in the j-th row and i-th column. This represents the column coordinate of the top-left pixel coordinate of the smallest bounding rectangle after the g-th filtering, located in the j-th row and i-th column;

[0280] Calculate the minimum, maximum, minimum, and maximum longitude of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering, as follows:

[0281]

[0282] in, This represents the minimum longitude of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the maximum longitude of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the minimum dimension of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. represents the maximum latitude of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering, long_min represents the minimum longitude of each point in the intersecting polygon, long_max represents the maximum longitude of each point in the intersecting polygon, lat_min represents the minimum latitude of each point in the intersecting polygon, and lat_max represents the maximum latitude of each point in the intersecting polygon.

[0283] Will( , ), ( , ), ( , ), ( , ) are used as four vertices to construct the geographic rectangle block corresponding to the sub-block in the q-th row and p-th column of the minimum bounding rectangle block after the g-th filtering;

[0284] Step 5 involves calculating the overlap rate between each sub-block of the smallest bounding rectangle after each filtering and the intersecting polygon, as detailed below:

[0285]

[0286] , ,

[0287] Where g represents the sub-block of the smallest bounding rectangle after filtering, K represents the number of the smallest bounding rectangles after filtering, q represents the sub-block number in the column direction, and p represents the sub-block number in the row direction. The number of grid divisions in the column direction is determined by the number of grids. The number of grid divisions in the row direction of the block. This represents the overlap rate of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. Represents the space intersection operation. This indicates the calculation of the area of ​​a polygon. This represents the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering. Represents intersecting polygons;

[0288] like ,but ;

[0289] like ,but 0;

[0290] in, Indicates the overlap rate threshold. This represents the attribute of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering;

[0291] The method for obtaining multiple filtered sub-blocks of the minimum bounding rectangle block after filtering in step 5 is as follows:

[0292] Filter from multiple sub-blocks of the minimum bounding rectangle after filtering. The sub-blocks are obtained by dividing the smallest bounding rectangle into multiple filtered sub-blocks, as defined below:

[0293]

[0294] , , ,

[0295] Where g represents the sub-block of the smallest bounding rectangle after filtering, and K represents the number of the smallest bounding rectangle blocks after filtering. Indicates the sub-block number in the column direction. The sub-block number indicating the row direction. The number of grid divisions in the column direction is determined by the number of grids. The number of grid divisions in the row direction is specified. This represents the h-th sub-block selected from the smallest bounding rectangle block after the g-th filtering, i.e., the h-th sub-block selected from the smallest bounding rectangle block after the g-th filtering. Line number Sub-blocks of the column, This represents the number of sub-blocks filtered within the smallest bounding rectangle block after the g-th filtering.

[0296] Step 6: Based on each filtered rectangular sub-block of the minimum bounding rectangle block after each filtering, and combined with DEM auxiliary data, perform local geometric correction on the first image to obtain multiple geometrically corrected sub-blocks of the first image. Based on the latitude and longitude of the four corners of each filtered sub-block of the minimum bounding rectangle block after each filtering, and combined with DEM auxiliary data, perform local geometric correction on the second image to obtain multiple geometrically corrected sub-blocks of the second image.

[0297] Step 6 involves performing local geometric correction on the first image to obtain multiple geometrically corrected sub-blocks of the first image, as detailed below:

[0298] Based on the smallest bounding rectangle block after the g-th filtering, the first... Line number Sub-blocks of a column The coordinates of the four corresponding vertices ( , ), ( , ), ( , ), ( , ), and resample the region according to the resampling resolution By combining the RPC positioning model of the first image with auxiliary DEM data, an indirect method is used for local geometric correction, thereby obtaining the geometrically corrected sub-block of the first image, defined as follows: , , .

[0299] in, This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number Sub-blocks of a column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number Minimum longitude of the sub-block in the column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number The maximum longitude of the sub-blocks in the column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number The minimum dimension of a sub-block in a column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number The maximum dimension of the sub-blocks in the column. K represents the minimum number of bounding rectangle blocks after filtering. This represents the number of sub-blocks filtered within the smallest bounding rectangle after the g-th filtering.

[0300] Step 6 involves performing local geometric correction on the second image to obtain multiple geometrically corrected sub-blocks of the second image, as detailed below:

[0301] Based on the smallest bounding rectangle block after the g-th filtering, the first... Line number Sub-blocks of a column The coordinates of the four corresponding vertices ( , ), ( , ), ( , ), ( , ), and resample the region according to the resampling resolution By combining the RPC positioning model of the second image with auxiliary DEM data, an indirect method is used for local geometric correction, thereby obtaining the geometrically corrected sub-block of the second image, defined as follows: , , ;

[0302] in, This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number Sub-blocks of a column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number Minimum longitude of the sub-block in the column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number The maximum longitude of the sub-blocks in the column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number The minimum dimension of a sub-block in a column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number The maximum dimension of the sub-blocks in the column. K represents the minimum number of bounding rectangle blocks after filtering. This represents the number of sub-blocks filtered within the smallest bounding rectangle block after the g-th filtering.

[0303] Step 7: Using each geometrically corrected sub-block of the first image as the reference image and the geometrically corrected sub-block of the second image as the registration image, a matching algorithm is used to obtain multiple sets of corresponding point pairs located at the geographic coordinates of the geometrically corrected sub-blocks of the first and second images; the geographic coordinates of each geometrically corrected sub-block of the first image are mapped to the corresponding pixel coordinates on the first image, and the geographic coordinates of each geometrically corrected sub-block of the second image are mapped to the corresponding pixel coordinates on the second image.

[0304] Step 7 describes the use of a matching algorithm to obtain multiple pairs of corresponding point coordinates located in the geometrically corrected sub-blocks of the first and second imagery, specifically as follows:

[0305] Match each corrected sub-block of the first image geometry with each corrected sub-block of the second image geometry to obtain the pixel coordinates of the corresponding points on each corrected sub-block of the first image geometry and the pixel coordinates of the corresponding points on each corrected sub-block of the second image geometry.

[0306] For each geometrically corrected sub-block of the first image, the pixel coordinates of the corresponding points are obtained based on the latitude and longitude of the top left corner of the sub-block, and the minimum longitude and minimum latitude values ​​of the sub-block are obtained. The pixel coordinates of the corresponding points are mapped to geographic coordinates by summing the product of the pixel coordinates and the resampling resolution with the minimum value.

[0307] For each geometrically corrected sub-block of the second image, the pixel coordinates of the corresponding points are obtained based on the latitude and longitude of the top left corner of the sub-block, and the minimum longitude and minimum latitude values ​​of the sub-block are obtained. The pixel coordinates of the corresponding points are mapped to geographic coordinates by summing the product of the pixel coordinates and the resampling resolution with the minimum value.

[0308] Step 7 describes mapping the geographic coordinates of each geometrically corrected sub-block of the first image to the corresponding pixel coordinates on the first image:

[0309] The elevation of each geometrically corrected sub-block of the first image is obtained from the DEM auxiliary data by bilinear interpolation based on the geographic coordinates of each geometrically corrected sub-block at the corresponding longitude and latitude.

[0310] The geographic 3D coordinates of each geometrically corrected sub-block of the first image are constructed by combining the geographic coordinates of each geometrically corrected sub-block of the first image and the elevation of each geometrically corrected sub-block of the first image.

[0311] By combining the geographic 3D coordinates of each geometrically corrected sub-block of the first image, the pixel coordinates corresponding to the geographic coordinates of each geometrically corrected sub-block on the first image are obtained by forward calculation using the RPC positioning model of the first image based on the RPC model.

[0312] Step 7, which maps the geographic coordinates of each geometrically corrected sub-block of the second image to the pixel coordinates on the second image, is as follows:

[0313] The elevation of each geometrically corrected sub-block of the second image is obtained from the DEM auxiliary data by bilinear interpolation based on the geographic coordinates of each geometrically corrected sub-block at the corresponding longitude and latitude position.

[0314] The geographic 3D coordinates of each geometrically corrected sub-block of the second image are constructed by combining the geographic coordinates of each geometrically corrected sub-block of the second image with the elevation of each geometrically corrected sub-block of the second image.

[0315] By combining the geographic 3D coordinates of each geometrically corrected sub-block in the second image, the pixel coordinates corresponding to the geographic coordinates of each geometrically corrected sub-block in the second image are obtained by forward calculation using the RPC positioning model of the second image based on the RPC model.

[0316] Step 8: Use a random sampling consensus algorithm to remove outliers from the pixel coordinates corresponding to the geographic coordinates of each geometrically corrected sub-block in the first image and the pixel coordinates corresponding to the geographic coordinates of each geometrically corrected sub-block in the second image, to obtain the pixel coordinates of the connection points in the first image and the pixel coordinates of the connection points in the second image.

[0317] Step 8: The pixel coordinates of the first image connection point are:

[0318] The pixel coordinates of each geometrically corrected sub-block in the first image after gross error removal are matched and then mapped to geographic coordinates.

[0319] Step 8: The pixel coordinates of the second image connection point are:

[0320] The pixel coordinates of each geometrically corrected sub-block in the second image after gross error removal are mapped to geographic coordinates.

[0321] A specific embodiment of the present invention also provides a computer-readable medium.

[0322] The computer-readable medium is a server workstation;

[0323] The server workstation stores a computer program executed by the electronic device. When the computer program runs on the electronic device, it causes the electronic device to perform the steps of the remote sensing image tie point matching method of the present invention.

[0324] It should be understood that any parts not described in detail in this specification belong to the prior art.

[0325] It should be understood that the above description of the preferred embodiments is quite detailed, but it should not be considered as a limitation on the scope of protection of this invention. Those skilled in the art, under the guidance of this invention, can make substitutions or modifications without departing from the scope of protection of the claims of this invention, and all such substitutions or modifications fall within the scope of protection of this invention. The scope of protection of this invention should be determined by the appended claims.

Claims

1. A method for matching tie points in remote sensing images, characterized in that, Includes the following steps: Step 1: Acquire the first image and its corresponding RPC positioning model. Obtain the pixel width and pixel height of the first image to construct the image-side four corner points of the first image. Use the RPC positioning model of the first image to calculate the object-side geographic coordinates corresponding to the image-side four corner points of the first image. Acquire the second image and its corresponding RPC positioning model. Obtain the pixel width and pixel height of the second image to construct the image-side four corner points of the second image. Use the RPC positioning model of the second image to calculate the object-side geographic coordinates corresponding to the image-side four corner points of the second image. Step 2: Construct a vector polygon of the first image using the object-side geographic coordinates corresponding to the four corner points of the image. Construct a vector polygon of the second image using the object-side geographic coordinates corresponding to the four corner points of the image. Calculate the intersecting polygon between the vector polygons of the first and second images using a spatial intersection calculation method and obtain the geographic range of the intersecting polygon. Construct the minimum bounding rectangle of the intersecting polygon based on the geographic range of the intersecting polygon. Step 3: Combine the DEM auxiliary data with the RPC positioning model of the first image to calculate the image resolution of the first image. Combine the DEM auxiliary data with the RPC positioning model of the second image to calculate the image resolution of the second image. Further obtain the resampling resolution. Calculate the pixel width and pixel height corresponding to the resampling resolution based on the minimum bounding rectangle of the intersecting polygon. Step 4: Divide the minimum bounding rectangle of the intersecting polygons into multiple minimum bounding rectangle blocks by grid division. Calculate the overlap rate between each minimum bounding rectangle block and the intersecting polygons in turn. Further, select the minimum bounding rectangle blocks with an overlap rate greater than the overlap rate threshold from the multiple minimum bounding rectangle blocks as multiple filtered minimum bounding rectangle blocks. Step 5: Perform grid division on multiple filtered minimum bounding rectangle blocks in sequence to obtain multiple sub-blocks of each filtered minimum bounding rectangle block. Calculate the overlap rate between each sub-block of each filtered minimum bounding rectangle block and the intersecting polygon. Further, select the sub-blocks of each filtered minimum bounding rectangle block whose overlap rate with the intersecting polygon is greater than the fine overlap rate threshold from the multiple sub-blocks of each filtered minimum bounding rectangle block as multiple filtered sub-blocks of each filtered minimum bounding rectangle block. Step 6: Based on each filtered rectangular sub-block of the minimum bounding rectangle block after each filtering, and combined with DEM auxiliary data, perform local geometric correction on the first image to obtain multiple geometrically corrected sub-blocks of the first image. Based on the latitude and longitude of the four corners of each filtered sub-block of the minimum bounding rectangle block after each filtering, and combined with DEM auxiliary data, perform local geometric correction on the second image to obtain multiple geometrically corrected sub-blocks of the second image. Step 7: Using each geometrically corrected sub-block of the first image as the reference image and the geometrically corrected sub-block of the second image as the registration image, a matching algorithm is used to obtain multiple sets of corresponding point pairs located at the geographic coordinates of the geometrically corrected sub-blocks of the first and second images; the geographic coordinates of each geometrically corrected sub-block of the first image are mapped to the corresponding pixel coordinates on the first image, and the geographic coordinates of each geometrically corrected sub-block of the second image are mapped to the corresponding pixel coordinates on the second image. Step 8: Use a random sampling consensus algorithm to remove outliers from the pixel coordinates corresponding to the geographic coordinates of each geometrically corrected sub-block in the first image and the pixel coordinates corresponding to the geographic coordinates of each geometrically corrected sub-block in the second image, to obtain the pixel coordinates of the connection points in the first image and the pixel coordinates of the connection points in the second image.

2. The remote sensing image tie point matching method according to claim 1, characterized in that, Step 2, obtaining the geographical extent of the intersecting polygons, is detailed below: Obtain the longitude and latitude of each point in the intersecting polygon in sequence; To construct the longitude range of an intersecting polygon, we select the minimum and maximum longitude values ​​of each point within the polygon from among the longitudes of multiple points in the polygon. This range is defined as follows: [long_min, long_max] Where long_min represents the minimum longitude of each point in the intersecting polygon, and long_max represents the maximum longitude of each point in the intersecting polygon; In an intersecting polygon, the minimum and maximum latitude values ​​of each point are selected from the latitude values ​​of multiple points within the polygon. These are then used as the latitude range of the intersecting polygon, defined as follows: [lat_min,lat_max] Where lat_min represents the minimum latitude of each point in the intersecting polygon, and lat_max represents the maximum latitude of each point in the intersecting polygon; The geographical extent of the intersecting polygons is constructed by the longitude range and the latitude range of the intersecting polygons. Step 2 involves constructing the minimum bounding rectangle of the intersecting polygons based on their geographical extent, as detailed below: Use (long_min, lat_min), (long_min, lat_max), (long_max, lat_max), and (long_min, lat_max) as the four vertices of the minimum bounding rectangle of the intersecting polygon to construct the minimum bounding rectangle of the intersecting polygon.

3. The remote sensing image tie point matching method according to claim 2, characterized in that, Step 3 involves calculating the image resolution of the first image using the RPC positioning model. The specific process is as follows: The first image is defined as: ImgA; Obtain the width of the first image in pixels ImgA, and define it as: ImgA_w; Obtain the pixel height of the first image, ImgA, and define it as: ImgA_h; The pixel coordinates of the first pixel in the first image are obtained and defined as follows: Where N represents the pixel step size, ( ) represents the column coordinate of the first pixel in the first image. () represents the row coordinate of the first pixel in the first image. This represents the first pixel of the first image. Combining DEM-assisted data with the RPC positioning model of the first image, the 3D geographic coordinates corresponding to the pixel coordinates of the first pixel in the first image are calculated using ray tracing, and are defined as follows: in, This represents the longitude of the three-dimensional geographic coordinates of the first pixel in the first image. This represents the latitude of the three-dimensional geographic coordinates of the first pixel in the first image. This indicates the elevation of the three-dimensional geographic coordinates of the first pixel in the first image. The pixel coordinates of the second pixel in the first image are obtained and defined as follows: Where N represents the pixel step size, ( ) represents the column coordinate of the second pixel in the first image. () represents the column coordinate of the second pixel in the first image. This represents the second pixel of the second image. Combining DEM-assisted data with the RPC positioning model of the first image, the 3D geographic coordinates corresponding to the pixel coordinates of the second pixel point in the first image are calculated using the ray tracing method and defined as follows: in, This indicates the longitude of the three-dimensional geographic coordinates of the second pixel in the first image. This represents the latitude of the three-dimensional geographic coordinates of the second pixel in the first image. This indicates the elevation of the three-dimensional geographic coordinates of the first pixel in the second image; The resolution of the first image is calculated as follows: in, The resolution of the first image is given by N, where N represents the pixel step size. Step 3 involves calculating the image resolution of the second image using the RPC positioning model. The specific process is as follows: The second image is defined as: ImgB; Obtain the width of the second image in pixels ImgB, defined as: ImgB_w; The pixel height of the second image, ImgB, is obtained and defined as: ImgB_h; The pixel coordinates of the first pixel in the second image are obtained and defined as follows: Where N represents the pixel step size, ( ) represents the column coordinate of the first pixel in the second image. () represents the row coordinate of the first pixel in the second image. This represents the first pixel of the second image. Combining DEM-assisted data with the RPC positioning model of the second image, the ray tracing method is used to calculate the 3D geographic coordinates corresponding to the pixel coordinates of the first pixel in the second image, defined as: in, This indicates the longitude of the three-dimensional geographic coordinates of the first pixel in the second image. This represents the latitude and three-dimensional geographic coordinates of the first pixel in the second image. This indicates the elevation of the three-dimensional geographic coordinates of the first pixel in the second image; The pixel coordinates of the second pixel in the second image are obtained and defined as follows: Where N represents the pixel step size, ( ) represents the column coordinate of the second pixel in the second image. () represents the column coordinate of the second pixel in the second image. This represents the second pixel of the second image. Combining DEM-assisted data with the RPC positioning model of the second image, the 3D geographic coordinates corresponding to the pixel coordinates of the second pixel point in the second image are calculated using the ray tracing method and defined as follows: in, This indicates the longitude of the three-dimensional geographic coordinates of the second pixel point in the second image. This represents the latitude and three-dimensional geographic coordinates of the second pixel in the second image. This indicates the elevation of the three-dimensional geographic coordinates of the second pixel in the second image; The resolution of the second image is calculated as follows: in, The resolution of the second image is given by N, which represents the pixel step size. Step 3 further obtains the resampling resolution, and the specific process is as follows: The larger value between the resolution of the first image and the resolution of the second image is used as the resampling resolution, which is defined as follows: ; Step 3 involves calculating the pixel width m_W and pixel height m_H at the resampling resolution based on the minimum bounding rectangle of the intersecting polygons, using the following formula: Where m_W represents the pixel width of the minimum bounding rectangle of the intersecting polygons at the resampling resolution, and m_H represents the pixel height of the minimum bounding rectangle of the intersecting polygons at the resampling resolution. The values ​​represent the resampling resolution, long_min represents the minimum longitude of each point in the intersecting polygon, long_max represents the maximum longitude of each point in the intersecting polygon, lat_min represents the minimum latitude of each point in the intersecting polygon, and lat_max represents the maximum latitude of each point in the intersecting polygon.

4. The remote sensing image tie point matching method according to claim 3, characterized in that, Step 4 involves dividing the minimum bounding rectangle of the intersecting polygons into multiple minimum bounding rectangle blocks using a grid. The specific process is as follows: Set the number of grid divisions in the column direction to be as follows Set the number of grid divisions in the row direction to be [number]. ; Calculate the pixel width and pixel height of each minimum bounding rectangle block, as follows: in, The pixel width of each block within the minimum bounding rectangle. The pixel height of each block of the minimum bounding rectangle. The number of grid divisions in the column direction, To set the number of grid divisions in the row direction, m_W represents the pixel width corresponding to the minimum bounding rectangle of the intersecting polygons at the resampling resolution, and m_H represents the pixel height corresponding to the minimum bounding rectangle of the intersecting polygons at the resampling resolution. The smallest bounding rectangle block in row j and column i is defined as: ， Where i represents the block number in the column direction, and j represents the block number in the row direction. The number of grid divisions in the column direction, To set the number of grid divisions in the row direction; The top-left pixel coordinates of the smallest bounding rectangle block in row j and column i are calculated as follows: in, The pixel width of each block within the minimum bounding rectangle. The pixel height of each block of the minimum bounding rectangle. Let represent the top-left pixel coordinates of the smallest bounding rectangle block in row j and column i. The row coordinate represents the top-left pixel coordinate of the smallest bounding rectangle block in row j and column i. The column coordinate represents the top-left pixel coordinate of the smallest bounding rectangle block in row j and column i; The bottom-right pixel coordinates of the smallest bounding rectangle block in row j and column i are calculated as follows: in, This represents the bottom-right pixel coordinates of the smallest bounding rectangle block in row j and column i. The row coordinate represents the bottom-right pixel coordinate of the smallest bounding rectangle block in row j and column i. The column coordinate represents the bottom right pixel coordinate of the smallest bounding rectangle block in row j and column i; Calculate the minimum longitude, maximum longitude, minimum latitude, and maximum latitude of the minimum bounding rectangle of the minimum bounding rectangle block in row j and column i, as follows: in, Let represent the minimum longitude of the smallest bounding rectangle block in row j and column i. This represents the maximum longitude of the smallest bounding rectangular block in row j and column i. This represents the minimum latitude of the smallest bounding rectangle block in the j-th row and i-th column. represents the maximum latitude of the smallest bounding rectangle block in row j and column i, long_min represents the minimum longitude of each point in the intersecting polygon, long_max represents the maximum longitude of each point in the intersecting polygon, lat_min represents the minimum latitude of each point in the intersecting polygon, and lat_max represents the maximum latitude of each point in the intersecting polygon. Will( , ), ( , ), ( , ), ( , ) are used as four vertices to construct the geographic rectangle corresponding to the minimum bounding rectangle of the j-th row and i-th column.

5. The remote sensing image tie point matching method according to claim 4, characterized in that, Step 4 involves sequentially calculating the overlap rate between each minimum bounding rectangle block and the intersecting polygon, as detailed below: The overlap rate of the smallest bounding rectangle block in row j and column i is calculated as follows: ， Where i represents the block number in the column direction, and j represents the block number in the row direction. The number of grid divisions in the column direction, To set the number of grid divisions in the row direction, Let represent the overlap rate of the smallest bounding rectangle block in the j-th row and i-th column. Represents the space intersection operation. This indicates the calculation of the area of ​​a polygon. Let represent the smallest bounding rectangle block in row j and column i. Represents intersecting polygons; like ,but ; like ,but 0; in, Indicates the overlap rate threshold. The attribute representing the smallest bounding rectangle block in the j-th row and i-th column; The method for obtaining the minimum bounding rectangle block after multiple filtering steps in step 4 is as follows: Select from multiple minimum bounding rectangle blocks The minimum bounding rectangle is used to divide the data into multiple filtered minimum bounding rectangles, as defined below: g∈[1,K], , Where i represents the block number in the column direction, and j represents the block number in the row direction. The number of grid divisions in the column direction, To set the number of grid divisions in the row direction, This represents the smallest bounding rectangle block after the g-th filtering, i.e., the smallest bounding rectangle block after the g-th filtering. Line number The smallest bounding rectangle of the column, where K represents the number of smallest bounding rectangles after filtering.

6. The remote sensing image tie point matching method according to claim 5, characterized in that, Step 5 involves sequentially dividing the multiple filtered minimum bounding rectangle blocks into a grid, resulting in multiple sub-blocks for each filtered minimum bounding rectangle block, as detailed below: Set the number of grid divisions in the column direction to be as follows: Set the number of grid cells to be divided in the row direction. ; Calculate the pixel width and pixel height of each sub-block within the smallest bounding rectangle after filtering, as follows: in, The pixel width of each sub-block of the smallest bounding rectangle after filtering. The pixel height of each sub-block of the smallest bounding rectangle after filtering; The number of grid divisions in the column direction is determined by the number of grids. The number of grid divisions in the row direction is specified. The pixel width of each block within the minimum bounding rectangle. The pixel height of each block of the smallest bounding rectangle; The sub-block in row q and column p of the minimum bounding rectangle after the g-th filtering is defined as: ， ， Where g represents the sub-block of the smallest bounding rectangle after filtering, K represents the number of the smallest bounding rectangles after filtering, q represents the sub-block number in the column direction, and p represents the sub-block number in the row direction. The number of grid divisions in the column direction is determined by the number of grids. Determine the number of grid divisions in the row direction of the block; The top-left pixel coordinates of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering are calculated as follows: in, The pixel width of each sub-block of the smallest bounding rectangle after filtering. The pixel height of each sub-block of the smallest bounding rectangle after filtering. This represents the top-left pixel coordinates of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the row coordinate of the top-left pixel coordinate of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the column coordinate of the top-left pixel coordinate of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the row coordinate of the top-left pixel of the smallest bounding rectangle after the g-th filtering, located in the j-th row and i-th column. This represents the column coordinate of the top-left pixel coordinate of the smallest bounding rectangle after the g-th filtering, located in the j-th row and i-th column; Calculate the bottom-right pixel coordinates of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering, as follows: in, The pixel width of each sub-block of the smallest bounding rectangle after filtering. The pixel height of each sub-block of the smallest bounding rectangle after filtering. This represents the bottom-right pixel coordinates of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the row coordinate of the bottom right pixel of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the column coordinate of the bottom right pixel coordinate of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering. This represents the row coordinate of the top-left pixel of the smallest bounding rectangle after the g-th filtering, located in the j-th row and i-th column. This represents the column coordinate of the top-left pixel coordinate of the smallest bounding rectangle after the g-th filtering, located in the j-th row and i-th column; Calculate the minimum, maximum, minimum, and maximum longitude of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering, as follows: in, This represents the minimum longitude of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the maximum longitude of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. This represents the minimum dimension of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. represents the maximum latitude of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering, long_min represents the minimum longitude of each point in the intersecting polygon, long_max represents the maximum longitude of each point in the intersecting polygon, lat_min represents the minimum latitude of each point in the intersecting polygon, and lat_max represents the maximum latitude of each point in the intersecting polygon. Will( , ), ( , ), ( , ), ( , ) are used as four vertices to construct the geographic rectangle block corresponding to the sub-block in row q and column p of the minimum bounding rectangle block after the g-th filtering.

7. The remote sensing image tie point matching method according to claim 6, characterized in that, Step 5 involves calculating the overlap rate between each sub-block of the smallest bounding rectangle after each filtering and the intersecting polygon, as detailed below: ， ， Where g represents the sub-block of the smallest bounding rectangle after filtering, K represents the number of the smallest bounding rectangles after filtering, q represents the sub-block number in the column direction, and p represents the sub-block number in the row direction. The number of grid divisions in the column direction is determined by the number of grids. Determine the number of grid divisions in the row direction of the block; This represents the overlap rate of the sub-block in the q-th row and p-th column within the smallest bounding rectangle after the g-th filtering. Represents the space intersection operation. This indicates the calculation of the area of ​​a polygon. This represents the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering. Represents intersecting polygons; like ,but ; like ,but 0; in, Indicates the overlap rate threshold. This represents the attribute of the sub-block in the q-th row and p-th column of the smallest bounding rectangle after the g-th filtering; The method for obtaining multiple filtered sub-blocks of the minimum bounding rectangle block after filtering in step 5 is as follows: Filter from multiple sub-blocks of the minimum bounding rectangle after filtering. The sub-blocks are obtained by dividing the smallest bounding rectangle into multiple filtered sub-blocks, as defined below: ， ， ， Where g represents the sub-block of the smallest bounding rectangle after filtering, and K represents the number of the smallest bounding rectangle blocks after filtering. Indicates the sub-block number in the column direction. The sub-block number indicating the row direction. The number of grid divisions in the column direction is determined by the number of grids. The number of grid divisions in the row direction is specified. This represents the h-th sub-block selected from the smallest bounding rectangle block after the g-th filtering, which is the h-th sub-block selected from the smallest bounding rectangle block after the g-th filtering. Line number Sub-blocks of the column, This represents the number of sub-blocks filtered within the smallest bounding rectangle after the g-th filtering.

8. The remote sensing image tie point matching method according to claim 7, characterized in that, Step 6 involves performing local geometric correction on the first image to obtain multiple geometrically corrected sub-blocks of the first image, as detailed below: Based on the smallest bounding rectangle block after the g-th filtering, the first... Line number Sub-blocks of a column The coordinates of the four corresponding vertices ( , ), ( , ), ( , ), ( , ), and resample the region according to the resampling resolution By combining the RPC positioning model of the first image with auxiliary DEM data, an indirect method is used for local geometric correction, thereby obtaining the geometrically corrected sub-block of the first image, defined as follows: , , ; in, This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number Sub-blocks of a column; This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number Minimum longitude of the sub-block in the column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number The maximum longitude of the sub-blocks in the column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number The minimum dimension of a sub-block in a column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number The maximum dimension of the sub-blocks in the column; K represents the minimum number of bounding rectangle blocks after filtering. This represents the number of sub-blocks filtered within the smallest bounding rectangle after the g-th filtering. Step 6 involves performing local geometric correction on the second image to obtain multiple geometrically corrected sub-blocks of the second image, as detailed below: Based on the smallest bounding rectangle block after the g-th filtering, the first... Line number Sub-blocks of a column The coordinates of the four corresponding vertices ( , ), ( , ), ( , ), ( , ), for this sub-block Regions according to resampling resolution By combining the RPC positioning model of the second image with auxiliary DEM data, an indirect method is used for local geometric correction, thereby obtaining the geometrically corrected sub-block of the second image, defined as follows: , , ; in, This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number Sub-blocks of a column; This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number Minimum longitude of the sub-block in the column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number The maximum longitude of the sub-blocks in the column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number The minimum dimension of a sub-block in a column. This represents the filtered element in the smallest bounding rectangle block after the g-th filtering. Line number The maximum dimension of the sub-blocks in the column; K represents the minimum number of bounding rectangle blocks after filtering. This represents the number of sub-blocks filtered within the smallest bounding rectangle after the g-th filtering.

9. The remote sensing image tie point matching method according to claim 8, characterized in that, Step 7 describes the use of a matching algorithm to obtain multiple pairs of corresponding point coordinates located in the geometrically corrected sub-blocks of the first and second imagery, specifically as follows: Match each corrected sub-block of the first image geometry with each corrected sub-block of the second image geometry to obtain the pixel coordinates of the corresponding points on each corrected sub-block of the first image geometry and the pixel coordinates of the corresponding points on each corrected sub-block of the second image geometry. For each geometrically corrected sub-block of the first image, the pixel coordinates of the corresponding points are obtained based on the latitude and longitude of the top left corner of the sub-block, and the minimum longitude and minimum latitude values ​​of the sub-block are obtained. The pixel coordinates of the corresponding points are mapped to geographic coordinates by summing the product of the pixel coordinates and the resampling resolution with the minimum value. For each geometrically corrected sub-block of the second image, the pixel coordinates of the corresponding points are obtained based on the latitude and longitude of the top left corner of the sub-block, and the minimum longitude and minimum latitude values ​​of the sub-block are obtained. The pixel coordinates of the corresponding points are mapped to geographic coordinates by summing the product of the pixel coordinates and the resampling resolution with the minimum value. Step 7 describes mapping the geographic coordinates of each geometrically corrected sub-block of the first image to the corresponding pixel coordinates on the first image: The elevation of each geometrically corrected sub-block of the first image is obtained from the DEM auxiliary data by bilinear interpolation based on the geographic coordinates of each geometrically corrected sub-block at the corresponding longitude and latitude. The geographic 3D coordinates of each geometrically corrected sub-block of the first image are constructed by combining the geographic coordinates of each geometrically corrected sub-block of the first image and the elevation of each geometrically corrected sub-block of the first image. By combining the geographic 3D coordinates of each geometrically corrected sub-block of the first image, the pixel coordinates corresponding to the geographic coordinates of each geometrically corrected sub-block on the first image are obtained by forward calculation using the RPC positioning model of the first image based on the RPC model. Step 7 maps the geographic coordinates of each geometrically corrected sub-block of the second image to the pixel coordinates on the second image as follows: The elevation of each geometrically corrected sub-block of the second image is obtained from the DEM auxiliary data by bilinear interpolation based on the geographic coordinates of each geometrically corrected sub-block at the corresponding longitude and latitude position. The geographic 3D coordinates of each geometrically corrected sub-block of the second image are constructed by combining the geographic coordinates of each geometrically corrected sub-block of the second image with the elevation of each geometrically corrected sub-block of the second image. By combining the geographic 3D coordinates of each geometrically corrected sub-block in the second image, the pixel coordinates corresponding to the geographic coordinates of each geometrically corrected sub-block in the second image are obtained by forward calculation using the RPC positioning model of the second image based on the RPC model.

10. A computer-readable medium, characterized in that, It stores a computer program executed by an electronic device, which, when run on the electronic device, causes the electronic device to perform the steps of the method as described in any one of claims 1-9.

Images