Quantitative Quality Inspection Method, Device and Electronic Equipment for Internal Geometric Distortion of Remote Sensing Images

Abstract

The present invention provides a method, device and electronic equipment for quantitatively inspecting internal geometric distortion of remote sensing images, belonging to the technical field of remote sensing image inspection. In this method, each side of the triangular network in the remote sensing image to be inspected is distinguished by different colors to represent the distribution of the internal geometric distortion of the remote sensing image to be inspected. Therefore, users can visually (i.e., apparently) and accurately capture the internal geometric distortion situation of the remote sensing image to be inspected. Moreover, the mean square error of the internal geometric distortion of the remote sensing image to be inspected and the gross error rate of the remote sensing image to be inspected are used as the inspection results of the internal geometric distortion of the remote sensing image to be inspected, that is, the internal geometric distortion value of the remote sensing image to be inspected can be quantitatively described.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing image quality inspection, and in particular, to a method, device, and electronic device for quantitatively inspecting internal geometric distortion of remote sensing images. Background Art

[0002] Due to various factors such as the distortion of the camera lens, atmospheric refraction, the curvature of the earth, terrain undulation, aerial triangulation encryption, and image interpolation, the original satellite remote sensing images and orthorectified remote sensing images will bring internal geometric distortion to the remote sensing images, which affects the most important measurement accuracy of the remote sensing images. Before applying the remote sensing images, it is often necessary to conduct quality inspection on the internal geometric accuracy of the remote sensing images.

[0003] Most traditional automated accuracy quality inspections of remote sensing images are carried out by automatically matching the homologous points of the remote sensing image to be inspected and the reference reference image (that is, the feature points in the remote sensing image to be inspected and the matching points corresponding to the feature points in the reference reference image), and then using the homologous points as the basic unit to calculate the absolute geometric accuracy error of each homologous point (that is, the position coordinate difference between any feature point A in the remote sensing image to be inspected and the matching point A' in the reference reference image that matches the above feature point A), and then automatically statistically obtaining the root mean square error of the absolute geometric accuracy of the entire remote sensing image to be inspected based on the absolute geometric accuracy errors of all the homologous points in the remote sensing image to be inspected, and automatically forming an accuracy quality inspection report. In the above process of automated accuracy quality inspection of remote sensing images, only the root mean square error of the absolute geometric accuracy of the remote sensing image to be inspected (that is, the geometric accuracy error between the remote sensing image to be inspected and the reference reference image) is inspected, but the internal geometric distortion of the remote sensing image to be inspected is not inspected, and the internal geometric distortion (that is, the geometric accuracy error between different feature points in the remote sensing image to be inspected) is crucial for the measurement accuracy of the remote sensing image.

[0004] In summary, how to visually represent the internal geometric distortion of the remote sensing image to be inspected and quantitatively calculate the internal geometric distortion of the remote sensing image to be inspected has become a technical problem that needs to be solved urgently at present. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide a method, device, and electronic device for quantitatively inspecting internal geometric distortion of remote sensing images, so as to alleviate the technical problems that the prior art cannot visually represent the internal geometric distortion of the remote sensing image to be inspected and cannot quantitatively calculate the internal geometric distortion of the remote sensing image to be inspected.

[0006] In the first aspect, an embodiment of the present invention provides a method for quantitatively inspecting internal geometric distortion of remote sensing images, including:

[0007] Obtain corresponding points of the remotely sensed image to be inspected and the reference image, where the reference image is a standard ortho-remotely sensed image, and the corresponding points are points with the same ground object position in the remotely sensed image to be inspected and the reference image;

[0008] Convert the geographic coordinates of the reference image among the corresponding points to obtain the theoretical pixel coordinates of the geographic coordinates of the reference image on the remotely sensed image to be inspected, and calculate the pixel error of each corresponding point according to each theoretical pixel coordinate and the pixel coordinate of the remotely sensed image to be inspected among the corresponding points;

[0009] Construct a triangular network in the remotely sensed image to be inspected based on the pixel coordinates of the remotely sensed image to be inspected among the corresponding points, and calculate the edge error of each side of each triangle in the triangular network according to the pixel error of each corresponding point;

[0010] Distinguish each side in the triangular network with different colors based on each edge error to characterize the distribution of the internal geometric distortion of the remotely sensed image to be inspected;

[0011] Calculate the mean square error of the internal geometric distortion of the remotely sensed image to be inspected according to each edge error, and calculate the gross error rate of the remotely sensed image to be inspected based on each edge error, a preset internal distortion error threshold, and the total number of sides in the triangular network;

[0012] Take the mean square error of the internal geometric distortion of the remotely sensed image to be inspected and the gross error rate of the remotely sensed image to be inspected as the quality inspection results of the internal geometric distortion of the remotely sensed image to be inspected.

[0013] Further, the remotely sensed image to be inspected includes any one of the following: a raw remotely sensed image with an initial positioning model file, an ortho-remotely sensed image. Obtaining the corresponding points of the remotely sensed image to be inspected and the reference image includes:

[0014] Divide the remotely sensed image to be inspected into grids to obtain a plurality of grids;

[0015] Use the Harris feature point operator to extract feature points from each grid to obtain the feature points of each grid;

[0016] Determine the matching points in the reference image that match each of the feature points, and use each of the feature points and its matching point as an initial homologous point. Among them, the initial homologous points include: homologous point number, column coordinate of the remote sensing image to be inspected, row coordinate of the remote sensing image to be inspected, when the remote sensing image to be inspected is an ortho-remote sensing image, the geographic X coordinate of the remote sensing image to be inspected, when the remote sensing image to be inspected is an ortho-remote sensing image, the geographic Y coordinate of the remote sensing image to be inspected, column coordinate of the reference image, row coordinate of the reference image, geographic X coordinate of the reference image, geographic Y coordinate of the reference image, and when the remote sensing image to be inspected is an original remote sensing image, the elevation value determined based on the geographic X coordinate and Y coordinate of the reference image;

[0017] If the remote sensing image to be inspected is the original remote sensing image, use the RPC model or collinearity equation model to perform least squares fitting on the coordinates of all the initial homologous points to obtain the absolute error of each initial homologous point;

[0018] If the remote sensing image to be inspected is the ortho-remote sensing image, use the geometric polynomial model to perform least squares fitting on the coordinates of all the initial homologous points to obtain the geometric polynomial error of each initial homologous point;

[0019] If the absolute error or geometric polynomial error of the first initial homologous point among the initial homologous points is greater than the preset error threshold, determine that the first initial homologous point is a mis-matched homologous point, delete the first initial homologous point from the initial homologous points, and use the remaining initial homologous points as the homologous points.

[0020] Further, the conversion of the geographic coordinates of the reference image in the homologous points includes:

[0021] If the remote sensing image to be inspected is the original remote sensing image, use the RPC model to convert the geographic coordinates of the reference image in the homologous points to obtain the theoretical pixel coordinates of the geographic coordinates of the reference image on the remote sensing image to be inspected;

[0022] If the remote sensing image to be inspected is the ortho-remote sensing image, use the six-parameter conversion model of the remote sensing image to be inspected to convert the geographic coordinates of the reference image in the homologous points to obtain the theoretical pixel coordinates of the geographic coordinates of the reference image on the remote sensing image to be inspected;

[0023] Calculate the pixel error of each homologous point according to each of the theoretical pixel coordinates and the pixel coordinates of the remote sensing image to be inspected in the corresponding homologous points, including:

[0024] According to the calculation formula of pixel error of homologous points Calculate the pixel error of each of the homologous points, where RMS COL represents the column error in the pixel error of the homologous points, and RMS ROW represents the row error in the pixel error of the homologous points, line represents the theoretical column coordinate in the theoretical pixel coordinates, ssmpie represents the theoretical row coordinate in the theoretical pixel coordinates, col src represents the column coordinate in the pixel coordinates of the to-be-inspected remote sensing image in the homologous points corresponding to the theoretical pixel coordinates, row src represents the row coordinate in the pixel coordinates of the to-be-inspected remote sensing image in the homologous points corresponding to the theoretical pixel coordinates.

[0025] Furthermore, based on the pixel coordinates of the to-be-inspected remote sensing image in the homologous points, construct a triangular network in the to-be-inspected remote sensing image, and calculate the edge error of each side of each triangle in the triangular network according to the pixel error of each homologous point, including:

[0026] Use the open-source library CGAL to construct a triangular network for the pixel coordinates of the to-be-inspected remote sensing image in the homologous points to obtain the triangular network in the to-be-inspected remote sensing image;

[0027] According to the edge error calculation formula Calculate the edge error of each side of each triangle in the triangular network, where line rms represents the edge error, and RMS COL1 represents the column error in the pixel error of a homologous point that constitutes an edge, and RMS COL2 represents the column error in the pixel error of another homologous point that constitutes an edge, and RMS ROW1 represents the row error in the pixel error of a homologous point that constitutes an edge, and RMS ROW2 represents the row error in the pixel error of another homologous point that constitutes an edge.

[0028] Furthermore, based on each of the edge errors, use different colors to distinguish each side in the triangular network, including:

[0029] If the first edge error in the edge errors is less than the internal distortion error threshold, the edge corresponding to the first edge error is set to the first color;

[0030] If the second edge error in the edge errors is not less than the internal distortion error threshold and the second edge error is less than 2 times the internal distortion error threshold, the edge corresponding to the second edge error is set to the second color;

[0031] If the third side error in the side errors is not less than twice the internal distortion error threshold, the side corresponding to the third side error is set to the third color.

[0032] Further, calculate the medium error of the internal geometric distortion of the remote sensing image to be inspected according to each of the side errors, and calculate the gross error rate of the remote sensing image to be inspected based on each of the side errors, the preset internal distortion error threshold, and the total number of sides in the triangular network, including:

[0033] According to the calculation formula for the medium error of the internal geometric distortion Calculate the medium error of the internal geometric distortion of the remote sensing image to be inspected, where RMS represents the medium error of the internal geometric distortion of the remote sensing image to be inspected, and line rmsi represents the side error of the i-th side, and n represents the total number of sides in the triangular network;

[0034] Determine the number of side errors that are not less than twice the internal distortion error threshold among the side errors according to each of the side errors and the internal distortion error threshold;

[0035] According to the calculation formula for the gross error rate Calculate the gross error rate of the remote sensing image to be inspected, where Cp represents the gross error rate of the remote sensing image to be inspected, and line num2 represents the number of side errors that are not less than twice the internal distortion error threshold among the side errors, and line num represents the total number of sides in the triangular network.

[0036] Further, the method further includes:

[0037] Determine whether the remote sensing image to be inspected is qualified according to the quality inspection result of the internal geometric distortion of the remote sensing image to be inspected. If the medium error of the internal geometric distortion of the remote sensing image to be inspected is less than the internal distortion error threshold, and the gross error rate of the remote sensing image to be inspected is less than the preset internal distortion error threshold ratio, determine that the remote sensing image to be inspected is qualified; otherwise, the remote sensing image to be inspected is unqualified;

[0038] Generate an internal precision quality inspection report for the remote sensing image to be inspected based on the quality inspection results of the internal geometric distortion of the remote sensing image to be inspected. Among them, the internal precision quality inspection report includes: the mean square error of the internal geometric distortion of the remote sensing image to be inspected, the gross error rate of the remote sensing image to be inspected, the number of side errors in the side errors that are less than the internal distortion error threshold, the number of side errors in the side errors that are not less than the internal distortion error threshold and less than 2 times the internal distortion error threshold, the number of side errors in the side errors that are not less than 2 times the internal distortion error threshold, the number of triangles in the triangular network, the total number of sides in the triangular network, the maximum side error in the side errors, and whether the remote sensing image to be inspected is qualified.

[0039] In a second aspect, an embodiment of the present invention further provides a quantitative quality inspection device for internal geometric distortion of remote sensing images, including:

[0040] An acquisition unit, configured to acquire corresponding points of the remote sensing image to be inspected and a reference reference image, where the reference reference image is a standard ortho-remote sensing image, and the corresponding points are points where the remote sensing image to be inspected and the reference reference image have the same ground object position;

[0041] A conversion and calculation unit, configured to convert the geographic coordinates of the reference reference image in the corresponding points to obtain the theoretical pixel coordinates of the geographic coordinates of the reference reference image on the remote sensing image to be inspected, and calculate the pixel error of each corresponding point according to each theoretical pixel coordinate and the pixel coordinate of the remote sensing image to be inspected in the corresponding point;

[0042] A triangular network construction and side error calculation unit, configured to construct a triangular network in the remote sensing image to be inspected based on the pixel coordinates of the remote sensing image to be inspected in the corresponding points, and calculate the side error of each side of each triangle in the triangular network according to the pixel error of each corresponding point;

[0043] A color differentiation unit, configured to differentiate each side in the triangular network with different colors based on each side error to characterize the distribution of the internal geometric distortion of the remote sensing image to be inspected;

[0044] A calculation unit, configured to calculate the mean square error of the internal geometric distortion of the remote sensing image to be inspected according to each side error, and calculate the gross error rate of the remote sensing image to be inspected based on each side error, a preset internal distortion error threshold, and the total number of sides in the triangular network;

[0045] A setting unit, configured to use the mean square error of the internal geometric distortion of the remote sensing image to be inspected and the gross error rate of the remote sensing image to be inspected as the quality inspection results of the internal geometric distortion of the remote sensing image to be inspected.

[0046] In a third aspect, an embodiment of the present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the steps of the method according to any one of the above first aspects are implemented.

[0047] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium storing machine-executable instructions. When the machine-executable instructions are called and run by a processor, the machine-executable instructions cause the processor to run the method according to any one of the above first aspects.

[0048] In an embodiment of the present invention, a method for quantitatively inspecting internal geometric distortion of remote sensing images is provided, including: obtaining corresponding points of a to-be-inspected remote sensing image and a reference reference image, where the reference reference image is a standard ortho-remote sensing image, and the corresponding points are points with the same ground object position in the to-be-inspected remote sensing image and the reference reference image; converting the geographic coordinates of the reference reference image among the corresponding points to obtain the theoretical pixel coordinates of the geographic coordinates of the reference reference image on the to-be-inspected remote sensing image, and calculating the pixel error of each corresponding point according to each theoretical pixel coordinate and the pixel coordinate of the to-be-inspected remote sensing image among the corresponding points; constructing a triangular network in the to-be-inspected remote sensing image based on the pixel coordinates of the to-be-inspected remote sensing image among the corresponding points, and calculating the edge error of each side of each triangle in the triangular network according to the pixel error of each corresponding point; differentiating each side in the triangular network with different colors based on each edge error to characterize the distribution of the internal geometric distortion of the to-be-inspected remote sensing image; calculating the root mean square error of the internal geometric distortion of the to-be-inspected remote sensing image according to each edge error, and calculating the gross error rate of the to-be-inspected remote sensing image based on each edge error, a preset internal distortion error threshold, and the total number of sides in the triangular network; using the root mean square error of the internal geometric distortion of the to-be-inspected remote sensing image and the gross error rate of the to-be-inspected remote sensing image as the quality inspection result of the internal geometric distortion of the to-be-inspected remote sensing image. Through the above description, it can be seen that in the method for quantitatively inspecting internal geometric distortion of remote sensing images of the present invention, each side of the triangular network in the to-be-inspected remote sensing image is differentiated with different colors to characterize the distribution of the internal geometric distortion of the to-be-inspected remote sensing image. Therefore, the user can visually (i.e., apparently) and accurately capture the internal geometric distortion situation of the to-be-inspected remote sensing image, and also use the root mean square error of the internal geometric distortion of the to-be-inspected remote sensing image and the gross error rate of the to-be-inspected remote sensing image as the quality inspection result of the internal geometric distortion of the to-be-inspected remote sensing image, that is, it can quantitatively describe the internal geometric distortion value of the to-be-inspected remote sensing image, alleviating the technical problems that the prior art cannot apparently characterize the internal geometric distortion of the to-be-inspected remote sensing image and cannot quantitatively calculate the internal geometric distortion of the to-be-inspected remote sensing image. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0050] Figure 1 It is a flowchart of a method for quantitatively inspecting internal geometric distortion of remote sensing images provided by an embodiment of the present invention;

[0051] Figure 2 It is a diagram showing the color separation representation of the triangulation of the internal geometric distortion of a remote sensing image to be inspected provided by an embodiment of the present invention;

[0052] Figure 3 It is another diagram showing the color separation representation of the triangulation of the internal geometric distortion of a remote sensing image to be inspected provided by an embodiment of the present invention;

[0053] Figure 4 It is a schematic diagram of a device for quantitatively inspecting internal geometric distortion of remote sensing images provided by an embodiment of the present invention;

[0054] Figure 5 It is a schematic diagram of an electronic device provided by an embodiment of the present invention. Specific Embodiments

[0055] The following will clearly and completely describe the technical solutions of the present invention in conjunction with the embodiments. Obviously, the described embodiments are some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0056] The traditional technology cannot visually represent the internal geometric distortion of the remote sensing image to be inspected, nor can it quantitatively calculate the internal geometric distortion of the remote sensing image to be inspected.

[0057] Based on this, in the method for quantitatively inspecting internal geometric distortion of remote sensing images of the present invention, each side of the triangulation in the remote sensing image to be inspected is distinguished by different colors to represent the distribution of the internal geometric distortion of the remote sensing image to be inspected. Therefore, users can visually (i.e., apparently) and accurately capture the internal geometric distortion situation of the remote sensing image to be inspected, and also use the mean square error of the internal geometric distortion of the remote sensing image to be inspected and the gross error rate of the remote sensing image to be inspected as the quality inspection results of the internal geometric distortion of the remote sensing image to be inspected, that is, it can quantitatively describe the internal geometric distortion value of the remote sensing image to be inspected.

[0058] To facilitate the understanding of this embodiment, a quantitative quality inspection method for internal geometric distortion of remote sensing images disclosed in the embodiments of the present invention will be introduced in detail first.

[0059] Embodiment 1:

[0060] According to an embodiment of the present invention, an embodiment of a quantitative quality inspection method for internal geometric distortion of remote sensing images is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0061] Figure 1 is a flowchart of a quantitative quality inspection method for internal geometric distortion of remote sensing images according to an embodiment of the present invention. As Figure 1 shown, the method includes the following steps:

[0062] Step S102, obtain the homologous points of the remote sensing image to be inspected and the reference reference image. Among them, the reference reference image is a standard ortho-remote sensing image, and the homologous points are the points with the same ground object position in the remote sensing image to be inspected and the reference reference image;

[0063] In the embodiment of the present invention, the accuracy of the remote sensing image is divided into absolute accuracy (i.e., the accuracy between the remote sensing image and the reference reference image), internal accuracy (i.e., the accuracy between different feature points in the remote sensing image), and relative accuracy. The present invention aims to perform automatic quantitative quality inspection on the internal accuracy of the remote sensing image.

[0064] Specifically, the homologous points refer to the points with the same ground object position in the remote sensing image to be inspected and the reference reference image, that is, the feature points in the remote sensing image to be inspected and the matching points corresponding to the feature points in the reference reference image. The above reference reference image refers to an ortho-remote sensing image that is considered to be the closest to accurate in both absolute accuracy and internal accuracy, that is, a standard ortho-remote sensing image.

[0065] Step S104, convert the geographic coordinates of the reference reference image among the homologous points to obtain the theoretical pixel coordinates of the geographic coordinates of the reference reference image on the remote sensing image to be inspected, and calculate the pixel error of each homologous point according to each theoretical pixel coordinate and the pixel coordinate of the remote sensing image to be inspected among the corresponding homologous points;

[0066] Specifically, the geographic coordinates of the reference reference image include: the geographic X coordinate of the reference reference image, the geographic Y coordinate of the reference reference image, and the elevation value Z read from the DEM data (i.e., digital elevation model data) according to the geographic X coordinate and Y coordinate of the reference reference image. Convert the above geographic coordinates to obtain the theoretical pixel coordinates of the geographic coordinates of the reference reference image on the remote sensing image to be inspected.

[0067] Step S106: Construct a triangular network in the remotely sensed image to be inspected based on the pixel coordinates of the corresponding points in the remotely sensed image to be inspected, and calculate the edge error of each side of each triangle in the triangular network according to the pixel error of each corresponding point.

[0068] Specifically, the pixel coordinates of the remotely sensed image to be inspected include the column coordinate and the row coordinate of the remotely sensed image to be inspected.

[0069] Step S108: Differentiate each side in the triangular network with different colors based on the edge errors to characterize the distribution of the internal geometric distortion of the remotely sensed image to be inspected.

[0070] Step S110: Calculate the mean square error of the internal geometric distortion of the remotely sensed image to be inspected according to the edge errors, and calculate the gross error rate of the remotely sensed image to be inspected based on the edge errors, the preset internal distortion error threshold, and the total number of sides in the triangular network.

[0071] Step S112: Take the mean square error of the internal geometric distortion of the remotely sensed image to be inspected and the gross error rate of the remotely sensed image to be inspected as the quality inspection results of the internal geometric distortion of the remotely sensed image to be inspected.

[0072] In an embodiment of the present invention, a method for quantitatively inspecting internal geometric distortion of remote sensing images is provided, including: obtaining corresponding points of the remote sensing image to be inspected and the reference reference image, where the reference reference image is a standard ortho-remote sensing image, and the corresponding points are points with the same ground object position in the remote sensing image to be inspected and the reference reference image; converting the geographic coordinates of the reference reference image in the corresponding points to obtain the theoretical pixel coordinates of the geographic coordinates of the reference reference image on the remote sensing image to be inspected, and calculating the pixel error of each corresponding point according to each theoretical pixel coordinate and the pixel coordinate of the remote sensing image to be inspected in the corresponding point; constructing a triangular network in the remote sensing image to be inspected based on the pixel coordinates of the remote sensing image to be inspected in the corresponding points, and calculating the edge error of each side of each triangle in the triangular network according to the pixel error of each corresponding point; distinguishing each side in the triangular network with different colors based on each edge error to characterize the distribution of the internal geometric distortion of the remote sensing image to be inspected; calculating the root mean square error of the internal geometric distortion of the remote sensing image to be inspected according to each edge error, and calculating the gross error rate of the remote sensing image to be inspected based on each edge error, a preset internal distortion error threshold, and the total number of sides in the triangular network; taking the root mean square error of the internal geometric distortion of the remote sensing image to be inspected and the gross error rate of the remote sensing image to be inspected as the inspection result of the internal geometric distortion of the remote sensing image to be inspected. Through the above description, it can be seen that in the method for quantitatively inspecting internal geometric distortion of remote sensing images of the present invention, each side of the triangular network in the remote sensing image to be inspected is distinguished by different colors to characterize the distribution of the internal geometric distortion of the remote sensing image to be inspected. Therefore, users can visually (i.e., apparently) and accurately capture the internal geometric distortion of the remote sensing image to be inspected, and also use the root mean square error of the internal geometric distortion of the remote sensing image to be inspected and the gross error rate of the remote sensing image to be inspected as the inspection result of the internal geometric distortion of the remote sensing image to be inspected, that is, it can quantitatively describe the internal geometric distortion value of the remote sensing image to be inspected, alleviating the technical problems that the prior art cannot apparently characterize the internal geometric distortion of the remote sensing image to be inspected and cannot quantitatively calculate the internal geometric distortion of the remote sensing image to be inspected.

[0073] The above content briefly introduces the method for quantitatively inspecting internal geometric distortion of remote sensing images of the present invention, and the following will describe the specific content involved in detail.

[0074] In an alternative embodiment of the present invention, the remote sensing image to be inspected includes any one of the following: a raw remote sensing image with an initial positioning model file, an ortho-remote sensing image. Obtaining the corresponding points of the remote sensing image to be inspected and the reference reference image specifically includes the following steps:

[0075] (1) Divide the remote sensing image to be inspected into grids to obtain a plurality of grids;

[0076] Specifically, the remote sensing image to be inspected can be an original remote sensing image with an initial positioning model file (such as an initial positioning model file with a strict imaging model, RPC (rational polynomial model of satellite images), POS (position file model of aerial images), etc.), or an ortho-rectified remote sensing image (i.e., a remote sensing image that has been ortho-rectified).

[0077] The obtained corresponding points need to satisfy the characteristics of uniform distribution and correct positions. To satisfy the uniform distribution of corresponding points, it is necessary to divide the remote sensing image to be inspected into grids, and then extract a most characteristic feature point in each grid.

[0078] (2) Use the Harris feature point operator to extract feature points for each grid to obtain the feature points of each grid;

[0079] (3) Determine the matching points in the reference image that match each feature point, and use each feature point and its matching point as initial corresponding points. Among them, the initial corresponding points include: corresponding point number, column coordinate of the remote sensing image to be inspected, row coordinate of the remote sensing image to be inspected, geographic X coordinate of the remote sensing image to be inspected when the remote sensing image to be inspected is an ortho-rectified remote sensing image, geographic Y coordinate of the remote sensing image to be inspected when the remote sensing image to be inspected is an ortho-rectified remote sensing image, column coordinate of the reference image, row coordinate of the reference image, geographic X coordinate of the reference image, geographic Y coordinate of the reference image, and elevation value determined based on the geographic X coordinate and Y coordinate of the reference image when the remote sensing image to be inspected is an original remote sensing image;

[0080] Specifically, phase consistency matching or correlation coefficient matching can be used to determine the matching points in the reference image that match each feature point, and then use each feature point and its matching point as initial corresponding points.

[0081] The format of the initial corresponding points is as follows:

[0082] ID <![CDATA[col src > <![CDATA[row src > <![CDATA[X src > <![CDATA[Y src > <![CDATA[col ref > <![CDATA[row ref > <![CDATA[X r e f > <![CDATA[Y ref > Z

[0083] ID: Corresponding point number;

[0084] col src : Column coordinate in the pixel coordinates of the remote sensing image to be inspected;

[0085] row src : Row coordinate in the pixel coordinates of the remote sensing image to be inspected;

[0086] X src : Geographic X coordinate in the geographic coordinates of the remote sensing image to be inspected when the remote sensing image to be inspected is an ortho-rectified remote sensing image, not required when the remote sensing image to be inspected is an original remote sensing image;

[0087] Ysrc : When the remotely sensed image to be inspected is an ortho-remotely sensed image, it is the geographic Y coordinate of the remotely sensed image to be inspected, and it is not required when the remotely sensed image to be inspected is an original remotely sensed image;

[0088] col ref : The column coordinate in the pixel coordinates of the reference reference image;

[0089] row ref : The row coordinate in the pixel coordinates of the reference reference image;

[0090] X ref : The geographic X coordinate in the geographic coordinates of the reference reference image;

[0091] Y ref : The geographic Y coordinate in the geographic coordinates of the reference reference image;

[0092] Z: When the remotely sensed image to be inspected is an original remotely sensed image, it is the elevation value determined based on the geographic X coordinate and Y coordinate of the reference reference image, and it may not be required when the remotely sensed image to be inspected is an ortho-remotely sensed image;

[0093] (4) If the remotely sensed image to be inspected is an original remotely sensed image, the RPC model or collinearity equation model is used to perform least squares fitting on all initial corresponding point coordinates to obtain the absolute error of each initial corresponding point;

[0094] Specifically, to satisfy the correctness of the corresponding points, usually after obtaining the initial corresponding points, it is necessary to automatically delete the initial corresponding points with incorrect matches to obtain the correct corresponding points.

[0095] In implementation, if the remotely sensed image to be inspected is an original remotely sensed image, the RPC model (i.e., the rational polynomial model of satellite images) or collinearity equation model is used to perform least squares fitting on all initial corresponding point coordinates to obtain the absolute error of each initial corresponding point.

[0096] (5) If the remotely sensed image to be inspected is an ortho-remotely sensed image, the geometric polynomial model is used to perform least squares fitting on all initial corresponding point coordinates to obtain the geometric polynomial error of each initial corresponding point;

[0097] (6) If the absolute error or geometric polynomial error of the first initial corresponding point among the initial corresponding points is greater than the preset error threshold, it is determined that the first initial corresponding point is an incorrectly matched corresponding point, and the first initial corresponding point is deleted from the initial corresponding points, and the remaining initial corresponding points are used as the corresponding points.

[0098] In an optional embodiment of the present invention, the conversion of the geographic coordinates of the reference reference image in the corresponding points specifically includes the following steps:

[0099] (1) If the remote sensing image to be inspected is the original remote sensing image, the RPC model is used to convert the geographic coordinates of the reference image in the corresponding points to obtain the theoretical pixel coordinates of the geographic coordinates of the reference image on the remote sensing image to be inspected.

[0100] Specifically, the geographic coordinates of the reference image are (X ref , Y ref , Z), and the RPC model is used to convert (X ref , Y ref , Z) to obtain the theoretical pixel coordinates of the geographic coordinates of the reference image on the remote sensing image to be inspected.

[0101] (2) If the remote sensing image to be inspected is an ortho-rectified remote sensing image, the six-parameter transformation model of the remote sensing image to be inspected (the six-parameter transformation model is provided in the open-source library GDAL) is used to convert the geographic coordinates of the reference image in the corresponding points to obtain the theoretical pixel coordinates of the geographic coordinates of the reference image on the remote sensing image to be inspected.

[0102] In an optional embodiment of the present invention, the pixel error of each corresponding point is calculated according to each theoretical pixel coordinate and the pixel coordinate of the remote sensing image to be inspected in the corresponding point, which specifically includes the following steps:

[0103] According to the calculation formula of the corresponding point pixel error calculate the pixel error of each corresponding point, where RMS COL represents the column error in the pixel error of the corresponding point, RMS ROW represents the row error in the pixel error of the corresponding point, line represents the theoretical column coordinate in the theoretical pixel coordinate, sample represents the theoretical row coordinate in the theoretical pixel coordinate, col src represents the column coordinate in the pixel coordinate of the remote sensing image to be inspected in the corresponding point corresponding to the theoretical pixel coordinate, row sr c represents the row coordinate in the pixel coordinate of the remote sensing image to be inspected in the corresponding point corresponding to the theoretical pixel coordinate.

[0104] In an optional embodiment of the present invention, a triangular network is constructed in the remote sensing image to be inspected based on the pixel coordinates of the remote sensing image to be inspected in the corresponding points, and the edge error of each side of each triangle in the triangular network is calculated according to the pixel error of each corresponding point, which specifically includes the following steps:

[0105] (1) Use the open-source library CGAL to construct a triangular network for the pixel coordinates of the remote sensing image to be inspected in the corresponding points to obtain the triangular network in the remote sensing image to be inspected.

[0106] (2) According to the edge error calculation formula Calculate the edge error of each edge of each triangle in the triangular network, where line rms represents the edge error of the AB edge, RMS COL1 represents the column error in the pixel error of the A corresponding point that constitutes the AB edge, RMS COL2 represents the column error in the pixel error of the B corresponding point that constitutes the AB edge, RMS ROW1 represents the row error in the pixel error of the A corresponding point that constitutes the AB edge, RMS ROW2 represents the row error in the pixel error of the B corresponding point that constitutes the AB edge.

[0107] The edge error of each of the above edges can characterize the relative error value between two corresponding points (specifically, the two feature points that form the edge in the remote sensing image to be inspected), that is, the internal error value. The edge errors of multiple edges can objectively and quantitatively reflect the internal distortion situation of the entire remote sensing image to be inspected.

[0108] In an alternative embodiment of the present invention, different colors are used to distinguish each edge in the triangular network based on the edge errors, which specifically includes the following steps:

[0109] (1) If the first edge error in the edge errors is less than the internal distortion error threshold, the edge corresponding to the first edge error is set to the first color;

[0110] The above first color can be green.

[0111] (2) If the second edge error in the edge errors is not less than the internal distortion error threshold and the second edge error is less than 2 times the internal distortion error threshold, the edge corresponding to the second edge error is set to the second color;

[0112] The above second color can be yellow.

[0113] (3) If the third edge error in the edge errors is not less than 2 times the internal distortion error threshold, the edge corresponding to the third edge error is set to the third color.

[0114] The above second color can be red, specifically as Figure 2 shown, from which it can be clearly judged which areas of the remote sensing image to be inspected have internal geometric distortion and the size of the internal geometric distortion.

[0115] In an alternative embodiment of the present invention, calculate the mean square error of the internal geometric distortion of the remote sensing image to be inspected based on the edge errors, and calculate the gross error rate of the remote sensing image to be inspected based on the edge errors, the preset internal distortion error threshold, and the total number of edges in the triangular network, which specifically includes the following steps:

[0116] (1) According to the calculation formula of the mean square error of the internal geometric distortion Calculate the root mean square (RMS) of the internal geometric distortion of the remotely sensed image to be inspected. Here, RMS represents the RMS of the internal geometric distortion of the remotely sensed image to be inspected, and line rmsi represents the edge error of the i-th edge, and n represents the total number of edges in the triangular network;

[0117] (2) Determine the number of edge errors that are not less than 2 times the internal distortion error threshold based on the edge errors and the internal distortion error threshold;

[0118] The number of edge errors that are not less than 2 times the internal distortion error threshold among the above edge errors is the number of edges that are not less than 2 times the internal distortion error threshold.

[0119] (3) Calculate the gross error rate of the remotely sensed image to be inspected according to the gross error rate calculation formula where Cp represents the gross error rate of the remotely sensed image to be inspected, and line num2 represents the number of edge errors that are not less than 2 times the internal distortion error threshold, and line num represents the total number of edges in the triangular network.

[0120] In an alternative embodiment of the present invention, the method further includes:

[0121] (1) Determine whether the remotely sensed image to be inspected is qualified according to the quality inspection result of the internal geometric distortion of the remotely sensed image to be inspected. Here, if the RMS of the internal geometric distortion of the remotely sensed image to be inspected is less than the internal distortion error threshold, and the gross error rate of the remotely sensed image to be inspected is less than the preset internal distortion error threshold ratio, it is determined that the remotely sensed image to be inspected is qualified; otherwise, the remotely sensed image to be inspected is unqualified;

[0122] (2) Generate an internal precision quality inspection report for the remotely sensed image to be inspected based on the quality inspection result of the internal geometric distortion of the remotely sensed image to be inspected. The internal precision quality inspection report includes: the RMS of the internal geometric distortion of the remotely sensed image to be inspected, the gross error rate of the remotely sensed image to be inspected, the number of edge errors less than the internal distortion error threshold (i.e., the number of edges less than the internal distortion error threshold), the number of edge errors not less than the internal distortion error threshold and less than 2 times the internal distortion error threshold, the number of edge errors not less than 2 times the internal distortion error threshold, the number of triangles in the triangular network, the total number of edges in the triangular network, the maximum edge error among the edge errors, and whether the remotely sensed image to be inspected is qualified.

[0123] Figure 3A schematic diagram of a remotely sensed image to be inspected with a triangulation network with color differentiation in an internal precision quality inspection report is shown. In addition, the internal precision quality inspection report also includes: the path of the remotely sensed image to be inspected (not specifically shown here); the path of the reference image (not specifically shown here); the total number of corresponding points: 175; the number of triangles: 330; the total number of edges in the triangulation network: 504; the internal distortion error threshold: 3.000 pixels; the internal distortion error threshold ratio: 0.050; the number of edge errors less than the internal distortion error threshold: 154; the number of edge errors not less than the internal distortion error threshold and less than 2 times the internal distortion error threshold: 254; the number of edge errors not less than 2 times the internal distortion error threshold: 96; the internal geometric distortion root mean square error of the remotely sensed image to be inspected: 4.696 pixels; the gross error rate (Cp) (explanation: the number of edge errors not less than 2 times the internal distortion error threshold divided by the total number of edges in the triangulation network * 100% = 96 / 504 * 100%): 19.048%; the maximum edge error: 10.453 pixels; the final quality inspection result (explanation: the qualified condition is that the gross error rate of the remotely sensed image to be inspected < the preset internal distortion error threshold ratio, and the internal geometric distortion root mean square error of the remotely sensed image to be inspected < the internal distortion error threshold): unqualified. The internal precision quality inspection report may also include the pixel errors of each corresponding point that makes up each edge, that is, the corresponding relationship between the row error in the pixel error of the A corresponding point of the AB edge - the column error in the pixel error of the A corresponding point - the row error in the pixel error of the B corresponding point - the column error in the pixel error of the B corresponding point.

[0124] The method of the present invention can clearly and quickly locate visually which parts of the remotely sensed image to be inspected have large internal geometric distortions, and can accurately give the internal geometric distortion root mean square error and the gross error rate of this remotely sensed image to be inspected quantitatively, and give a final conclusion on whether it is qualified.

[0125] Embodiment 2:

[0126] The embodiment of the present invention also provides a device for quantitatively inspecting the internal geometric distortion of a remotely sensed image. The device for quantitatively inspecting the internal geometric distortion of a remotely sensed image is mainly used to execute the method for quantitatively inspecting the internal geometric distortion of a remotely sensed image provided in Embodiment 1 of the present invention. The following is a specific introduction to the device for quantitatively inspecting the internal geometric distortion of a remotely sensed image provided in the embodiment of the present invention.

[0127] Figure 4 is a schematic diagram of a device for quantitatively inspecting the internal geometric distortion of a remotely sensed image according to an embodiment of the present invention, as Figure 4As shown in the figure, the device mainly includes: an acquisition unit 10, a conversion and calculation unit 20, a triangular network construction and edge error calculation unit 30, a color discrimination unit 40, a calculation unit 50, and a setting unit 60, where:

[0128] The acquisition unit is used to acquire the corresponding points of the remote sensing image to be inspected and the reference reference image. Among them, the reference reference image is a standard ortho-remote sensing image, and the corresponding points are the points where the remote sensing image to be inspected and the reference reference image have the same ground object position;

[0129] The conversion and calculation unit is used to convert the geographical coordinates of the reference reference image in the corresponding points to obtain the theoretical pixel coordinates of the geographical coordinates of the reference reference image on the remote sensing image to be inspected, and calculate the pixel error of each corresponding point according to each theoretical pixel coordinate and the pixel coordinate of the remote sensing image to be inspected in the corresponding point;

[0130] The triangular network construction and edge error calculation unit is used to construct a triangular network in the remote sensing image to be inspected based on the pixel coordinates of the remote sensing image to be inspected in the corresponding points, and calculate the edge error of each side of each triangle in the triangular network according to the pixel error of each corresponding point;

[0131] The color discrimination unit is used to distinguish each side in the triangular network with different colors based on each edge error to characterize the distribution of the internal geometric distortion of the remote sensing image to be inspected;

[0132] The calculation unit is used to calculate the mean square error of the internal geometric distortion of the remote sensing image to be inspected according to each edge error, and calculate the gross error rate of the remote sensing image to be inspected based on each edge error, a preset internal distortion error threshold, and the total number of sides in the triangular network;

[0133] The setting unit is used to use the mean square error of the internal geometric distortion of the remote sensing image to be inspected and the gross error rate of the remote sensing image to be inspected as the quality inspection results of the internal geometric distortion of the remote sensing image to be inspected.

[0134] In an embodiment of the present invention, a quantitative quality inspection device for internal geometric distortion of remote sensing images is provided, including: obtaining corresponding points of a to-be-inspected remote sensing image and a reference reference image, where the reference reference image is a standard ortho-remote sensing image, and the corresponding points are points with the same ground object position in the to-be-inspected remote sensing image and the reference reference image; converting the geographical coordinates of the reference reference image in the corresponding points to obtain the theoretical pixel coordinates of the geographical coordinates of the reference reference image on the to-be-inspected remote sensing image, and calculating the pixel errors of each corresponding point according to the respective theoretical pixel coordinates and the pixel coordinates of the to-be-inspected remote sensing image in the corresponding points; constructing a triangular network in the to-be-inspected remote sensing image based on the pixel coordinates of the to-be-inspected remote sensing image in the corresponding points, and calculating the edge errors of each side of each triangle in the triangular network according to the pixel errors of each corresponding point; differentiating each side in the triangular network with different colors based on the respective edge errors to characterize the distribution of the internal geometric distortion of the to-be-inspected remote sensing image; calculating the mean square error of the internal geometric distortion of the to-be-inspected remote sensing image according to the respective edge errors, and calculating the gross error rate of the to-be-inspected remote sensing image based on the respective edge errors, a preset internal distortion error threshold, and the total number of sides in the triangular network; using the mean square error of the internal geometric distortion of the to-be-inspected remote sensing image and the gross error rate of the to-be-inspected remote sensing image as the quality inspection result of the internal geometric distortion of the to-be-inspected remote sensing image. Through the above description, it can be seen that in the quantitative quality inspection device for internal geometric distortion of remote sensing images of the present invention, each side of the triangular network in the to-be-inspected remote sensing image is differentiated with different colors to characterize the distribution of the internal geometric distortion of the to-be-inspected remote sensing image. Therefore, the user can visually (i.e., apparently) and accurately capture the internal geometric distortion situation of the to-be-inspected remote sensing image, and also uses the mean square error of the internal geometric distortion of the to-be-inspected remote sensing image and the gross error rate of the to-be-inspected remote sensing image as the quality inspection result of the internal geometric distortion of the to-be-inspected remote sensing image, that is, it can quantitatively describe the internal geometric distortion value of the to-be-inspected remote sensing image, alleviating the technical problems in the prior art that the internal geometric distortion of the to-be-inspected remote sensing image cannot be characterized apparently and the internal geometric distortion of the to-be-inspected remote sensing image cannot be quantitatively calculated.

[0135] Optionally, the remote sensing image to be inspected includes any one of the following: the original remote sensing image with an initial positioning model file, the ortho-rectified remote sensing image. The obtaining unit is further configured to: divide the remote sensing image to be inspected into grids to obtain a plurality of grids; extract feature points from each grid by using the Harris feature point operator to obtain the feature points of each grid; determine matching points in the reference image that match the respective feature points, and use each feature point and its matching point as initial homologous points. The initial homologous points include: homologous point number, column coordinate of the remote sensing image to be inspected, row coordinate of the remote sensing image to be inspected, geographic X coordinate of the remote sensing image to be inspected when the remote sensing image to be inspected is an ortho-rectified remote sensing image, geographic Y coordinate of the remote sensing image to be inspected when the remote sensing image to be inspected is an ortho-rectified remote sensing image, column coordinate of the reference image, row coordinate of the reference image, geographic X coordinate of the reference image, geographic Y coordinate of the reference image, and elevation value determined based on the geographic X coordinate and Y coordinate of the reference image when the remote sensing image to be inspected is an original remote sensing image; if the remote sensing image to be inspected is an original remote sensing image, perform least squares fitting on the coordinates of all the initial homologous points by using the RPC model or collinearity equation model to obtain the absolute error of each initial homologous point; if the remote sensing image to be inspected is an ortho-rectified remote sensing image, perform least squares fitting on the coordinates of all the initial homologous points by using the geometric polynomial model to obtain the geometric polynomial error of each initial homologous point; if the absolute error or geometric polynomial error of the first initial homologous point among the initial homologous points is greater than a preset error threshold, determine that the first initial homologous point is a mis-matched homologous point, delete the first initial homologous point from the initial homologous points, and use the remaining initial homologous points as homologous points.

[0136] Optionally, the conversion and calculation unit is further configured to: if the remote sensing image to be inspected is an original remote sensing image, convert the geographic coordinates of the reference image among the homologous points by using the RPC model to obtain the theoretical pixel coordinates of the geographic coordinates of the reference image on the remote sensing image to be inspected; if the remote sensing image to be inspected is an ortho-rectified remote sensing image, convert the geographic coordinates of the reference image among the homologous points by using the six-parameter conversion model of the remote sensing image to be inspected to obtain the theoretical pixel coordinates of the geographic coordinates of the reference image on the remote sensing image to be inspected; according to the pixel error calculation formula of the homologous points calculate the pixel error of each homologous point, where RMS COL represents the column error in the pixel error of the homologous points, RMS ROW represents the row error in the pixel error of the homologous points, line represents the theoretical column coordinate in the theoretical pixel coordinates, sample represents the theoretical row coordinate in the theoretical pixel coordinates, col src represents the column coordinate in the pixel coordinates of the remote sensing image to be inspected among the homologous points corresponding to the theoretical pixel coordinates, row srcRepresents the row coordinate in the pixel coordinates of the remote sensing image to be inspected among the homologous points corresponding to the theoretical pixel coordinates.

[0137] Optionally, the triangulation network construction and edge error calculation unit is further configured to: use the open-source library CGAL to construct a triangulation network for the pixel coordinates of the remote sensing image to be inspected among the homologous points, obtaining the triangulation network in the remote sensing image to be inspected; according to the edge error calculation formula Calculate the edge error of each edge of each triangle in the triangulation network, where line rms Represents the edge error, and RMS COL1 Represents the column error in the pixel error of a homologous point constituting the edge, and RMS COL2 Represents the column error in the pixel error of another homologous point constituting the edge, and RMS ROW1 Represents the row error in the pixel error of a homologous point constituting the edge, and RMS ROW2 Represents the row error in the pixel error of another homologous point constituting the edge.

[0138] Optionally, the color differentiation unit is further configured to: if the first edge error among the edge errors is less than the internal distortion error threshold, set the edge corresponding to the first edge error to the first color; if the second edge error among the edge errors is not less than the internal distortion error threshold and is less than 2 times the internal distortion error threshold, set the edge corresponding to the second edge error to the second color; if the third edge error among the edge errors is not less than 2 times the internal distortion error threshold, set the edge corresponding to the third edge error to the third color.

[0139] Optionally, the calculation unit is further configured to: according to the internal geometric distortion mean error calculation formula Calculate the internal geometric distortion mean error of the remote sensing image to be inspected, where RMS represents the internal geometric distortion mean error of the remote sensing image to be inspected, and line rmsi Represents the edge error of the i-th edge, and n represents the total number of edges in the triangulation network; determine the number of edge errors not less than 2 times the internal distortion error threshold among the edge errors according to the edge errors and the internal distortion error threshold; according to the gross error rate calculation formula Calculate the gross error rate of the remote sensing image to be inspected, where Cp represents the gross error rate of the remote sensing image to be inspected, and line num2 Represents the number of edge errors not less than 2 times the internal distortion error threshold among the edge errors, and line num Represents the total number of edges in the triangulation network.

[0140] Optionally, the device is further configured to: determine whether the remote sensing image to be inspected is qualified according to the quality inspection result of the internal geometric distortion of the remote sensing image to be inspected, where if the root mean square error of the internal geometric distortion of the remote sensing image to be inspected is less than the internal distortion error threshold, and the gross error rate of the remote sensing image to be inspected is less than the preset internal distortion error threshold ratio, it is determined that the remote sensing image to be inspected is qualified; otherwise, the remote sensing image to be inspected is unqualified;

[0141] Generate an internal precision quality inspection report for the remote sensing image to be inspected based on the quality inspection result of the internal geometric distortion of the remote sensing image to be inspected, where the internal precision quality inspection report includes: the root mean square error of the internal geometric distortion of the remote sensing image to be inspected, the gross error rate of the remote sensing image to be inspected, the number of side errors less than the internal distortion error threshold in the side errors, the number of side errors not less than the internal distortion error threshold and less than 2 times the internal distortion error threshold in the side errors, the number of side errors not less than 2 times the internal distortion error threshold in the side errors, the number of triangles in the triangular network, the total number of sides in the triangular network, the maximum side error in the side errors, and whether the remote sensing image to be inspected is qualified.

[0142] The device provided by the embodiments of the present invention has the same implementation principle and the same technical effects as those of the foregoing method embodiments. For a brief description, for the parts not mentioned in the device embodiments, reference may be made to the corresponding content in the foregoing method embodiments.

[0143] As Figure 5 shown, an electronic device 600 provided by an embodiment of the present application includes: a processor 601, a memory 602, and a bus. The memory 602 stores machine-readable instructions executable by the processor 601. When the electronic device runs, the processor 601 communicates with the memory 602 through the bus, and the processor 601 executes the machine-readable instructions to perform the steps of the above-mentioned method for quantitatively inspecting the internal geometric distortion of a remote sensing image.

[0144] Specifically, the above-mentioned memory 602 and processor 601 can be general-purpose memory and processor, which are not specifically limited here. When the processor 601 runs the computer program stored in the memory 602, it can execute the above-mentioned method for quantitatively inspecting the internal geometric distortion of a remote sensing image.

[0145] The processor 601 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit in the hardware of the processor 601 or instructions in the form of software. The above-mentioned processor 601 can be a general-purpose processor, including a central processing unit (CPU for short), a network processor (NP for short), etc.; it can also be a digital signal processor (DSP for short), an application specific integrated circuit (ASIC for short), a field-programmable gate array (FPGA for short), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as being executed and completed by the hardware decoding processor, or can be executed and completed by a combination of the hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. This storage medium is located in the memory 602, and the processor 601 reads the information in the memory 602 and combines its hardware to complete the steps of the above method.

[0146] Corresponding to the above method for quantitatively inspecting the internal geometric distortion of remote sensing images, an embodiment of the present application also provides a computer-readable storage medium. The computer-readable storage medium stores machine-executable instructions. When the computer-executable instructions are called and run by a processor, the computer-executable instructions cause the processor to run the steps of the above method for quantitatively inspecting the internal geometric distortion of remote sensing images.

[0147] The device for quantitatively inspecting the internal geometric distortion of remote sensing images provided by the embodiments of the present application can be specific hardware on the device or software or firmware installed on the device, etc. For the device provided by the embodiments of the present application, the implementation principle and the technical effects produced are the same as those of the foregoing method embodiments. For the sake of brief description, for the parts not mentioned in the device embodiments, reference can be made to the corresponding content in the foregoing method embodiments. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices, and units described above can all refer to the corresponding processes in the above method embodiments and will not be elaborated here.

[0148] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For another example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some communication interfaces. The indirect coupling or communication connection of the devices or units can be in electrical, mechanical or other forms.

[0149] For another example, the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions and operations of devices, methods and computer program products according to multiple embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment or a part of code, and the part of the module, the program segment or the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0150] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0151] In addition, the functional units in the embodiments provided in the present application can be integrated in one processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0152] When the above-mentioned functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing an electronic device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method for quantitatively inspecting internal geometric distortion of remote sensing images described in various embodiments of this application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM for short), random access memories (RAM for short), magnetic disks, or optical discs that can store program codes.

[0153] It should be noted that: similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. In addition, the terms "first", "second", "third", etc. are only used for descriptive distinction and cannot be understood as indicating or implying relative importance.

[0154] Finally, it should be noted that: the above-mentioned embodiments are only specific implementation manners of this application, used to illustrate the technical solution of this application, rather than limiting it. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: any person skilled in the art within the technical scope disclosed in this application can still modify the technical solutions recorded in the foregoing embodiments, or can easily think of changes, or perform equivalent replacements on some of the technical features; and these modifications, changes, or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of this application. All should be covered within the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

1. A quantitative quality inspection method for internal geometric distortion of remote sensing images, characterized in that: include: Obtaining the same-name points of the remote sensing image to be checked and the benchmark reference image, wherein the benchmark reference image is a standard orthophoto remote sensing image, and the same-name points are points where the remote sensing image to be checked and the benchmark reference image have the same ground object position; The geographic coordinates of the benchmark reference image in the same-name point are converted to obtain theoretical pixel coordinates of the geographic coordinates of the benchmark reference image on the remote sensing image to be checked, and the pixel error of each of the same-name points is calculated according to each of the theoretical pixel coordinates and the pixel coordinates of the remote sensing image to be checked in the corresponding same-name point; Constructing a triangulated network in the remote sensing image to be checked based on the pixel coordinates of the remote sensing image to be checked in the same name points, and calculating the edge error of each edge of each triangle in the triangulated network according to the pixel errors of each of the same name points; Differentiating the edges in the triangulated network by different colors based on the edge errors to characterize the distribution of internal geometric distortion of the remote sensing image to be inspected; Calculating the internal geometric distortion mean error of the remote sensing image to be checked according to each of the edge errors, and calculating the gross error rate of the remote sensing image to be checked based on each of the edge errors, a preset internal distortion error threshold and the total number of edges in the triangulated network; Taking the mean error of the internal geometric distortion of the remote sensing image to be inspected and the gross error rate of the remote sensing image to be inspected as the quality inspection result of the internal geometric distortion of the remote sensing image to be inspected; Wherein, different colors are used to distinguish the edges in the triangulated network based on the edge errors, including: If a first edge error in the edge errors is less than the internal distortion error threshold, setting the edge corresponding to the first edge error to a first color; If a second edge error in the edge errors is not less than the internal distortion error threshold, and the second edge error is less than 2 times the internal distortion error threshold, then the edge corresponding to the second edge error is set to a second color; If a third edge error in the edge errors is not less than 2 times the internal distortion error threshold, the edge corresponding to the third edge error is set to a third color.

2. The method according to claim 1, characterized in that: The remote sensing image to be checked includes any of the following: an original remote sensing image with an initial positioning model file, an orthophoto remote sensing image, and obtaining the same-name points of the remote sensing image to be checked and the benchmark reference image includes: Dividing the remote sensing image to be inspected into grids to obtain a plurality of grids; Using Harris feature point operator to extract feature points from each grid to obtain feature points of each grid; Determine a matching point that matches each of the feature points in the benchmark reference image, and use each of the feature points and the matching point that matches the feature point as an initial same-name point, wherein the initial same-name point includes: a same-name point number, a column coordinate of the remote sensing image to be checked, a row coordinate of the remote sensing image to be checked, a geographic X coordinate of the remote sensing image to be checked when the remote sensing image to be checked is an orthophoto remote sensing image, a geographic Y coordinate of the remote sensing image to be checked when the remote sensing image to be checked is an orthophoto remote sensing image, a column coordinate of the benchmark reference image, a row coordinate of the benchmark reference image, a geographic X coordinate of the benchmark reference image, a geographic Y coordinate of the benchmark reference image, and an elevation value determined based on the geographic X coordinate and Y coordinate of the benchmark reference image when the remote sensing image to be checked is an original remote sensing image; If the remote sensing image to be checked is the original remote sensing image, the RPC model or the collinear equation model is used to perform least square fitting on the coordinates of all the initial same-name points to obtain the absolute error of each of the initial same-name points; If the remote sensing image to be checked is the orthophoto remote sensing image, a geometric polynomial model is used to perform least square fitting on the coordinates of all the initial same-name points to obtain a geometric polynomial error of each of the initial same-name points; If the absolute error or geometric polynomial error of the first initial homonymous point among the initial homonymous points is greater than a preset error threshold, the first initial homonymous point is determined to be an incorrectly matched homonymous point, and the first initial homonymous point is deleted from the initial homonymous points, and the remaining initial homonymous points are used as the homonymous points.

3. The method according to claim 2, characterized in that The geographic coordinates of the benchmark reference image in the same-name point are converted, including: If the remote sensing image to be checked is the original remote sensing image, the geographic coordinates of the benchmark reference image in the same-name point are converted using the RPC model to obtain theoretical pixel coordinates of the geographic coordinates of the benchmark reference image on the remote sensing image to be checked; If the remote sensing image to be checked is the orthophoto remote sensing image, the geographical coordinates of the benchmark reference image in the same-name point are converted using the six-parameter conversion model of the remote sensing image to be checked to obtain the theoretical pixel coordinates of the geographical coordinates of the benchmark reference image on the remote sensing image to be checked; Calculating the pixel error of each of the same-name points according to the theoretical pixel coordinates and the pixel coordinates of the remote sensing image to be checked in the corresponding same-name point includes: Calculate the pixel error of the same name point according to the formula Calculate the pixel error of each of the same-name points, where RMS COL Represents the column error in the pixel error of the same name point, RMS ROW represents the row error in the pixel error of the same name point, line represents the theoretical column coordinates in the theoretical pixel coordinates, sample represents the theoretical row coordinates in the theoretical pixel coordinates, col src represents the column coordinates of the pixel coordinates of the remote sensing image to be inspected in the same name point corresponding to the theoretical pixel coordinates, row src Represents the row coordinates in the pixel coordinates of the remote sensing image to be inspected in the same-name point corresponding to the theoretical pixel coordinates.

4. The method according to claim 1, characterized in that Constructing a triangulated network in the remote sensing image to be checked based on the pixel coordinates of the remote sensing image to be checked in the same name points, and calculating the edge error of each edge of each triangle in the triangulated network according to the pixel errors of each of the same name points, including: Using the open source library CGAL to construct a triangulated network for the pixel coordinates of the remote sensing image to be inspected in the same-name points, to obtain a triangulated network in the remote sensing image to be inspected; According to the edge error calculation formula Calculate the edge error of each edge of each triangle in the triangulated network, where line rms represents the edge error, RMS COL1 Represents the column error in the pixel error of a same-name point that constitutes an edge, RMS COL2 Represents the column error in the pixel error of another point with the same name that constitutes the edge, RMS ROW1 Indicates the row error in the pixel error of a same-name point that constitutes an edge, RMS ROW2 Represents the row error in the pixel error of another point of the same name that forms the edge.

5. The method according to claim 1, characterized in that Calculating the internal geometric distortion mean error of the remote sensing image to be checked according to each of the edge errors, and calculating the gross error rate of the remote sensing image to be checked based on each of the edge errors, a preset internal distortion error threshold and the total number of edges in the triangulated network, including: According to the calculation formula of the error in internal geometric distortion Calculate the mean square error of the internal geometric distortion of the remote sensing image to be checked, where RMS represents the mean square error of the internal geometric distortion of the remote sensing image to be checked, line rmsi represents the edge error of the i-th edge, and n represents the total number of edges in the triangulated network; Determining the number of edge errors among the edge errors that are not less than 2 times the internal distortion error threshold according to each of the edge errors and the internal distortion error threshold; According to the calculation formula of gross error rate Calculate the gross error rate of the remote sensing image to be checked, where Cp represents the gross error rate of the remote sensing image to be checked, line num2 Indicates the number of edge errors in the edge errors that are not less than 2 times the internal distortion error threshold, line num Represents the total number of edges in the triangulated network.

6. The method according to claim 1, characterized in that The method further comprises: Determining whether the remote sensing image to be inspected is qualified according to the quality inspection result of the internal geometric distortion of the remote sensing image to be inspected, wherein if the error in the internal geometric distortion of the remote sensing image to be inspected is less than the internal distortion error threshold, and the gross error rate of the remote sensing image to be inspected is less than a preset internal distortion error threshold ratio, then it is determined that the remote sensing image to be inspected is qualified, otherwise, the remote sensing image to be inspected is unqualified; An internal accuracy quality inspection report of the remote sensing image to be inspected is generated based on the quality inspection result of the internal geometric distortion of the remote sensing image to be inspected, wherein the internal accuracy quality inspection report includes: the error in the internal geometric distortion of the remote sensing image to be inspected, the gross error rate of the remote sensing image to be inspected, the number of edge errors in the edge errors that are less than the internal distortion error threshold, the number of edge errors in the edge errors that are not less than the internal distortion error threshold and less than 2 times the internal distortion error threshold, the number of edge errors in the edge errors that are not less than 2 times the internal distortion error threshold, the number of triangles in the triangulated network, the total number of edges in the triangulated network, the maximum edge error in the edge errors, and whether the remote sensing image to be inspected is qualified.

7. A quantitative quality inspection device for internal geometric distortion of remote sensing images, characterized in that: include: An acquisition unit, used for acquiring the same-name points of the remote sensing image to be checked and the benchmark reference image, wherein the benchmark reference image is a standard orthophoto remote sensing image, and the same-name points are points where the remote sensing image to be checked and the benchmark reference image have the same ground object position; a conversion and calculation unit, configured to convert the geographic coordinates of the benchmark reference image in the same name point to obtain theoretical pixel coordinates of the geographic coordinates of the benchmark reference image on the remote sensing image to be checked, and calculate the pixel error of each of the same name points according to each of the theoretical pixel coordinates and the corresponding pixel coordinates of the remote sensing image to be checked in the same name point; A triangulated network construction and edge error calculation unit, used for constructing a triangulated network in the remote sensing image to be checked based on the pixel coordinates of the remote sensing image to be checked in the same name points, and calculating the edge error of each edge of each triangle in the triangulated network according to the pixel errors of each of the same name points; A color distinguishing unit, used for distinguishing each edge in the triangulated network by using different colors based on each edge error, so as to characterize the distribution of internal geometric distortion of the remote sensing image to be inspected; A calculation unit, used for calculating the internal geometric distortion mean error of the remote sensing image to be checked according to each of the edge errors, and calculating the gross error rate of the remote sensing image to be checked based on each of the edge errors, a preset internal distortion error threshold and the total number of edges in the triangulated network; A setting unit, used for taking the mean error of the internal geometric distortion of the remote sensing image to be checked and the gross error rate of the remote sensing image to be checked as the quality inspection result of the internal geometric distortion of the remote sensing image to be checked; Wherein, the color distinction unit is also used for: if a first edge error in the edge errors is less than the internal distortion error threshold, the edge corresponding to the first edge error is set to the first color; if a second edge error in the edge errors is not less than the internal distortion error threshold, and the second edge error is less than 2 times the internal distortion error threshold, the edge corresponding to the second edge error is set to the second color; if a third edge error in the edge errors is not less than 2 times the internal distortion error threshold, the edge corresponding to the third edge error is set to the third color.

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

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores machine-executable instructions. When the machine-executable instructions are called and executed by a processor, the machine-executable instructions cause the processor to execute the method according to any one of claims 1 to 6.

Images