Photogrammetry target and positioning method facing on-orbit strong solar stray light environment

Abstract

The application discloses a photogrammetry target and positioning method for an on-orbit strong solar stray light environment, and belongs to the technical field of satellite antenna on-orbit shape measurement. The target comprises a central glass microsphere directional back reflection area and a low reflectivity edge area arranged at the periphery of the central glass microsphere directional back reflection area, and the low reflectivity edge area is used for forming a local low response transition zone between the central back reflection area and a strong stray light background. An original image containing the target is collected; a top hat transformation is performed on the original image to obtain an enhanced image; edge detection is performed on the enhanced image, and candidate areas are screened according to the geometric features of the target to determine a target effective area; the gray values of the corresponding areas in the original image are recalled, and a gray weighted centroid method is used to calculate the target center coordinates. The application solves the problem that the target edge cannot be identified in the strong solar stray light environment, and improves the stability and precision of the target center positioning.

Description

Technical Field

[0001] This invention belongs to the field of satellite antenna on-orbit surface measurement technology, specifically relating to photogrammetric targets and positioning methods for strong solar stray light environments on-orbit. Background Technology

[0002] Large deployable space antennas are widely used in high-resolution Earth observation, space communication, meteorological remote sensing, and deep space exploration missions. These antennas are typically folded and stowed during the launch phase, then unfolded into a large-aperture working surface after entering orbit. Affected by factors such as thermal loads, residual stress after unfolding, attitude disturbances, and changes in the space environment, the antenna may experience surface deformation or configuration shifts during on-orbit operation, thus affecting antenna gain, beam pointing, and payload performance. Therefore, on-orbit surface measurement of large deployable space antennas is of significant engineering importance.

[0003] Close-range photogrammetry offers advantages such as non-contact, full-field measurement, and high precision, making it suitable for 3D measurement of large space structures. Since satellite antenna surfaces typically lack stable and identifiable natural textures, engineering practice generally requires deploying artificial collaborative targets on the antenna surface. Multiple cameras acquire images of the targets, and the target center coordinates are extracted for spatial intersection and 3D reconstruction. Among these, circular reflective targets can generate strong echo signals under active camera illumination, and are characterized by simple structure, easy identification, and high positioning accuracy, thus frequently used in close-range photogrammetry tasks.

[0004] In existing engineering projects, photogrammetric target center localization schemes typically include the following steps: A measuring camera acquires an image of the surface of the structure under active illumination, causing the reflected light target to form a bright area in the image; the acquired image undergoes preprocessing such as grayscale conversion, filtering, thresholding, or background suppression; candidate target regions are selected based on features such as connected region area, roundness, aspect ratio, and edge closure; centroid extraction is performed on the target region image, mainly including methods based on target edge geometric features (by extracting the target contour and fitting a circle, ellipse, or polynomial curve, using the center of the fitted shape as the centroid of the target image) and methods based on regional grayscale distribution features (such as grayscale weighted centroid method and grayscale square weighted centroid method); the extracted target center coordinates are used as image point observations for subsequent multi-view intersection, camera orientation, and 3D reconstruction.

[0005] However, the above methods have significant limitations when applied to satellites in orbit under strong solar stray light conditions. During satellite operation, sunlight illuminates the antenna surface as the satellite's orbital position and attitude change. After reflection and scattering by the antenna reflector, support structure, or other components, it enters the camera's field of view, forming a large area of ​​bright background. Strong solar stray light significantly increases the grayscale of the background around the target, compressing the local grayscale difference on both sides of the target boundary. In areas of strong reflection, it may also cause pixel saturation or edge defects. After the camera point spread function weakens the edge sharpness, the edge discernibility of traditional circular reflective targets decreases significantly. Under these conditions, traditional threshold segmentation, edge detection, and connected component screening methods are prone to problems such as target region defects, background mixing, boundary adhesion, or misjudgment of candidate regions. For the grayscale weighted centroid method, its positioning results are highly sensitive to the target region involved in the calculation. If a bright background is mixed into the extracted region, or if there are defects in the actual boundary, it will cause a systematic shift in the centroid coordinates, thus affecting the accuracy of subsequent photogrammetric 3D solution.

[0006] Therefore, there is an urgent need for a solution that can be designed collaboratively from both the physical structure of the target and the image localization process: on the one hand, enhance the local gray-scale transition at the edge of the target structure to improve the recognizability of the target boundary against a strong background; on the other hand, use background suppression methods to stably extract the target region during image processing and return to the original image to calculate the centroid, so as to ensure the reliability and physical consistency of the center localization result. Summary of the Invention

[0007] To address the aforementioned technical problems, this invention provides a photogrammetric target and positioning method for in-orbit strong solar stray light environments. By enhancing the boundary grayscale transition through a low-reflection edge target, and after determining the target area by combining top-hat transformation, the effective area of ​​the target is used as a mask. The corresponding pixel grayscale values ​​in the original image are called to perform grayscale weighted centroid calculation, thereby solving the problem of unidentifiable target edges under strong solar stray light.

[0008] To achieve the above objectives, the present invention adopts the following technical solution:

[0009] Photogrammetric targets for use in strong solar stray light environments in orbit include:

[0010] Polyimide substrate;

[0011] A central glass microsphere directional backlight reflection region is provided at the center of the front side of the polyimide substrate, and a low reflectivity edge region is provided around the central glass microsphere directional backlight reflection region;

[0012] The central glass microsphere directional backlight reflection region is used to generate a high-brightness echo signal under near-coaxial lighting conditions of the camera; the low-reflectivity edge region is used to form a local low-response transition zone between the central glass microsphere directional backlight reflection region and the strong stray light background, so as to enhance the grayscale transition characteristics at the target boundary.

[0013] Furthermore, the target is a circular thin-film structure, and the central glass microbead directional backlight reflection area is arranged concentrically with the polyimide substrate; the low reflectivity edge area is formed by the outer ring area of ​​the polyimide substrate where the central glass microbead directional backlight reflection area is not provided.

[0014] Furthermore, the central glass microsphere directional backlight reflection area is composed of a glass microsphere-type backlight reflection coating, in which high-refractive-index glass microspheres are distributed, so that the incident light returns along a direction close to the incident direction after refraction and reflection.

[0015] Furthermore, an adhesive layer is provided on the back side of the polyimide substrate for fixing the target to the surface of the satellite antenna and for making the front side of the target face the photogrammetry camera.

[0016] Furthermore, the polyimide substrate has a diameter of 16 mm, the central glass microsphere directional reflective region has a diameter of 10 mm, the low reflectivity edge region has a radial width of 3 mm, the target has a total thickness of 0.23 mm ± 0.05 mm, and the coating thickness of the central glass microsphere directional reflective region is 0.10 mm ± 0.01 mm.

[0017] Furthermore, the overall imaging diameter of the target on the image plane is equal to the sum of the imaging diameter of the directional retroreflection region of the central glass microsphere and twice the radial pixel width of the low reflectivity edge region. The imaging diameter of the directional retroreflection region of the central glass microsphere on the image plane is determined by the camera focal length, working distance, pixel size, and the actual diameter of the corresponding region. The radial pixel width of the low reflectivity edge region on the image plane is determined by the camera focal length, working distance, pixel size, and the actual radial width of the corresponding region.

[0018] On the other hand, the present invention provides a method for locating a photogrammetric target based on the aforementioned method, comprising:

[0019] Acquire raw images containing the photogrammetric target;

[0020] The original image is subjected to a top-hat transform to obtain an enhanced image with background suppression;

[0021] Edge detection is performed on the enhanced image, and candidate regions are selected based on the geometric features of the photogrammetric target to determine the effective target region;

[0022] Using the effective target region as a region mask, the gray values ​​of the pixels in the original image corresponding to the spatial position of the effective target region are called, and the target center coordinates are calculated using the gray-weighted centroid method.

[0023] Furthermore, the equivalent diameter of the structuring element used in the top-hat transformation is greater than or equal to the overall imaging diameter of the photogrammetric target in the image.

[0024] Furthermore, the equivalent diameter of the structural element is 1.0 to 1.5 times the overall imaging diameter of the target.

[0025] Furthermore, when performing edge detection on the enhanced image, the Canny operator is used to extract candidate edge regions of the target.

[0026] The beneficial effects of this invention are as follows:

[0027] First, this invention artificially constructs a local grayscale transition zone at the target's physical structure level by setting a directional backlight reflection region in the central glass microsphere and its surrounding low-reflectivity edge region. This low-reflectivity edge region forms a low-response transition zone between the bright background formed by strong solar stray light and the backlight reflection center, ensuring that the target boundary retains detectable grayscale gradient characteristics even after the point spread function becomes blurred. This fundamentally solves the problem of decreased target edge discernibility under strong stray light conditions.

[0028] Second, this invention employs top-hat transformation to suppress the background of the original image, effectively weakening the gradually bright background formed by strong solar stray light while preserving the local bright target features. Based on this, edge detection and region selection are performed, significantly improving the stability and accuracy of effective target region extraction.

[0029] Third, after determining the effective target area, this invention uses the effective target area as a region mask and calls the grayscale values ​​of the pixels in the original image corresponding to the spatial position of that region to perform grayscale weighted centroid calculation. Since the top-hat transform is a non-linear morphological process that will change the original grayscale distribution, this invention only uses the enhanced image to determine the effective target area, while using the corresponding pixel grayscale values ​​of the original image in the centroid calculation stage. This avoids the positioning error introduced by grayscale distortion in the enhanced image and ensures the physical consistency and sub-pixel accuracy of the center coordinates.

[0030] Fourth, this invention integrates a target structure with a low reflectivity edge region with top-hat transformation background suppression and original image grayscale weighted centroid positioning in a collaborative design, forming a complete technical solution from hardware to algorithm, which is particularly suitable for antenna surface photogrammetry tasks in satellite on-orbit strong solar stray light environment. Attached Figure Description

[0031] Figure 1 This is a structural diagram of the photogrammetric target for the present invention in an on-orbit environment with strong solar stray light.

[0032] Figure 2 A schematic diagram illustrating the grayscale distribution principle of a typical backlight reflection target and the low reflectivity edge target of this invention under a strong solar stray light background;

[0033] Figure 3 To simulate the actual image of the low reflectivity edge photogrammetry target of the present invention under the background of strong solar stray light in orbit and the image after top-hat transformation, (a) represents the actual image under the background of strong stray light, and (b) represents the image after suppressing the stray light background by top-hat transformation.

[0034] Figure 4 The image shows a comparison of target edge extraction results under strong solar stray light background in orbit. In the image, (a) represents the result of edge extraction after the original real image is processed by the Canny operator, and (b) represents the result of edge extraction after the stray light background is suppressed by the top-hat transform and then processed by the Canny operator.

[0035] Figure label:

[0036] 1-Directional backlight reflection region of glass microspheres, 2-Low reflectivity edge region. Detailed Implementation

[0037] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0038] This invention proposes a photogrammetric target and positioning method for use in on-orbit environments with strong solar stray light. This method can be applied to on-orbit photogrammetric systems for satellite antennas, as well as to spacecraft structural component measurement, large deployable space structures measurement, and other scenarios involving artificial target identification under strong stray light interference.

[0039] like Figure 1As shown, the photogrammetric target of the present invention has a circular thin-film structure, including a polyimide substrate, a glass microsphere directional backlight reflection region 1 disposed on the front side of the polyimide substrate, a low reflectivity edge region 2 located around the glass microsphere directional backlight reflection region, and an adhesive layer disposed on the back side of the polyimide substrate. The polyimide substrate is circular. The glass microsphere directional backlight reflection region 1 is formed in the central region of the front side of the polyimide substrate by spraying a glass microsphere-type reflective coating, and is concentrically arranged with the circular substrate. The glass microsphere-type backlight reflection coating contains high-refractive-index glass microspheres, used to generate a directional backlight reflection high-brightness signal under near-coaxial lighting conditions of the camera (e.g., causing incident light to return along a direction close to the incident direction after refraction and reflection). The low reflectivity edge region 2 is formed by the outer ring region of the front side of the polyimide substrate where no glass microspheres are sprayed.

[0040] like Figure 2 The diagram shows the grayscale distribution principle of a normal backlight reflection target and the low reflectivity edge target of the present invention under a strong solar stray light background, which is used to illustrate the enhancement effect of the low reflectivity edge region on the grayscale transition of the target boundary.

[0041] For ordinary retroreflective targets, under strong solar stray light, the grayscale of the outer background is significantly increased, the grayscale difference between the central retroreflective area and the surrounding bright background is compressed, and the grayscale transition at the target boundary is weakened, easily manifesting as a local continuous high grayscale distribution, thus making it difficult to stably identify the target boundary. In contrast, this invention sets a low-reflectivity edge region around the directional retroreflective area of ​​the central glass microsphere, forming a local low-response transition zone between the central bright retroreflective area and the outer bright background. In terms of grayscale distribution, it exhibits a structure of "bright center - low-response edge - bright background", thereby enhancing the grayscale transition characteristics at the target boundary and improving the detectability of the target outline under strong solar stray light.

[0042] In a preferred embodiment, the diameter and radial width of each part of the target are determined based on the results of subsequent formula derivation. The design principle is to minimize its impact on the antenna surface structure and electrical performance while meeting the imaging pixel width requirements of the experimental system to obtain the minimum design diameter. Different diameters can be calculated based on different system requirements in different systems. In this experimental system, the diameter of the polyimide substrate is 16 mm, the glass microsphere directional backlight reflection area 1 is a circular sprayed area with a diameter of 10 mm, and the low reflectivity edge area 2 is an annular area surrounding the glass microsphere directional backlight reflection area 1 with a radial width of approximately 3 mm. The total thickness of the target is 0.23 mm ± 0.05 mm, and the thickness of the coating in the glass microsphere directional backlight reflection area 1 is 0.10 mm ± 0.01 mm. The adhesive layer is disposed on the back of the polyimide substrate and is used to fix the target to the satellite antenna surface according to the preset measurement points, with the front of the target facing the photogrammetry camera.

[0043] The glass microsphere directional backlight reflection region 1 is used to form a bright echo signal under near-coaxial lighting conditions of the camera; the low reflectivity edge region 2 is a black polyimide ring region, which is used to form a local low-response transition zone between the central bright backlight reflection region and the strong stray light background, thereby enhancing the gray-level transition characteristics at the target boundary, reducing the possibility of gray-level adhesion between the strong stray light background and the central backlight reflection region in the image, and improving the stability of target edge detection, region segmentation and gray-level weighted centroid localization.

[0044] Because circular targets in industrial photogrammetry offer advantages such as orientation independence, regular image projection shape, stable boundary extraction, and high centroid positioning accuracy, this invention preferably employs a circular target structure. By concentrically arranging the central circular backlight reflection area and the outer annular low-reflectivity edge area, the detectability of the target's boundary under strong stray light backgrounds can be enhanced while maintaining target recognition stability.

[0045] Furthermore, the target dimensions include the diameter of the central glass microbead directional retroreflection region 1, the outer diameter of the polyimide substrate, and the radial width of the low-reflectivity edge region 2. The diameter of the central glass microbead directional retroreflection region 1 is mainly determined based on camera imaging parameters and target positioning requirements, ensuring that the retroreflection region has sufficient pixel coverage in the image to meet the needs of target recognition, edge extraction, and grayscale weighted centroid positioning. The outer diameter of the polyimide substrate is minimized while meeting imaging positioning requirements to reduce the impact on the antenna's electrical performance and surface condition after the target is attached to the satellite antenna surface.

[0046] Let the actual diameter of the directional retroreflection region 1 of the central glass microsphere be... The outer diameter of the polyimide substrate is The radial width of the low reflectivity edge region 2 is Then all three conditions are met:

[0047] ,

[0048] Let the working distance of the camera be... focal length is The pixel size is The approximate imaging diameter of the directional retroreflection region 1 of the central glass microsphere on the image plane is... It can be represented as:

[0049] ,

[0050] Approximate radial pixel width of low reflectivity edge region 2 on the image plane It can be represented as:

[0051] ,

[0052] in, It should meet the minimum pixel coverage required for target recognition, edge extraction, and grayscale weighted centroid localization; The low-reflectivity edge region 2 should be distinguishable in the image, forming a detectable low-response transition band between the central bright backlight reflection region and the external strong stray light background. Therefore, the imaging size requirement of the central backlight reflection region can be determined first based on the localization algorithm's requirements. Then, based on the distinguishability requirements of the black low-reflectivity edge region 2, determine... The final outer diameter of the polyimide substrate is obtained as follows:

[0053] ,

[0054] Under the imaging geometry of the photogrammetry system, the 10 mm diameter oriented retroreflection region 1 of the central glass microsphere corresponds to an imaging diameter of approximately 10 pixels in the image, which meets the pixel coverage requirements for high-precision target identification, orientation, and centroid positioning in photogrammetry. The low-reflectivity black annular edge region with a radial width of approximately 3 mm corresponds to an annular low-response region of approximately 2-3 pixels in the image. It can be stably resolved under conditions permissible by the camera's imaging quality, and forms a clear grayscale transition band between the central bright retroreflection region and the external strong stray light background. Therefore, the low-reflectivity edge region 2 enhances the grayscale transition characteristics at the target boundary, which is beneficial for accurate target edge extraction, reliable target region identification, and precise target center positioning.

[0055] The above-mentioned size combination meets the requirements for target imaging recognition and positioning accuracy while minimizing the overall attachment area of ​​the target, thereby helping to reduce the impact of the target on the electrical performance, thermal control performance and structural surface condition of the satellite antenna surface.

[0056] Based on the aforementioned target design, this invention also provides a method for locating a retroreflection target based on top-hat background suppression and gray-level weighted centroid, specifically including:

[0057] Acquire the original image I containing the aforementioned target.

[0058] A top-hat transform is applied to the original image I to obtain an enhanced image with background suppression, which is used to weaken the gradually bright background formed by strong solar stray light. Let the structuring element of the original image be... Morphological opening operation is The enhanced image after top-hat transformation It can be represented as:

[0059] ,

[0060] Among them, structural elements The scale should be set according to the target imaging size. Structural elements equivalent diameter It should be greater than or equal to the overall imaging diameter of the target. ,in:

[0061] ,

[0062] Preferred, It is 1.0 to 1.5 times the overall imaging diameter of the target, that is... This is to effectively estimate the gradually changing background and preserve the local bright target features of the target.

[0063] Enhanced image after top-hat transformation Edge detection is performed using the Canny operator to extract candidate edge regions of the target. Since the target comprises a central glass microsphere directional backlight reflection region 1 and a peripheral low-reflectivity edge region 2, under strong solar stray light conditions, multi-level grayscale variations will occur between the central backlight reflection region, the low-reflectivity edge region, and the external bright background. Without background suppression, additional edge responses may occur between the external bright background and the low-reflectivity edge region, potentially causing adjacency or adhesion to the effective edge of the central backlight reflection region.

[0064] After weakening the gradually varying brightness of the background through top-hat transformation, the edge response of the peripheral background interference is reduced, and the effective edges of the central reflective region become more prominent. Based on this enhanced image, the Canny operator is used for edge detection, which can obtain more stable candidate edge regions. Subsequently, the effective target region is selected based on the geometric and grayscale features of the candidate edge regions.

[0065] Therefore, in the enhanced image The Canny operator is used for edge detection, which can reduce background interference edges while preserving the local boundary features of the target, thereby improving the effectiveness of candidate edge regions.

[0066] Then, candidate edge regions are filtered based on the geometric and grayscale features of the target to determine the effective target region. The filtering criteria include indicators such as the area range of connected regions, roundness, edge closure, and regional gray-scale statistical characteristics.

[0067] Specifically, firstly, based on the camera projection geometry and considering the working distance, focal length, pixel size, and actual target size of the specific photogrammetric imaging system, the expected imaging size of the target in the image is calculated. Then, based on this expected imaging size, the areas of candidate connected regions are initially screened, eliminating noise regions with significantly smaller areas and background interference regions with significantly larger areas. Secondly, based on the roundness and edge closure of the candidate regions, candidate regions with near-circular shapes and good boundary continuity are selected, with a roundness of not less than 0.8. Finally, combining the mean gray level, maximum gray level, and gray level contrast between the candidate regions and the surrounding background regions, candidate regions with high-brightness backlight reflection characteristics are retained and determined as the effective target areas. .

[0068] Since top-hat transformation is a non-linear morphological process, it enhances images. The grayscale distribution no longer strictly maintains the original radiation relationship, therefore this invention does not directly use the top cap to enhance the image. Calculate the final centroid.

[0069] In determining the effective target area Then, the effective area of ​​the target is... As a region mask, extract the original image. The grayscale values ​​of the pixels corresponding to the spatial location of this region are used to calculate the target center coordinates using the grayscale weighted centroid method.

[0070] , ,

[0071] in, The final calculated target centroid coordinates, pixels in the original image grayscale value at that location The effective target region is determined after top-cap enhancement, edge extraction, and region filtering.

[0072] After completing the above steps, it is possible to achieve close-range photogrammetry target structure, target extraction, and center positioning for monitoring satellite antenna surface deformation under high-brightness background conditions of on-orbit stray solar light.

[0073] To verify the effectiveness of the above structural design and positioning method, in one embodiment, a ground-based simulated solar stray light experiment was conducted using a satellite antenna four-eye photogrammetry system. In the experiment, a circular reflective target was placed on the surface of the satellite antenna's unfolded lobe, and four measuring cameras were used to observe the target on the antenna surface from multiple perspectives. The measuring cameras had a field of view of 60° and an image resolution of [missing information]. The system features a pixel count, a focal length of approximately 16.5 mm, a pixel size of 4.5 μm, a depth of field of 3–5 m, and a working distance of approximately 4 m. A near-coaxial supplemental lighting device is configured to enhance the imaging brightness of the retroreflective target. Simultaneously, a standard solar simulator is used to illuminate the antenna and target area to simulate strong background light and local stray light interference caused by solar irradiation during satellite orbit. The experimental system parameters described above are consistent with the configuration in the document. For ordinary retroreflective targets without low-reflectivity edges, under strong solar stray light, the grayscale of the outer background is significantly increased, and the grayscale difference between the central retroreflective area and the surrounding bright background is compressed, easily appearing as a localized continuous high-grayscale area in the image. In this case, the grayscale transition characteristics between the target boundary and the background are weakened, and the target outline easily merges with the bright background, making it difficult to stably identify the target boundary.

[0074] The original image of the photogrammetric target with low reflectivity edges proposed in this invention under strong background light conditions, and the image after top-hat transformation processing, are shown below. Figure 3 As shown, (a) represents a real-world image against a strong stray light background, and (b) represents an image after top-hat transformation to suppress the stray light background. Figure 3 It can be seen that even under bright background illumination, the back-reflection target with low reflectivity edges still retains a relatively obvious grayscale change near the boundary, and the target outline can be effectively identified. This is because the low reflectivity edge region forms a local low-response transition zone between the central bright back-reflection region and the external strong stray light background, thereby enhancing the grayscale transition characteristics near the target boundary.

[0075] Furthermore, the top-hat transformation can suppress the gradually brightening background caused by stray solar light, making the grayscale difference between the target area and the surrounding background more obvious, thereby further improving the discernibility of target features. Thus, low-reflectivity edges do not simply change the overall brightness of the target, but rather enhance the local grayscale transition between the reflective area and the surrounding background at the level of the target's physical structure. Combined with the top-hat transformation background suppression processing, the detectability of the target edge area under strong background interference can be improved, providing a more stable image foundation for subsequent effective target area extraction and centroid localization.

[0076] To further analyze the impact of stray light background on target region extraction, this embodiment also compares the edge extraction results of the target affected by stray light interference before and after the top-hat transformation, such as... Figure 4 As shown, (a) represents the result of edge extraction from the original real-shot image using the Canny operator, and (b) represents the result of edge extraction using the Canny operator after suppressing stray light background through top-hat transformation. Before top-hat transformation, due to the abrupt grayscale change between the bright background and the low-reflectivity edge region, additional edge responses caused by background interference may occur around the target. These additional edges may be adjacent to, adhere to, or partially overlap with the effective edge of the target in local areas, complicating the candidate edge structure and thus affecting the accurate determination of the effective target region. After top-hat transformation, the gradually changing bright background formed by strong solar stray light is suppressed, the edge response of the peripheral background interference is weakened, and the separation between the effective edge of the target and the external interference edge is improved, which is beneficial for the stable screening of effective pixels inside the target and the grayscale weighted centroid localization.

[0077] Under the aforementioned experimental conditions, 150 images were acquired under different sunlight incidence directions. Experimental results show that targets with low-reflectivity edges and reflective surfaces can be stably identified with a 100% success rate. Further positioning statistics indicate that after top-hat transformation preprocessing combined with gray-scale weighted centroid positioning, the average planar positioning deviation of the target is approximately 0.06 pixels. These results demonstrate that the low-reflectivity edge target and corresponding positioning method described in this invention can improve the stability of target edge recognition, effective region extraction, and center positioning under simulated strong solar stray light conditions in orbit, providing a stable and reliable basis for image point observations in satellite antenna in-orbit photogrammetry.

[0078] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. Photogrammetric target oriented to in-orbit strong solar stray light environment, characterized in that, include: Polyimide substrate; A central glass microsphere directional backlight reflection region is provided at the center of the front side of the polyimide substrate, and a low reflectivity edge region is provided around the central glass microsphere directional backlight reflection region; The central glass microsphere directional backlight reflection region is used to generate a high-brightness echo signal under near-coaxial lighting conditions of the camera; the low-reflectivity edge region is used to form a local low-response transition zone between the central glass microsphere directional backlight reflection region and the strong stray light background, so as to enhance the grayscale transition characteristics at the target boundary.

2. Photogrammetric target facing in-orbit high solar stray light environment according to claim 1, characterized in that, The target is a circular thin-film structure, and the central glass microsphere directional backlight reflection area is arranged concentrically with the polyimide substrate; the low reflectivity edge area is formed by the outer ring area of ​​the polyimide substrate where the central glass microsphere directional backlight reflection area is not provided.

3. The photogrammetric target for on-orbit strong solar stray light environment according to claim 2, characterized in that, The central glass microsphere directional backlight reflection area is composed of a glass microsphere-type backlight reflection coating, in which high-refractive-index glass microspheres are distributed, so that the incident light returns along a direction close to the incident direction after refraction and reflection.

4. The photogrammetric target for on-orbit strong solar stray light environment according to claim 2, characterized in that, An adhesive layer is provided on the back of the polyimide substrate for fixing the target to the surface of the satellite antenna and for making the front of the target face the photogrammetry camera.

5. The photogrammetric target for on-orbit strong solar stray light environment according to claim 2, characterized in that, The polyimide substrate has a diameter of 16 mm, the central glass microsphere directional reflective region has a diameter of 10 mm, and the radial width of the low reflectivity edge region is 3 mm; the total thickness of the target is 0.23 mm ± 0.05 mm, and the coating thickness of the central glass microsphere directional reflective region is 0.10 mm ± 0.01 mm.

6. The photogrammetric target for on-orbit strong solar stray light environment according to claim 2, characterized in that, The overall imaging diameter of the target on the image plane is equal to the sum of the imaging diameter of the directional retroreflection region of the central glass microsphere and twice the radial pixel width of the low reflectivity edge region. The imaging diameter of the directional retroreflection region of the central glass microsphere on the image plane is determined by the camera focal length, working distance, pixel size, and the actual diameter of the corresponding region. The radial pixel width of the low reflectivity edge region on the image plane is determined by the camera focal length, working distance, pixel size, and the actual radial width of the corresponding region.

7. A method for locating a photogrammetric target based on any one of claims 1 to 6, characterized in that, include: Acquire raw images containing the photogrammetric target; The original image is subjected to a top-hat transform to obtain an enhanced image with background suppression; Edge detection is performed on the enhanced image, and candidate regions are selected based on the geometric features of the photogrammetric target to determine the effective target region; Using the effective target region as a region mask, the gray values ​​of the pixels in the original image corresponding to the spatial position of the effective target region are called, and the target center coordinates are calculated using the gray-weighted centroid method.

8. The positioning method according to claim 7, characterized in that, The equivalent diameter of the structuring element used in the top-hat transformation is greater than or equal to the overall imaging diameter of the photogrammetric target in the image.

9. The positioning method according to claim 8, characterized in that, The equivalent diameter of the structural element is 1.0 to 1.5 times the overall imaging diameter of the target.

10. The positioning method according to claim 7, characterized in that, When performing edge detection on the enhanced image, the Canny operator is used to extract candidate edge regions of the target.

Images