A method for screening products of ZY-3 satellite stereo image pair sensor calibration

Abstract

The present invention provides a method for screening sensor correction products for the Ziyuan-3 satellite stereo image pair. The method comprises S1: obtaining image products and DEM products from the Ziyuan-3 satellite stereo image pair; S2: generating a single-view digital surface model from the image products; S3: constructing a spatially consistent feature descriptor based on the single-view digital surface model and the DEM product; and S4: judging the sensor correction product based on the spatially consistent feature descriptor. The present invention constructs a spatially consistent feature descriptor based on the single-view digital surface model generated from the single-view Ziyuan-3 satellite stereo image pair sensor correction product. The feature descriptor is used to determine whether the single-view digital surface model meets application requirements. The single-view Ziyuan-3 satellite stereo image pair sensor correction product corresponding to the single-view digital surface model that meets the application requirements is judged to be qualified.

Description

Technical Field

[0001] The present invention belongs to the technical field of data screening, and in particular relates to a method for screening correction products of ZY-3 satellite stereo image pairs sensors. Background Art

[0002] The Ziyuan-3 mapping satellite uses stereo imaging technology, which has several significant advantages in remote sensing applications, including: 1. Providing high-precision three-dimensional surface information. Stereo imaging technology can obtain three-dimensional information of the surface and generate high-precision digital elevation models (DFMs) and three-dimensional terrain models. This technology can accurately reflect the undulations and terrain features of the surface, providing precise data support for terrain analysis, natural resource management, etc. 2. Enhancing the geometric and geographic positioning accuracy of images. By simultaneously acquiring images from multiple perspectives, stereo imaging can improve the geometric and geographic positioning accuracy of images, which is particularly important for tasks such as high-precision map production, boundary confirmation, and geological exploration. 3. Improving the interpretation capability of remote sensing data. Stereo imaging technology can provide more perspectives and depth information, which facilitates the interpretation and analysis of remote sensing data. For example, in urban planning, stereo imaging can better identify urban features such as buildings and road networks. 4. Suitable for complex terrain and environments. For complex terrain and environments such as mountains, forests, and rivers, traditional single-view images may not provide sufficient information. Stereoscopic imaging technology can overcome these challenges, providing more comprehensive surface data and facilitating remote sensing missions in complex terrain. 5. Supporting a variety of application needs: Due to its high precision and multi-angle imaging capabilities, stereoscopic imaging technology is suitable for a wide range of applications, including but not limited to urban planning, resource management, environmental monitoring, and disaster assessment. It provides these fields with rich data resources and analytical tools.

[0003] Sensor-calibrated imagery products are created by performing sensor calibration on radiometrically corrected images. Sensor calibration corrects for geometric distortion caused by platform motion and scanning rate, eliminates detector alignment errors and optical system distortion, and eliminates or mitigates various distortions or systematic errors in satellite imaging. It also enables seamless integration of sliced CCD images and constructs geometric models for image formation. Sensor-calibrated imagery products are pre-processed, entry-level products ready for subsequent applications.

[0004] Currently, the quality inspection of sensor correction products is oriented towards single-view images. Due to the particularity of remote sensing and mapping satellite stereo imaging technology, there is still a lack of data screening methods for stereo image pair sensor correction products.

[0005] Image quality issues caused by atmospheric conditions and cloud cover, such as large areas of thin cloud, while not affecting feature interpretation in subsequent applications, can reduce the clarity and usability of single-view digital surface models, making it impossible to produce high-precision digital elevation models. Image quality and resolution, such as garbled or tapped images in single-view stereo images, can directly impact the accuracy and detail of the resulting 3D terrain model.

[0006] Therefore, it is of great significance to screen and develop the ZY-3 satellite stereo image pair sensor correction products and related technical methods that meet the needs of subsequent engineering applications. Summary of the Invention

[0007] The present invention provides a method for screening products of sensor calibration for stereo image pairs of ZY-3 satellites, characterized in that the method comprises the following steps:

[0008] S1: Obtain stereo image products and DEM products from the Ziyuan-3 satellite;

[0009] S2: Generate a single-scene digital surface model from image products;

[0010] S3: Construct spatial consistency feature descriptors based on single-scene digital surface models and DEM products;

[0011] S4: Determine the sensor correction product based on the spatial consistency feature descriptor.

[0012] Furthermore, the step S2 includes

[0013] S21 performs stereo matching on satellite stereo image pairs and calculates disparity maps;

[0014] S22 generates a three-dimensional point cloud based on the disparity map;

[0015] S23 converts the 3D point cloud coordinates to WGS84 geographic coordinates;

[0016] S24 filters and corrects the 3D point cloud in WGS84 geographic coordinates;

[0017] S25 rasterizes the corrected three-dimensional point cloud to generate grid DSM data.

[0018] Furthermore, step S21 includes stereo matching of the satellite stereo image pair, that is, finding corresponding pixel pairs in the left and right images. After completing the stereo matching, the horizontal displacement of the corresponding pixel in the left image to the corresponding pixel in the right image is calculated based on the value of each pixel (u, v), that is, the disparity value r(u, v), to obtain a disparity map.

[0019] Furthermore, the step S22 includes calculating the corresponding ground object coordinates X(u, v), Y(u, v), and Z(u, v) for each disparity value r(u, v) in the disparity map to form an initial three-dimensional point cloud;

[0020] X(u, v): X-axis coordinate value of the object point in the camera coordinate system;

[0021] Y(u, v): Y-axis coordinate value of the object point in the camera coordinate system;

[0022] Z(u, v): The Z-axis coordinate value of the object point in the camera coordinate system, that is, the distance from the camera to the object point.

[0023] Furthermore, for each pixel (u, v) at a disparity r(u, v), the object coordinate Z(u, v) can be calculated using the following formula:

[0024]

[0025] Where: w is the focal length of the camera, which is obtained from the parameters of the satellite image; F is the baseline length between cameras, which is obtained from the parameters of the satellite image; r(u, v) is the disparity value measured at the pixel (u, v).

[0026] Furthermore, the step S3 includes

[0027] S31 builds terrain dataset;

[0028] S32 constructs the spatial consistency feature function β;

[0029] S33 constructs the spatial consistency feature descriptor μ.

[0030] Furthermore, the step S31 includes

[0031] Set (x h ,y h , z h ) represents the DSM grid point dataset E, (x h ,y h , z h ) represents the spatial coordinates of the grid points, h = [1, 2, ..., M], M represents the number of points in E, and the spatial separation distance between adjacent grid points in the x and y directions is 1,

[0032] According to the spatial coordinates of each grid point of DSM, the inclination angle θ of each grid point of the auxiliary data DEM product is found and calculated;

[0033] DSM grid points with θ < 2° are selected to construct the terrain dataset A, which is the first dataset;

[0034] DSM grid points with 2°≤θ<6° are selected to construct terrain dataset B, i.e. the second dataset;

[0035] DSM grid points with 6°≤θ<25° are selected to construct terrain dataset C, which is the third dataset;

[0036] The DSM grid points with θ ≥ 25° are selected to construct the terrain dataset D, which is the fourth dataset;

[0037] Get four terrain data sets A, B, C, and D;

[0038] A={a i}, i = [1, 2, ..., O(g)], O(g) represents the number of points in A, Represents grid point a i The spatial coordinates of

[0039] B={b j}, j = [1, 2, ..., P(g)], P(g) represents the number of points in B, Represents grid point b j The spatial coordinates of

[0040] C={c k}, k = [1, 2, ..., Q(g)], Q(g) represents the number of points in C, Represents grid point c k The spatial coordinates of

[0041] D={d l}, l = [1, 2, ..., R(g)], R(g) represents the number of points in D, Represents the grid point d l The spatial coordinates of .

[0042] Furthermore, the step S32 includes

[0043] Let Z(x, y) be a random variable, Z(x, y) is the value of attribute Z at the spatial position (x, y), g is the spatial separation distance between the two sample points, and They are the spatial positions of the variables Z(x, y) and z h The calculation formula of spatial consistency characteristic function β is:

[0044]

[0045] T={β a (g)|β a (g) < γ};

[0046] U={βb (g)|β b (g) < δ};

[0047] V={β c (g)|β c (g)<ε};

[0048] W={β d (g)|β d (g)<τ};

[0049] in:

[0050] β a (g) is the spatial consistency feature function of data set A;

[0051] β b (g) is the spatial consistency feature function of data set B;

[0052] β c (g) is the spatial consistency feature function of data set C;

[0053] β c (g) is the spatial consistency feature function of data set D;

[0054] Datasets T, U, V, and W are spatial consistency feature function datasets for different terrains;

[0055] γ, δ, ε, and τ are the spatial consistency feature thresholds.

[0056] Furthermore, the step S33 includes establishing a correspondence between the data sets T, U, V, W and the spatial consistency feature descriptor μ;

[0057]

[0058] Furthermore, step S4 includes judging the sensor correction product based on the spatial consistency feature descriptor μ. When μ=1, the single-scene digital surface model meets the application requirements, and the single-scene Ziyuan-3 satellite stereo image pair sensor correction product corresponding to the single-scene digital surface model that meets the application requirements is judged to be qualified.

[0059] Beneficial effects:

[0060] The present invention provides a data screening method based on the Ziyuan-3 satellite stereo image pair sensor correction product for the construction of three-dimensional terrain models. A spatial consistency feature descriptor is constructed based on a single-view digital surface model generated from the single-view Ziyuan-3 satellite stereo image pair sensor correction product. The feature descriptor is used to determine whether the single-view digital surface model meets application requirements. Single-view Ziyuan-3 satellite stereo image pair sensor correction products corresponding to single-view digital surface models that meet application requirements are determined to be qualified. This achieves data screening.

[0061] It should be understood that the above general description and the following detailed description are only exemplary and illustrative and do not limit the scope of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a flow chart of a method for screening products of the ZY-3 satellite stereo image pair sensor calibration according to the present invention;

[0063] It should be understood that the drawings are not necessarily drawn to scale, presenting a somewhat simplified representation of various features illustrative of the basic principles of the present disclosure. The specific design features of the present invention as disclosed herein, including, for example, specific dimensions, orientations, positions, and shapes will be determined in part by the particular intended application and use environment.

[0064] In the drawings, reference numbers refer to the same or equivalent parts of the present invention throughout the several figures of the drawing. DETAILED DESCRIPTION

[0065] Reference will now be made in detail to various embodiments of the present invention, examples of which are illustrated in the accompanying drawings and described below. Although the present invention will be described in conjunction with exemplary embodiments of the present invention, it should be understood that this description is not intended to limit the present invention to those exemplary embodiments. On the other hand, the present invention is intended to cover not only the exemplary embodiments of the present invention, but also various alternatives, modifications, equivalents and other embodiments, which may be included within the spirit and scope of the present invention as defined by the appended claims.

[0066] Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. The specific structures and functions described in the exemplary embodiments of the present invention are for illustrative purposes only. The embodiments according to the concepts of the present invention may be implemented in various forms, and it should be understood that they should not be interpreted as being limited to the exemplary embodiments described in the exemplary embodiments, but include all modifications, equivalents or alternatives included in the spirit and scope of the present invention.

[0067] Throughout the specification, the technical terms used herein are for the purpose of describing various exemplary embodiments only and are not intended to be limiting. It will be further understood that the terms "comprises," "comprising," "having," etc., when used in exemplary embodiments, specifically refer to the presence of the stated parts, steps, operations, or elements, but do not preclude the presence or addition of one or more other parts, steps, operations, or elements.

[0068] like Figure 1 As shown, the present invention provides a method for screening sensor calibration products of the Ziyuan-3 satellite stereo image pair, the method comprising

[0069] (1) Obtain stereo image products and DEM products from the Ziyuan-3 satellite;

[0070] Obtain the single-view Ziyuan-3 satellite stereo image pair sensor correction product. The stereo image pair contains two image products, one of which is the left image and the other is the right image; obtain the auxiliary data DEM product: SRTM 30m product.

[0071] (2) Generate a single-scene digital surface model (DSM) from image products;

[0072] (2.1) Stereo matching of satellite stereo image pairs and calculation of disparity maps;

[0073] Stereo matching of satellite stereo image pairs involves finding corresponding pixel pairs in the left and right images. This can be accomplished using various stereo matching algorithms. After stereo matching, the value of each pixel (u, v), or the disparity value r(u, v), represents the horizontal displacement from the corresponding pixel in the left image to the corresponding pixel in the right image. Based on the required disparity accuracy, only the disparity values that meet the accuracy requirements are retained, i.e., the valid disparity values r(u, v), resulting in a valid disparity map.

[0074] (2.2) Generate a 3D point cloud based on the disparity map;

[0075] For each valid disparity value r(u, v) in the valid disparity map, the corresponding object coordinates (X, Y, Z) are calculated; these object coordinates constitute the initial 3D point cloud;

[0076] X: The X-axis coordinate value of the object point in the camera coordinate system;

[0077] Y: Y-axis coordinate value of the object point in the camera coordinate system;

[0078] Z: The Z-axis coordinate value of the object point in the camera coordinate system, that is, the distance from the camera to the object point;

[0079] For each pixel (u, v) at the disparity r(u, v), the object coordinates Z(u, v) can be calculated using the following formula:

[0080]

[0081] in:

[0082] w is the focal length of the camera, which can be obtained from the parameters of the satellite image;

[0083] F is the baseline length between cameras, which can be obtained from the parameters of satellite images;

[0084] r(u, v) is the disparity value measured at pixel (u, v);

[0085] The object coordinates X(u, v) and Y(u, v) can be derived from the geometric relationship between the parallax r(u, v) and the internal and external parameters of the camera;

[0086] Once the object coordinates Z(u, v) are obtained, the coordinates (X c , Y c , Z c ) is converted into coordinates (X, Y, Z) in the ground object coordinate system.

[0087] Specifically, let the coordinates in the camera coordinate system be (X c , Y c , Z c ), then the feature coordinates (X, Y) can be calculated by the following steps:

[0088]

[0089] (2.3) Convert the initial 3D point cloud coordinates to WGS84 geographic coordinates;

[0090] (2.4) Filtering and correcting the initial 3D point cloud in WGS84 geographic coordinates to remove inaccurate points caused by mismatching or noise, and obtaining a corrected 3D point cloud;

[0091] (2.5) The corrected 3D point cloud is rasterized to generate grid DSM data.

[0092] (3) Construct spatial consistency feature descriptors based on single-scene digital surface models and DEM products;

[0093] (3.1) Construct terrain dataset;

[0094] Based on the result of step (2.4), set (x h ,y h , z h ) represents the DSM grid point dataset E, (x h ,y h , z h ) represents the spatial coordinates of the grid points, h = [1, 2, ..., M], M represents the number of points in E. Assume that the spatial separation distance between adjacent grid points in the x-direction and the y-direction is 1.

[0095] According to the spatial coordinates of each grid point of DSM, the inclination angle θ of each grid point of the auxiliary data DEM product is found and calculated;

[0096] DSM grid points with θ < 2° are selected to construct the terrain dataset A, which is the first dataset;

[0097] DSM grid points with 2°≤θ<6° are selected to construct terrain dataset B, i.e. the second dataset;

[0098] DSM grid points with 6°≤θ<25° are selected to construct terrain dataset C, which is the third dataset;

[0099] The DSM grid points with θ ≥ 25° are selected to construct the terrain dataset D, which is the fourth dataset;

[0100] Get four terrain data sets A, B, C, and D. E = A∪B∪C∪D;

[0101] A={a i}, i = [1, 2, ..., O(g)], O(g) represents the number of points in A, Represents grid point a i The spatial coordinates of

[0102] B={b j}, j = [1, 2, ..., P(g)], P(g) represents the number of points in B, Represents grid point b j The spatial coordinates of

[0103] C={c k}, k = [1, 2, ..., Q(g)], Q(g) represents the number of points in C, Represents the spatial coordinates of the grid point cx;

[0104] D={d l}, l = [1, 2, ..., R(g)], R(g) represents the number of points in D, Represents the grid point d l The spatial coordinates of .

[0105] (3.2) Constructing spatial consistency feature function β;

[0106] Let Z(x, y) be a random variable, Z(x, y) is the value of attribute Z at the spatial position (x, y), g is the spatial separation distance between the two sample points, and They are the spatial positions of the variables Z(x, y) and z h value, then the calculation formula of the spatial consistency characteristic function β is:

[0107]

[0108]

[0109] T={β a(g)|β a (g) < γ};

[0110] U={β b (g)|β b (g) < δ};

[0111] V={β c (g)|β c (g)<ε};

[0112] W={β d (g)|β d (g)<τ};

[0113] in:

[0114] β a (g) is the spatial consistency feature function of data set A;

[0115] β b (g) is the spatial consistency feature function of data set B;

[0116] β c (g) is the spatial consistency feature function of data set C;

[0117] β d (g) is the spatial consistency feature function of data set D;

[0118] Datasets T, U, V, and W are spatial consistency feature function datasets for different terrains;

[0119] γ, δ, ε, and τ are the spatial consistency feature thresholds.

[0120] (3.3) Construct spatial consistency feature descriptor μ;

[0121] Establish the corresponding relationship between the data set T, U, V, W and the spatial consistency feature descriptor μ;

[0122]

[0123] (4) Determination of sensor correction products based on spatial consistency feature descriptors for single-view ZY-3 satellite stereo images;

[0124] The sensor-corrected products are evaluated based on the spatial consistency feature descriptor μ. When μ = 1, the single-view digital surface model meets the application requirements. The sensor-corrected products corresponding to the single-view Ziyuan-3 satellite stereo image pair that meet the application requirements are considered qualified. In summary, data screening is achieved.

[0125] The above-mentioned embodiments of the present application can be implemented in various hardware, software coding, or a combination of the two. For example, the embodiments of the present application can also represent program code for executing the above-mentioned method in a digital signal processor. The present application can also relate to various functions performed by a computer processor, a digital signal processor, a microprocessor, or a field programmable gate array. The above-mentioned processor can be configured according to the present application to perform specific tasks, which are completed by executing machine-readable software code or firmware code that defines the specific methods disclosed in the present application. The software code or firmware code can be developed to represent different programming languages and different formats or forms. It can also represent different target platform compiled software codes. However, the different code styles, types, and languages of the software code for performing tasks according to the present application and other types of configuration code do not depart from the spirit and scope of the present application.

[0126] The foregoing description of specific exemplary embodiments of the present invention has been presented for purposes of illustration and description. It is not intended to exclude or limit the invention to the precise form disclosed, and it is apparent that many modifications and variations are possible in light of the above teachings. The exemplary embodiments have been selected and described to explain certain principles of the present invention and their practical application so as to enable others skilled in the art to make or utilize the various exemplary embodiments of the present invention, and various alternatives and modifications thereof. It is intended that the scope of the present invention be defined by the claims appended hereto and their equivalents.

[0127] It will be understood that the above embodiments are merely exemplary embodiments for illustrating the principles of the present invention, and the present invention is not limited thereto. Those skilled in the art will appreciate that various modifications and improvements can be made without departing from the spirit and substance of the present invention, and such modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A method for screening products of ZY-3 satellite stereo image pair sensor calibration, characterized in that: The method comprises S1: Obtain stereo image products and DEM products from the Ziyuan-3 satellite; S2: Generate a single-scene digital surface model from image products; S3: Construct spatial consistency feature descriptors based on single-scene digital surface models and DEM products; including: S31 builds terrain dataset, set up Represents a DSM grid point dataset, where represents the spatial coordinates of the grid points, , represents the number of points in E, and the adjacent grid points are Direction and The directional space separation distance is 1; According to the spatial coordinates of each grid point of DSM, the inclination angle of each grid point of the auxiliary data DEM product is found and calculated. ; Select The DSM grid points are used to construct the terrain dataset A, which is the first dataset; Select The DSM grid points are used to construct the terrain dataset B, which is the second dataset; Select The DSM grid points are used to construct the terrain dataset C, which is the third dataset; Select The DSM grid points are used to construct the terrain dataset D, which is the fourth dataset; Get four terrain data sets A, B, C, and D; , , represents the number of points in A, Represents grid points The spatial coordinates of , , represents the number of points in B, Represents grid points The spatial coordinates of , , represents the number of points in C, Represents grid points The spatial coordinates of , , represents the number of points in D, Represents grid points The spatial coordinates of S32 builds spatial consistency feature function , set up is a random variable, is an attribute In spatial position The value at is the spatial separation distance between two sample points, and The variables are In spatial position and at Value, spatial consistency feature function The calculation formula is: ； ； ； ； ； ； ； ； in: is the spatial consistency feature function of data set A; is the spatial consistency feature function of data set B; is the spatial consistency feature function of data set C; is the spatial consistency feature function of the dataset D; Dataset It is a dataset of spatially consistent characteristic functions for different terrains; is the spatial consistency feature threshold; S33 builds spatial consistency feature descriptor , including building a dataset Spatially consistent feature descriptors The corresponding relationship; ； S4: Determine the sensor correction product based on the spatial consistency feature descriptor.

2. The method for screening ZY-3 satellite stereo image pair sensor calibration products according to claim 1, characterized in that: The step S2 includes: S21 performs stereo matching on satellite stereo image pairs and calculates disparity maps; S22 generates a three-dimensional point cloud based on the disparity map; S23 converts the 3D point cloud coordinates to WGS84 geographic coordinates; S24 filters and corrects the 3D point cloud in WGS84 geographic coordinates; S25 rasterizes the corrected three-dimensional point cloud to generate grid DSM data.

3. The method for screening ZY-3 satellite stereo image pair sensor calibration products according to claim 2, characterized in that: The step S21 includes performing stereo matching on the satellite stereo image pair, that is, finding the corresponding pixel point pairs in the left and right images, and after completing the stereo matching, according to each pixel The value of the horizontal displacement from the corresponding pixel in the left image to the corresponding pixel in the right image, that is, the disparity value , and get the disparity map.

4. The method for screening ZY-3 satellite stereo image pair sensor calibration products according to claim 2, characterized in that: The step S22 includes: for each disparity value in the disparity map , calculate the corresponding ground object coordinates , forming the initial three-dimensional point cloud; :The ground point in the camera coordinate system Axis coordinate values; :The ground point in the camera coordinate system Axis coordinate values; :The ground point in the camera coordinate system Axis coordinate value, that is, the distance from the camera to the object point.

5. The method for screening ZY-3 satellite stereo image pair sensor calibration products according to claim 4, characterized in that: For each pixel Parallax on , the coordinates of the object can be calculated by the following formula ; ； in: is the focal length of the camera, obtained from the parameters of the satellite image; is the baseline length between cameras, obtained from the parameters of satellite images; It is in pixels The parallax value measured at .

6. The method for screening ZY-3 satellite stereo image pair sensor calibration products according to claim 1, characterized in that: The step S4 includes the following steps: To judge the sensor calibration product, set When the single-scene digital surface model meets the application requirements, the single-scene Ziyuan-3 satellite stereo image pair sensor correction product corresponding to the single-scene digital surface model that meets the application requirements is judged to be qualified.

Images