Positioning accuracy evaluation method
By resampling and edge extraction of remote sensing images, and combining high-resolution shoreline data to calculate positioning deviations, the limitations of relying on test sites or other satellite images in the existing technology are solved, and the geolocation accuracy of remote sensing images that does not rely on external data is achieved. Evaluation simplifies the calculation process and improves the universal applicability of the evaluation.
Patent Information
- Application Number
- CN202210089008.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-01-25
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2042-01-25
AI Technical Summary
The existing remote sensing image geolocation accuracy evaluation methods rely mostly on actual data of the test site control point or other satellite image information. The implementation is limited, making it difficult to quickly and conveniently obtain control points for positioning accuracy evaluation.
By resampling and edge extraction of remote sensing images, high-resolution shoreline data are used to calculate the arc length of the earth's surface as the positioning deviation, quantitative evaluation is performed based on the global self-consistent stratification high-resolution geodatabase, and multiple sets of positioning results are statistically analyzed. Evaluate geolocation errors.
The geolocation accuracy evaluation of remote sensing images that do not rely on the same scene data of other satellites is realized, which simplifies the calculation process and improves the universal applicability and simplicity of the evaluation.
Smart Images

Figure CN114529520B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of remote sensing image geo-positioning, and in particular to a positioning accuracy evaluation method, in particular to a remote sensing image geo-positioning accuracy evaluation method based on high-resolution shoreline data. Background Art
[0002] With the rapid development of remote sensing applications, the geographic positioning accuracy of remote sensing images has become one of the important indicators to measure the application value of remote sensing images. This requires a quantitative evaluation of the positioning accuracy of remote sensing images.
[0003] The existing remote sensing image geolocation evaluation methods are all based on the comparison of the positions of control points with the same name. How to quickly and conveniently obtain and match control points is of great significance for the evaluation of remote sensing image geolocation accuracy. Existing positioning accuracy evaluation methods mostly rely on the measured data of test site control points or other satellite image information, and their implementation is subject to certain limitations.
[0004] The control point acquisition methods given in GJB 4407-2002 "Space Remote Sensing Image Positioning Accuracy Detection Method" include: field detection method, topographic map (library) reading point detection method and photogrammetry method. Existing literature 1 [Chen Yixia, Research on Remote Sensing Image Geometric Positioning Accuracy Evaluation Method, Nanjing University of Science and Technology, Master's Thesis, 2013] gives a method for automatic selection and matching of control points based on SURF algorithm and field edge information. Existing literature 2 [Wei Dandan, Gan Fuping, Shang Kun, et al., Geometric Positioning Accuracy Evaluation of CBERS-04 Satellite PAN / MUX Images, Radio Engineering, 2018, 48(5)] gives a positioning accuracy evaluation method based on field measured control points and other satellite images. Existing literature 3 [Lin Hongdi, Qu Lina, Research on positioning accuracy evaluation of GeoEye-1 satellite under uncontrolled autonomous positioning and sparse control points, Urban Surveying, 2018, 6] and existing literature 4 [Wang Jianbu, Zhang Jie, Ma Yi, Positioning accuracy evaluation of secondary products of Resource-1 02C satellite remote sensing images, Marine Surveying and Mapping, 2013, 33(5)] respectively give methods for evaluating positioning accuracy based on control points obtained from the test site. The patent document with publication number CN104574347A discloses a method for evaluating the geometric positioning accuracy of an on-orbit satellite image based on multi-source remote sensing data, which adopts the following steps: Step 1, adjusting the image to be evaluated and the reference image to two images under the same ellipsoid, reference plane and resolution; Step 2, downsampling the above two images and performing radiation enhancement processing; Step 3, using the accelerated robust feature Surf algorithm to perform coarse matching on the above two images, and using epipolar geometric constraints to eliminate mismatched point pairs; Step 4, based on the coarse matching result, performing geometric relationship compensation on the image to be evaluated, and accurately dividing the image to be evaluated and the reference image after geometric compensation into blocks; Step 5, using the Surf algorithm to perform precise matching on the block pairs of the image to be evaluated and the epipolar geometric constraints to eliminate mismatched point pairs; Step 6, calculating the external geometric positioning accuracy, and at the same time calculating the internal geometric positioning accuracy based on the screened control point pairs in each direction.
[0005] The above-mentioned literature still has the defect of relying too much on measured data of test site control points or other satellite image information, and its implementation is subject to certain limitations. Summary of the invention
[0006] In view of the defects in the prior art, an object of the present invention is to provide a positioning accuracy evaluation method.
[0007] A positioning accuracy evaluation method provided by the present invention comprises the following steps:
[0008] Step 1: Resample the original image according to the geolocation result to obtain the resampled image and the longitude and latitude corresponding to each pixel;
[0009] Step 2: Perform edge extraction on the remote sensing image with the coastline to obtain pixels corresponding to the coastline in the remote sensing image;
[0010] Step 3: For the shoreline pixels in the remote sensing image, search for corresponding points in the high-resolution shoreline data;
[0011] Step 4: For the shoreline pixels in the remote sensing image, calculate the corresponding arc length of the earth's surface as the positioning deviation according to the longitude and latitude corresponding to the pixel and the longitude and latitude of the corresponding point in the high-resolution shoreline data;
[0012] Step 5: Statistically analyze the remote sensing image geographic positioning error based on multiple sets of positioning results.
[0013] Preferably, in step 1, the resampled grid is set to an equal longitude and equal latitude interval grid generated by geographic cylindrical projection, and the discrete points of the original positioning results are resampled to points on the equal longitude and equal latitude interval grid.
[0014] Preferably, step 2 comprises the following steps:
[0015] Step 2.1: Extract the edge of the remote sensing image to obtain a binary image of the remote sensing image edge;
[0016] Step 2.2: Project the longitude and latitude discrete points in the high-resolution shoreline data to the equal longitude and latitude interval grids in step 1 to generate a binary image with the same resolution as the remote sensing image;
[0017] Step 2.3: Perform a dilation operation in image morphological processing on the binary image in step 2.2;
[0018] Step 2.4: Use the binary image obtained in step 2.3 as a mask to filter the edge of the remote sensing image in step 2.1 as the shoreline pixels extracted from the remote sensing image.
[0019] Preferably, in step 2.1, the Sobel operator is used to extract the edge of the remote sensing image.
[0020] Preferably, in step 2.3, the structural element is D,
[0021]
[0022] Pixels where there is a high-resolution shoreline are marked as 1, and 0 otherwise.
[0023] Preferably, in step 3, high-resolution shoreline data is obtained from a global self-consistent hierarchical high-resolution geographic database.
[0024] Preferably, in step 3, for the shoreline pixels in the remote sensing image, corresponding points in the high-resolution shoreline data are searched according to the principle of closest surface distance.
[0025] Preferably, in step 4, if the latitude corresponding to the shoreline pixel in the remote sensing image is a and the longitude is b (unit: radians), and the α and longitude of the corresponding point in the high-resolution shoreline data are β, the calculation formula for the positioning deviation d is:
[0026] d=Aarccos(sin(a)cos(b)sin(α)cos(β)+sin(a)sin(b)sin(α)sin(β)+cos(a)cos(α))
[0027] Where A is the average radius of the Earth.
[0028] Preferably, the calculation method of the positioning error δ in step 5 is:
[0029]
[0030] in, is the average value of multiple positioning deviations.
[0031] Preferably, the number of groups of positioning deviations is not less than 50.
[0032] Compared with the prior art, the present invention has the following beneficial effects:
[0033] 1. The present invention can evaluate the geographic positioning accuracy of remote sensing images without relying on positioning data of the same scene from other satellites;
[0034] 2. The present invention is based on the Global Self-Consistent Hierarchical High-Resolution Geographic Database (GSHHG), and compares the water-land boundary position in the remote sensing image with the high-resolution coastline database to quantitatively evaluate the geographic positioning accuracy of the remote sensing image;
[0035] 3. The method of the present invention is reasonable, simple to calculate, and easy to implement, and can be widely used in the accuracy evaluation of remote sensing image theorems. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:
[0037] Figure 1 is a flow chart of the present invention;
[0038] Figure 2 It is a satellite original remote sensing image;
[0039] Figure 3 is the resampled remote sensing image;
[0040] Figure 4 The high-resolution shoreline data is projected to the grid result with the same resolution as the remote sensing image;
[0041] Figure 5 is the pixel corresponding to the shoreline in the remote sensing image;
[0042] Figure 6 It is the statistical result of positioning deviation based on high-resolution shoreline data. DETAILED DESCRIPTION
[0043] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.
[0044] Embodiment 1:
[0045] like Figure 1 As shown, a positioning accuracy evaluation method provided in this embodiment includes the following steps:
[0046] Step 1: Resample the original image according to the geolocation result to obtain the resampled image and the longitude and latitude corresponding to each pixel. In step 1, the resampled grid is set to an equal longitude and latitude interval grid generated by the geographic cylindrical projection, and the discrete points of the original positioning result are resampled to points on the equal longitude and latitude interval grid.
[0047] Step 2: Perform edge extraction on the remote sensing image with the coastline to obtain the pixels corresponding to the coastline in the remote sensing image.
[0048] Step 2 includes the following steps:
[0049] Step 2.1: Extract the edge of the remote sensing image to obtain a binary image of the remote sensing image edge;
[0050] Step 2.2: Project the longitude and latitude discrete points in the high-resolution shoreline data to the equal longitude and latitude interval grids in step 1 to generate a binary image with the same resolution as the remote sensing image. In step 2.1, the Sobel operator is used to extract the edge of the remote sensing image;
[0051] Step 2.3: Perform a dilation operation on the binary image in step 2.2 in image morphological processing. In step 2.3, the structural element is D,
[0052]
[0053] Pixels with high-resolution shorelines are marked as 1, otherwise 0;
[0054] Step 2.4: Use the binary image obtained in step 2.3 as a mask to filter the edge of the remote sensing image in step 2.1 as the shoreline pixels extracted from the remote sensing image.
[0055] Step 3: For the shoreline pixels in the remote sensing image, search for corresponding points in the high-resolution shoreline data. In step 3, the high-resolution shoreline data is obtained from the global self-consistent layered high-resolution geographic database. For the shoreline pixels in the remote sensing image, search for corresponding points in the high-resolution shoreline data according to the principle of the shortest distance to the surface.
[0056] Step 4: For the shoreline pixels in the remote sensing image, calculate the corresponding arc length of the earth's surface as the positioning deviation according to the longitude and latitude corresponding to the pixel and the longitude and latitude of the corresponding point in the high-resolution shoreline data. In step 4, if the latitude and longitude corresponding to the shoreline pixel in the remote sensing image are a and b (unit: radians), and the α and longitude of the corresponding point in the high-resolution shoreline data are β, the positioning deviation d is calculated as follows:
[0057] d=Aarccos(sin(a)cos(b)sin(α)cos(β)+sin(a)sin(b)sin(α)sin(β)+cos(a)cos(α))
[0058] Where A is the average radius of the Earth.
[0059] Step 5: Statistically analyze the geographic positioning error of remote sensing images based on multiple sets of positioning results. The calculation method of the positioning error δ in step 5 is:
[0060]
[0061] in, It is the average value of multiple groups of positioning deviations. The number of groups of positioning deviations shall not be less than 50.
[0062] The method of the present invention is verified by combining a certain satellite remote sensing image. Figure 2 The figure shows the original remote sensing image of a satellite. Since a 45° rotating scanning reflector is used in the imaging process of the payload, image rotation exists in the original image. Figure 3 is the resampled remote sensing image obtained after executing step 1. In step 2, the high-resolution shoreline data is projected to the grid with the same resolution as the remote sensing image. Figure 4 As shown in the figure, after executing step 2, the pixels corresponding to the shoreline in the remote sensing image are obtained as follows Figure 5 As shown. Figure 5 It can be seen that due to factors such as clouds, the shoreline pixels extracted from remote sensing images are not continuous. Figure 5 For the shoreline pixels in the image, execute steps 3 to 5, and the statistical results of 1349 sets of positioning deviations are as follows: Figure 6 shown.
[0063] The present invention is independent of other satellite images of the same scene and does not require field measurement of ground control points. It is easy to implement and can be widely applied to the evaluation of the geographic positioning accuracy of remote sensing images.
[0064] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.
Claims
1. A positioning accuracy evaluation method, characterized in that: The following steps are involved: Step 1: Resample the original image according to the geolocation result to obtain the resampled image and the longitude and latitude corresponding to each pixel; Step 2: Perform edge extraction on the remote sensing image with the coastline to obtain pixels corresponding to the coastline in the remote sensing image; Step 3: For the shoreline pixels in the remote sensing image, search for corresponding points in the high-resolution shoreline data; Step 4: For the shoreline pixels in the remote sensing image, calculate the corresponding arc length of the earth's surface as the positioning deviation according to the longitude and latitude corresponding to the pixel and the longitude and latitude of the corresponding point in the high-resolution shoreline data; Step 5: Statistically analyze the geographic positioning error of remote sensing images based on multiple sets of positioning results; In step 3, high-resolution shoreline data is obtained from a global self-consistent hierarchical high-resolution geographic database.
2. The positioning accuracy evaluation method according to claim 1, characterized in that: In step 1, the resampled grid is set to an equal longitude and equal latitude interval grid generated by geographic cylindrical projection, and the discrete points of the original positioning result are resampled to points on the equal longitude and equal latitude interval grid.
3. The positioning accuracy evaluation method according to claim 2, characterized in that: The step 2 comprises the following steps: Step 2.1: Extract the edge of the remote sensing image to obtain a binary image of the remote sensing image edge; Step 2.2: Project the longitude and latitude discrete points in the high-resolution shoreline data to the equal longitude and latitude interval grids in step 1 to generate a binary image with the same resolution as the remote sensing image; Step 2.3: Perform a dilation operation in image morphological processing on the binary image in step 2.2; Step 2.4: Use the binary image obtained in step 2.3 as a mask to filter the edge of the remote sensing image in step 2.1 as the shoreline pixels extracted from the remote sensing image.
4. The positioning accuracy evaluation method according to claim 3, characterized in that: In step 2.1, the Sobel operator is used to extract the edge of the remote sensing image.
5. The positioning accuracy evaluation method according to claim 3, characterized in that: In step 2.3, the structural element is D, Pixels where there is a high-resolution shoreline are marked as 1, and 0 otherwise.
6. The positioning accuracy evaluation method according to claim 1, characterized in that: In step 3, for the shoreline pixels in the remote sensing image, corresponding points in the high-resolution shoreline data are searched according to the principle of the shortest distance to the surface.
7. The positioning accuracy evaluation method according to claim 1, characterized in that: In step 4, if the latitude corresponding to the shoreline pixel in the remote sensing image is a, the longitude is b, and the unit is radian, and the latitude of the corresponding point in the high-resolution shoreline data is α, and the longitude is β, the positioning deviation d is calculated as follows: d=Aarccos(sin(a)cos(b)sin(α)cos(β)+sin(a)sin(b)sin(α)sin(β)+cos(a)cos(α)) where A is the average radius of the earth.
8. The positioning accuracy evaluation method according to claim 1, characterized in that: The calculation method of positioning error δ in step 5 is: in, is the average value of multiple positioning deviations.
9. The positioning accuracy evaluation method according to claim 8, characterized in that: The number of positioning deviation groups shall not be less than 50.
