Adaptive grid-based star map identification method
By using quadtree adaptive grid partitioning and binary encoding to optimize star map recognition, the problem of the inability to adjust the number of grids in the traditional grid method is solved, and efficient and accurate star map recognition is achieved.
Patent Information
- Application Number
- CN202511561962.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-29
- Publication Date
- 2026-02-03
AI Technical Summary
Traditional grid-based methods cannot adaptively adjust the number of grids when identifying star maps, leading to problems such as missing codes or slow recognition speed.
A quadtree method is used for adaptive grid partitioning, combined with binary encoding and a similarity function to optimize the star map recognition process.
It improves the efficiency and accuracy of star map recognition, avoids the problems of missing codes and slow recognition speed, and maintains the advantages of low algorithm complexity and fast recognition speed of traditional grid method.
Smart Images
Figure CN121453069A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application belongs to the technical field of navigation, and particularly relates to a star map recognition method based on adaptive grid. BACKGROUND
[0002] As a high-precision and high-reliability autonomous attitude measurement instrument taking stars as measurement objects, a star sensor has been widely applied to space loads such as space vehicles. When the star sensor just enters a working state or loses attitude, after imaging of the starry sky, the star sensor completes attitude measurement after star point extraction, centroid positioning, star map recognition and attitude solution. The main working principle is as follows: the starry sky is imaged by using an optical system and an image sensor, the position and brightness information of star point targets in a star map are obtained through star point extraction and centroid positioning, then the star point targets are matched with corresponding navigation stars in a star catalog through star map recognition, and finally the attitude of the star sensor relative to an inertial coordinate system is measured according to the recognition result.
[0003] Star map recognition is the most complex and time-consuming link in the working process of the star sensor. In the development process of the star map recognition algorithm, the grid method has been widely applied due to its low algorithm complexity and fast recognition speed. However, the grid method has the following shortcomings: when the star map is divided into grids, the number of rows and columns can only be set, and the number of rows and columns cannot be adjusted according to the number of star point targets in the star map. When the number of grids is too small, multiple star point targets may exist in some grids, and encoding of the grids will result in missing codes, which seriously affects the star map recognition rate when the star catalog grid codes are used for recognition. Conversely, when the number of grids is too large, the number of codes obtained by encoding the grids is huge, which seriously affects the star map recognition speed when the star catalog grid codes are used for recognition. Therefore, adaptive grid division of the star map and optimization of the traditional grid method are important key technologies in the field of star map recognition algorithm research. SUMMARY
[0004] Therefore, the present application aims to provide a star map recognition method based on adaptive grid, which solves the problem that the number of grids cannot be adaptively adjusted when the star map is recognized by using the traditional grid method.
[0005] To achieve the above object, the technical scheme of the present application is as follows: A star map recognition method based on adaptive grid, comprising: S1: respectively dividing the star map and the star catalog into grids according to the principle of quadtree; S2: respectively carrying out binary coding on the star map grid and the star catalog grid divided in step S1, and obtaining a star map feature vector and a star catalog feature vector correspondingly; S3: defining a similarity function, and calculating a similarity coefficient of the star map feature vector and the star catalog feature vector obtained in step S2 by using the similarity function; S4: obtaining a navigation star matched with the star point target in the star map in the star catalog based on the similarity coefficient obtained in step S3.
[0006] Further, the grid division operations on the star map and the star catalog in step S1 are consistent; in the process of carrying out grid division on the star map to obtain the star map grid: the star map is divided into a star map grid including four sub-grids of the same size; if the number of star point targets in the divided star map grid is still greater than 1, the sub-grid with the number of star point targets greater than 1 in the current star map grid is further divided according to the principle of quadtree, until the number of star point targets contained in each sub-grid in the star map grid is not greater than 1, at this time, the grid division on the star map is completed, and the final star map grid is obtained.
[0007] Further, the process of carrying out binary coding on the star map grid and the star catalog grid to obtain the corresponding feature vectors in step S2 is consistent; the star map grid is divided with the smallest sub-grid as a benchmark, and the divided star map grid is coded from left to right and from top to bottom, if a sub-grid of the star map grid contains a star point target, the corresponding coding of the sub-grid is 1, otherwise, the coding is 0, and the vector obtained by coding is the star map feature vector.
[0008] Further, the similarity function in step S3 is: ; Wherein, ξ represents the similarity coefficient, n represents the length of the star map feature vector or the star catalog feature vector, w i represents the i-th element in the star map feature vector, v i represents the i-th element in the star catalog feature vector.
[0009] Further, in step S4, a similarity threshold is set, if the similarity coefficient is less than the similarity threshold, the navigation star corresponding to the star catalog feature vector is the navigation star corresponding to the star point target in the star map in the star catalog.
[0010] Compared with the prior art, the application can achieve the following beneficial effects: The adaptive grid-based star map recognition method provided by the application comprises the following steps: performing adaptive grid division on a star map in a quadtree manner to obtain a star map feature vector and a star table feature vector; calculating a similarity coefficient of the star map feature vector and the star table feature vector by using a defined similarity function; and obtaining a navigation star in a star table that matches a star point target in the star map, thereby retaining the advantages of low algorithm complexity and fast recognition speed of the traditional grid method, ensuring the integrity of the code in the subsequent binary coding process due to the adaptive grid division of the star map and the star table in the quadtree manner, improving the star map recognition efficiency and accuracy, and effectively avoiding the technical problems of missing code in the traditional grid method, which seriously affects the star map recognition rate when the star table grid code is used for recognition, or a large number of code values, which seriously affects the star map recognition speed when the star table grid code is used for recognition. BRIEF DESCRIPTION OF DRAWINGS
[0011] The accompanying drawings, which form a part of this application, are included to provide a further understanding of the application and are incorporated in and constitute a part of this application. The embodiments of the application illustrated in the drawings are provided to explain the present application and are not meant to limit the application. Figure 1 A flowchart of the adaptive grid-based star map recognition method according to the embodiments of the application is shown in FIG. 2. Figure 2 A process diagram of the quadtree grid division of the star map according to the embodiments of the application is shown in FIG. 3. Figure 3 A diagram of the star map grid divided in step S2 according to the embodiments of the application is shown in FIG. 4. DETAILED DESCRIPTION
[0012] In order to make the objectives, technical solutions and advantages of the application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the application and do not limit the application.
[0013] It should be noted that the embodiments in the application and the features in the embodiments can be combined with each other without conflict.
[0014] In the description of the present application, it should be understood that the terms "center", "longitudinal", "transverse", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation of the present application. In addition, the terms "first", "second" and the like are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined with "first", "second" and the like can be explicitly or implicitly included one or more. In the description of the present application, unless otherwise specified, the meaning of "a plurality of" is two or more.
[0015] In the description of the present application, it should be noted that unless otherwise specified and limited, the terms "mounting", "connection", "connection" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be connected inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.
[0016] The present application will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
[0017] As Figure 1 shown, the star map recognition method based on adaptive grid in the embodiment of the present application comprises: S1: The star map and the star catalog are respectively divided into grids according to the principle of quadtree.
[0018] In some embodiments, the grid division operations of the star map and the star catalog are consistent; in the process of dividing the star map into grids, the star map is divided into a star map grid comprising four sub-grids of the same size; the sub-grid whose number of star point targets is still greater than 1 after division is further divided according to the principle of quadtree, until each sub-grid of the star map grid contains no more than 1 star point target, at which time the grid division of the star map is completed, and the final star map grid is obtained.
[0019] In the embodiment of the present application, the star map grid after the star map is divided into grids according to the principle of quadtree in step S1 is as shown in Figure 2 The process of dividing the star map into grids to obtain the star map grid specifically comprises: the size of the star map is MxN, if the number of star point targets in the original star map (i.e. Figure 2If the number of red stars in the star map is greater than 1, then the star map is divided into grids according to the quadtree principle. The specific method is as follows: divide the star map (such as...) Figure 2 (a) is divided into a star map grid consisting of four sub-grids, each with a scale of (M / 2)×(N / 2). It is found that the number of star targets in each of the four sub-grids is greater than 1, i.e., as shown in (a). Figure 2 As shown in (b), the four subgrids need to be further divided according to the quadtree principle. At this point, the scale of each subgrid in the star map grid is (M / 4) × (N / 4), as follows: Figure 2 As shown in (c), it is found that in the current star map grid, some subgrids have no more than 1 star target, while some subgrids have more than 1 star target (e.g., Figure 2 (The red box in (c)) requires subgrids with more than 1 star target (i.e., such as...). Figure 2 (c) The red box) continues to perform grid division according to the quadtree principle, such as Figure 2 As shown in (d) above, the smallest subgrid in the star map grid at this point has a size of (M / 8) × (N / 8), and the number of star targets in each subgrid does not exceed 1. The partitioning ends here, yielding the final star map grid. The method for obtaining the star table grid in this embodiment is the same as the method for obtaining the star map grid, and will not be repeated here.
[0020] S2: Perform binary encoding on the star map grid and star catalog grid after step S1 to obtain the star map feature vector and star catalog feature vector respectively.
[0021] In some embodiments, the process of binary encoding the star map grid and the star table grid to obtain the corresponding feature vector is the same. In the process of binary encoding the star map grid to obtain the star map feature vector: the star map grid is divided based on the smallest sub-grid in the star map grid. The star map grid divided at this time is binary encoded from left to right and from top to bottom. If a sub-grid of the star map grid contains a star point target, the corresponding sub-grid is encoded as 1, otherwise it is encoded as 0. The encoded vector is the star map feature vector.
[0022] In this embodiment of the invention, the smallest subgrid in the star map grid has a scale of (M / 8) × (N / 8). Therefore, the star map grid is divided with a scale of (M / 8) × (N / 8), as follows: Figure 3 As shown by the dashed lines, the area is divided into 8×8=64 sub-grids. These 64 sub-grids are then binary encoded. Specifically, sub-grids containing star targets are encoded as 1, and sub-grids without star targets are encoded as 0. The encodings are collected from left to right and top to bottom to obtain the corresponding star map feature vectors. The method for obtaining the star catalog feature vectors in this embodiment is the same as the method for obtaining the star map feature vectors, and will not be repeated here.
[0023] S3: defining a similarity function, and calculating a similarity coefficient of the star map feature vector and the catalog feature vector by using the similarity function.
[0024] In some embodiments, the similarity function in step S3 is: ; wherein ξ represents the similarity coefficient, n represents the length of the star map feature vector or the catalog feature vector, w i represents the i-th element in the star map feature vector, and v i represents the i-th element in the catalog feature vector. The smaller the similarity coefficient ξ is, the more similar the star map feature vector and the catalog feature vector are.
[0025] S4: obtaining the navigation star matching the star point target in the star map from the catalog based on the similarity coefficient obtained in step S3.
[0026] In some embodiments, step S4 comprises: setting a similarity threshold, and if the similarity coefficient is smaller than the similarity threshold, the navigation star corresponding to the catalog feature vector is the navigation star corresponding to the star point target in the star map in the catalog. The similarity threshold is adaptively adjusted according to actual conditions.
[0027] It should be understood that the steps can be reordered, added or deleted using the various forms of flowcharts shown above. For example, the steps described in the present disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions of the present disclosure can be achieved, which is not limited herein.
[0028] The above detailed description does not constitute a limitation on the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements within the spirit and principles of the present application shall be included in the protection scope of the present application.
Claims
1. A star map recognition method based on adaptive grids, characterized in that, include: S1: Divide the star map and star catalog into grids according to the quadtree principle; S2: Perform binary encoding on the star map grid and star catalog grid after step S1 to obtain the star map feature vector and star catalog feature vector respectively; S3: Define a similarity function and use the similarity function to calculate the similarity coefficient between the star map feature vector and the star catalog feature vector obtained in step S2; S4: Based on the similarity coefficient obtained in step S3, obtain the navigation star that matches the star point target in the star table.
2. The star map recognition method based on adaptive grid according to claim 1, characterized in that, The meshing operation for the star map and star catalog in step S1 is the same. In the process of meshing the star map to obtain the star map mesh: the star map is divided into a star map mesh containing four sub-mesh of the same size; if the number of star targets in the divided star map mesh is still greater than 1, then continue to divide the sub-mesh of the current star map mesh with the number of star targets greater than 1 according to the quadtree principle, until the number of star targets in each sub-mesh of the star map mesh is no greater than 1. At this time, the meshing of the star map is completed, and the final star map mesh is obtained.
3. The star map recognition method based on adaptive grid according to claim 2, characterized in that, The process of binary encoding the star map grid and star table grid in step S2 to obtain the corresponding feature vector is the same. In the process of binary encoding the star map grid to obtain the star map feature vector: the star map grid is divided based on the smallest sub-grid in the star map grid. The star map grid divided at this time is binary encoded from left to right and from top to bottom. If a sub-grid of the star map grid contains star point targets, the corresponding sub-grid is encoded as 1, otherwise it is encoded as 0. The encoded vector is the star map feature vector.
4. The star map recognition method based on adaptive grid according to claim 1, characterized in that, The similarity function in step S3 is: ； Where ξ represents the similarity coefficient, n represents the length of the star map feature vector or star catalog feature vector, and w i v represents the i-th element in the star map eigenvector. i This represents the i-th element in the eigenvector of the star catalog.
5. The star map recognition method based on adaptive grid according to claim 1, characterized in that, In step S4, a similarity threshold is set. If the similarity coefficient is less than the similarity threshold, then the navigation star corresponding to the star feature vector in the star table is the navigation star corresponding to the star point target in the star table.