Edge-enhanced multi-exposure space target three-dimensional reconstruction method, device and equipment

Abstract

The invention relates to an edge-enhanced multi-exposure space target three-dimensional reconstruction method, device and equipment, and the method comprises the steps: randomly selecting an optical image from data sets with different exposure levels as a group of to-be-reconstructed data, and employing a 3DGS method to carry out the reconstruction of the to-be-reconstructed data under the randomly initialized three-dimensional Gaussian distribution, reconstructing the space target according to the group of data to obtain three-dimensional Gaussian data, carrying out two-dimensional projection according to different poses of the space target in the data to be reconstructed to obtain a plurality of corresponding rendering images, and respectively extracting the plurality of rendering images and high-frequency feature images of the corresponding optical images by adopting a Sobel operator; the method comprises the following steps: constructing a three-dimensional Gaussian distribution model, calculating to obtain a structure sensing gradient loss function, optimizing the three-dimensional Gaussian distribution in combination with a multi-exposure joint optimization strategy, and performing three-dimensional reconstruction based on the finally optimized three-dimensional Gaussian distribution so as to overcome the problem of uneven illumination of a space target image and obtain an accurate three-dimensional reconstruction model of a space target.

Description

Technical Field

[0001] The present application relates to the technical field of computer vision three-dimensional reconstruction, and in particular to a method, device and equipment for three-dimensional reconstruction of space targets with edge enhancement and multiple exposures. Background Art

[0002] In today's aerospace field, with the rapid advancement of science and technology, many spacecraft have been successfully deployed into space orbits, greatly promoting human exploration and utilization of the universe. However, long-term on-orbit operation has put some satellites at risk of failure, or even out of control and becoming space debris. These debris seriously threaten the safety of other spacecraft and pose a severe challenge to the sustainability of the space environment. On-orbit maintenance technology has emerged as the times require, becoming a key means to ensure space safety and achieve effective utilization of space resources. In this technical system, accurately describing the current geometry of the target spacecraft is the basis and prerequisite for effective maintenance. However, due to the complex and changeable cosmic lighting conditions, images taken by satellite observation platforms often have uneven lighting, which brings great difficulties to the description of the geometry of the target spacecraft. To solve this problem, exposure synthesis technology is widely used in on-orbit rendezvous observations. Through multiple shots with different exposure times, comprehensive information about the target spacecraft can be obtained as much as possible.

[0003] Although exposure synthesis technology has alleviated the problem of uneven illumination to a certain extent, existing spatial target processing technologies still have many limitations. Traditional 3D reconstruction methods based on the Structure from Motion (SfM) framework rely on feature point (such as SIFT) extraction and camera parameter estimation to complete reconstruction. However, in multi-exposure observation scenarios, the luminosity changes of the same component in different images make feature point extraction and matching more difficult, and the reconstruction quality is difficult to guarantee, and may even lead to reconstruction failure. Although the neural radiance field (NeRF) in deep learning implicitly represents the scene through a neural network and achieves high-quality image rendering, the training and rendering time is too long, and it can only be optimized for a single image at a time, which is prone to overfitting and is greatly limited in practical applications.

[0004] Although some studies have applied it to the field of space targets and verified its feasibility in space target reconstruction, no in-depth and systematic improvements have been made for special imaging scenarios such as multi-exposure images. Summary of the invention

[0005] Based on this, it is necessary to provide a method, device and equipment for three-dimensional reconstruction of space targets with edge enhancement and multiple exposures, which can overcome the problem of uneven illumination of space target images and improve the reconstruction quality, in order to solve the above-mentioned technical problems.

[0006] A method for three-dimensional reconstruction of a space target by edge enhancement and multiple exposures, the method comprising: Acquire multiple optical images of the same space target at different exposure levels and multiple poses, and form a data set of optical images at the same exposure level to obtain multiple data sets; Selecting an optical image from each of the data sets as a group of data to be reconstructed, wherein the spatial target in each optical image in the data to be reconstructed presents a different posture; The space target is reconstructed according to a set of data to be reconstructed under a randomly initialized three-dimensional Gaussian distribution by using a 3DGS method to obtain three-dimensional Gaussian data; Based on the three-dimensional Gaussian data, two-dimensional projection is performed according to different postures of the space target in the data to be reconstructed to obtain a corresponding plurality of rendered images, and a Sobel operator is used to respectively extract high-frequency feature images of the plurality of rendered images and the corresponding optical images; Using the high-frequency feature image corresponding to the rendered image and the optical image, a structure-aware gradient loss function is calculated, and the three-dimensional Gaussian distribution is optimized in combination with a multi-exposure joint optimization strategy, and an optimization iteration is completed. After the three-dimensional Gaussian distribution is iteratively optimized for a preset number of times, a final optimized three-dimensional Gaussian distribution is obtained; The three-dimensional Gaussian distribution after the final optimization is converted into a point cloud for three-dimensional reconstruction to obtain a three-dimensional reconstruction model of the space target.

[0007] In one embodiment, the multiple optical images of the same space target include three exposure levels: high, medium, and low.

[0008] In one embodiment, when obtaining a plurality of two-dimensional rendering images according to three-dimensional Gaussian data: According to the spatial target posture in each optical image in the data to be reconstructed, a corresponding transformation matrix is ​​given; Different transformation matrices are used to perform two-dimensional projection according to the three-dimensional Gaussian data to obtain a rendering image corresponding to each optical image in the data to be reconstructed.

[0009] In one embodiment, the structure-aware gradient loss function is expressed as: ; In the above formula, and represent the rendered image and the optical image respectively, and They are high-frequency feature extraction in the horizontal and vertical directions respectively. express Norm.

[0010] In one embodiment, the multi-exposure joint optimization strategy is expressed as: ; In the above formula, Optical images representing three different exposure levels, represents the mean absolute error at each exposure level, represents the structural similarity loss at each exposure level, Represents the weight coefficient used to balance the two losses.

[0011] The present application also provides a device for three-dimensional reconstruction of a space target by edge-enhanced multi-exposure, the device comprising: A multi-exposure level data acquisition module is used to acquire multiple optical images of the same space target at different exposure levels and multiple different postures, and to form a data set of optical images at the same exposure level to obtain multiple data sets; A module for selecting data to be reconstructed, used for selecting an optical image from each of the data sets as a group of data to be reconstructed, wherein the spatial targets in each optical image in the data to be reconstructed present different postures; A three-dimensional Gaussian data generation module is used to reconstruct the space target according to a set of data to be reconstructed under a randomly initialized three-dimensional Gaussian distribution using a 3DGS method to obtain three-dimensional Gaussian data; A high-frequency feature image extraction module is used to perform two-dimensional projection based on the three-dimensional Gaussian data according to different postures of the space target in the data to be reconstructed, obtain a corresponding plurality of rendered images, and use a Sobel operator to respectively extract high-frequency feature images of the plurality of rendered images and the corresponding optical image; An optimization iteration module, used to calculate a structure-aware gradient loss function using the high-frequency feature images corresponding to the rendered image and the optical image, and optimize the three-dimensional Gaussian distribution in combination with a multi-exposure joint optimization strategy, and complete an optimization iteration, and after performing a preset number of iterative optimizations on the three-dimensional Gaussian distribution, obtain a final optimized three-dimensional Gaussian distribution; The space target three-dimensional reconstruction module is used to perform three-dimensional reconstruction based on the point cloud converted into the final optimized three-dimensional Gaussian distribution to obtain a three-dimensional reconstruction model of the space target.

[0012] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented: Acquire multiple optical images of the same space target at different exposure levels and multiple poses, and form a data set of optical images at the same exposure level to obtain multiple data sets; Selecting an optical image from each of the data sets as a group of data to be reconstructed, wherein the spatial target in each optical image in the data to be reconstructed presents a different posture; The space target is reconstructed according to a set of data to be reconstructed under a randomly initialized three-dimensional Gaussian distribution by using a 3DGS method to obtain three-dimensional Gaussian data; Based on the three-dimensional Gaussian data, two-dimensional projection is performed according to different postures of the space target in the data to be reconstructed to obtain a corresponding plurality of rendered images, and a Sobel operator is used to respectively extract high-frequency feature images of the plurality of rendered images and the corresponding optical images; Using the high-frequency feature image corresponding to the rendered image and the optical image, a structure-aware gradient loss function is calculated, and the three-dimensional Gaussian distribution is optimized in combination with a multi-exposure joint optimization strategy, and an optimization iteration is completed. After the three-dimensional Gaussian distribution is iteratively optimized for a preset number of times, a final optimized three-dimensional Gaussian distribution is obtained; The three-dimensional Gaussian distribution after the final optimization is converted into a point cloud for three-dimensional reconstruction to obtain a three-dimensional reconstruction model of the space target.

[0013] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps: Acquire multiple optical images of the same space target at different exposure levels and multiple poses, and form a data set of optical images at the same exposure level to obtain multiple data sets; Selecting an optical image from each of the data sets as a group of data to be reconstructed, wherein the spatial target in each optical image in the data to be reconstructed presents a different posture; The space target is reconstructed according to a set of data to be reconstructed under a randomly initialized three-dimensional Gaussian distribution by using a 3DGS method to obtain three-dimensional Gaussian data; Based on the three-dimensional Gaussian data, two-dimensional projection is performed according to different postures of the space target in the data to be reconstructed to obtain a corresponding plurality of rendered images, and a Sobel operator is used to respectively extract high-frequency feature images of the plurality of rendered images and the corresponding optical images; Using the high-frequency feature image corresponding to the rendered image and the optical image, a structure-aware gradient loss function is calculated, and the three-dimensional Gaussian distribution is optimized in combination with a multi-exposure joint optimization strategy, and an optimization iteration is completed. After the three-dimensional Gaussian distribution is iteratively optimized for a preset number of times, a final optimized three-dimensional Gaussian distribution is obtained; The three-dimensional Gaussian distribution after the final optimization is converted into a point cloud for three-dimensional reconstruction to obtain a three-dimensional reconstruction model of the space target.

[0014] The above-mentioned edge-enhanced multi-exposure three-dimensional reconstruction method, device and equipment for space target are as follows: by arbitrarily selecting an optical image from data sets with different exposure levels as a group of data to be reconstructed, the 3DGS method is used to reconstruct the space target according to a group of data under a randomly initialized three-dimensional Gaussian distribution to obtain three-dimensional Gaussian data, and based on the three-dimensional Gaussian data, two-dimensional projection is performed according to different postures of the space target in the data to be reconstructed to obtain corresponding multiple rendered images, and the Sobel operator is used to extract high-frequency feature images of multiple rendered images and corresponding optical images respectively, and the structure-aware gradient loss function is calculated using the high-frequency feature images corresponding to the rendered images and the optical images, and the three-dimensional Gaussian distribution is optimized in combination with the multi-exposure joint optimization strategy, and three-dimensional reconstruction is performed based on the finally optimized three-dimensional Gaussian distribution to overcome the problem of uneven illumination of the real mission heavy space target image and obtain an accurate three-dimensional reconstruction model of the space target. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic diagram of the flow of a method for three-dimensional reconstruction of a space target by edge enhancement and multiple exposure in one embodiment; Figure 2 A schematic block diagram of a method for three-dimensional reconstruction of a space target using edge-enhanced multi-exposure in one embodiment; Figure 3 It is a structural block diagram of a device for three-dimensional reconstruction of a space object with edge enhancement and multiple exposures in one embodiment; Figure 4 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0017] Accurate description of the geometric features of unknown spacecraft is crucial for on-orbit service tasks (such as on-orbit maintenance and condition monitoring). However, due to the changes in observation angle and lighting conditions caused by the movement of the observation platform, the image of the target spacecraft has uneven lighting problems, such as Figure 1 As shown, in this application, a method for three-dimensional reconstruction of a space target with edge enhancement and multiple exposures is proposed, comprising the following steps: Step S100, obtaining multiple optical images of the same space target at different exposure levels and multiple postures, and forming a data set from the optical images at the same exposure level to obtain multiple data sets.

[0018] Step S110 , arbitrarily selecting one optical image from each data set as a group of data to be reconstructed, wherein the spatial target in each optical image in the data to be reconstructed presents a different posture.

[0019] Step S120 , using the 3DGS method under a randomly initialized three-dimensional Gaussian distribution, reconstructing the space target according to a set of data to be reconstructed to obtain three-dimensional Gaussian data.

[0020] Step S130, based on the three-dimensional Gaussian data, two-dimensional projection is performed according to different postures of the spatial target in the data to be reconstructed to obtain corresponding multiple rendered images, and the Sobel operator is used to extract high-frequency feature images of the multiple rendered images and the corresponding optical images.

[0021] In step S140, the structure-aware gradient loss function is calculated using the high-frequency feature image corresponding to the rendered image and the optical image, and the three-dimensional Gaussian distribution is optimized in combination with the multi-exposure joint optimization strategy. An optimization iteration is completed, and the three-dimensional Gaussian distribution is iteratively optimized for a preset number of times to obtain the final optimized three-dimensional Gaussian distribution.

[0022] Step S150 , converting the finally optimized three-dimensional Gaussian distribution into a point cloud for three-dimensional reconstruction to obtain a three-dimensional reconstructed model of the space target.

[0023] In this application, a Gaussian scene representation method (ME-GS) based on multi-exposure joint constraints and edge gradient guidance is proposed to address the uneven illumination of space target images. Based on the 3D Gaussian Splatting (3DGS) framework, this method innovatively introduces a low, medium, and high exposure constraint mechanism, which improves the quality of scene representation through the complementary information of images with different exposures, and effectively improves the accuracy of detail rendering. At the same time, the Sobel edge detection operator is introduced to extract the structural features of spacecraft targets, enhancing the representation capability of high-frequency information of the scene. In view of the problem of the small number of space target datasets, a multi-exposure synthetic dataset for space target reconstruction is also constructed.

[0024] In step S100, in order to solve the problem of less space target data set, multiple exposures are performed on the same space target to enrich the data.

[0025] In this embodiment, the multiple optical images of the space target in different positions are obtained by continuously acquiring multiple photos of the moving space target within a period of time.

[0026] In this embodiment, the multiple optical images of the same space target include three exposure levels: high, medium, and low, that is, they include a high exposure data set, medium exposure data, and low exposure data, respectively.

[0027] In step S110, when using 3DGS for three-dimensional reconstruction, multiple optical images of the same target with different postures are required to obtain a three-dimensional model from a two-dimensional image. Therefore, a set of data to be reconstructed includes three optical images with different exposure levels and different spatial target postures.

[0028] In step S120, the 3DGS method is used for 3D reconstruction. This method is an emerging neural rendering technology. Different from the previous NeRF method, it uses a large number of anisotropic 3D Gaussian distributions to represent the scene. Each 3D Gaussian follows a 3D Gaussian distribution (Formula (1)): (1) In formula (1), represents the mean of the Gaussian points, represents the covariance matrix.

[0029] In addition, each 3D Gaussian also has an opacity and spherical harmonic parameters ( represents degrees of freedom), which is used to model view-dependent color.

[0030] Furthermore, when 3DGS is first used for modeling, its 3D Gaussian distribution is randomly initialized, which means that there is still a big difference between the three-dimensional Gaussian data of the space target constructed at this time and the real space target structure. Therefore, it is necessary to continuously optimize the 3D Gaussian distribution so that the generated three-dimensional Gaussian data, that is, the three-dimensional model of the space target, can gradually solve the real structure.

[0031] In step S130, after obtaining the three-dimensional Gaussian data, it is necessary to project the 3D Gaussian sphere onto the 2D plane according to a given perspective transformation matrix to obtain the corresponding rendered image. The difference between the rendered image and the real image is used to guide the optimization of the 3D Gaussian distribution.

[0032] Specifically, when obtaining multiple two-dimensional rendered images based on three-dimensional Gaussian data: a corresponding transformation matrix is ​​given according to the posture of the spatial target in each optical image in the data to be reconstructed, and two-dimensional projection is performed according to the three-dimensional Gaussian data using different transformation matrices to obtain a rendered image corresponding to each optical image in the data to be reconstructed.

[0033] In this embodiment, when the 3D Gaussian sphere is projected onto the 2D plane, the corresponding 2D covariance matrix It can be expressed as: (2) In formula (2), The affine approximate Jacobian matrix representing the projective transformation, Represents the viewing angle transformation matrix, which projects the 3D Gaussian sphere onto the 2D plane.

[0034] Furthermore, after projecting the 3D Gaussian onto the 2D plane, the color of each pixel in the image is calculated. For each pixel in the image, the color of the pixels that overlap with it is calculated by blending the pixels that overlap with it. An ordered 3D Gaussian is used to calculate the color of a pixel using the following formula: (3) In formula (3), Represents the final color value of a pixel in the image, represents the spherical harmonic parameters of the ith Gaussian distribution, represents the effective opacity of the ith Gaussian distribution, represents the effective opacity of the jth Gaussian distribution, represents the original opacity of the ith Gaussian distribution, Represents the mean vector of the ith Gaussian distribution.

[0035] Furthermore, after rendering the image, 3DGS optimizes and Loss, the rendered image and the real image are compared pixel by pixel to optimize the 3D Gaussian distribution. The loss function is expressed as: (4) In formula (4), represents a hyperparameter, preferably, in 3DGS, Set to 0.2. represents the structural similarity index loss function.

[0036] Considering that in the process of optimizing the 3D Gaussian distribution, if only the pixel-level loss function ( and ) is optimized. This method mainly focuses on the direct difference of pixel values, which is difficult to fully capture the geometric structure characteristics of spatial targets and easily causes structural distortion and blurred details and edges. In order to accurately capture and enhance the geometric structure and surface details of the target object, this method introduces a structure-aware gradient loss based on the Sobel operator. As a classic first-order differential operator, the Sobel operator calculates The Sobel operator extracts high-frequency information of the image by using pixel differences within the neighborhood, which can effectively capture high-frequency structural features such as the edge and surface contour of the object. The convolution kernels are used to calculate the horizontal direction ( ) and vertical direction ( )’s gradient: (5) (6) In step S140, when optimizing the three-dimensional Gaussian distribution, after respectively extracting high-frequency feature images of multiple rendered images and corresponding optical images (ie, real images) using the above-mentioned Sobel operator, The norm defines the structure-aware gradient loss, expressed as: (7) In formula (7), and represent the rendered image and the optical image respectively, and They are high-frequency feature extraction in the horizontal and vertical directions respectively. express Norm.

[0037] Furthermore, since 3DGS follows the convention of NeRF, a single image is randomly selected for optimization in each iteration. However, such a random single optimization strategy cannot fully utilize the detail information obtained by the multi-exposure imaging mechanism. To solve this problem, a multi-view exposure joint optimization strategy is proposed. In each optimization iteration, three images with different exposure levels are selected at the same time. ,in , and Represent low exposure, medium exposure and high exposure respectively. This strategy effectively utilizes the complementary information under different exposure conditions: low exposure images capture the details of high exposure parts; high exposure images show the structural information of shadow areas; medium exposure images maintain the overall structural balance.

[0038] In this embodiment, when constructing the multi-exposure joint optimization loss, the mean absolute error is also considered. and D-SSIM loss , then the multi-exposure joint optimization strategy is expressed as: (8) In formula (8), Optical images representing three different exposure levels, represents the mean absolute error at each exposure level, Represents the structural similarity loss at each exposure level. represents the weight coefficient used to balance the two losses. Preferably, Set to 0.2.

[0039] In order to achieve high-fidelity reconstruction of spatial objects, in this embodiment, the two key components of structure perception and multi-exposure constraint are unified into an optimization framework, that is, the structure-aware gradient loss function is combined with the multi-exposure joint optimization strategy, and the overall optimization goal is defined as: (9) In formula (9), The hyperparameter representing the gradient loss weight is preferably set to 0.2. This joint optimization strategy is specially designed for the characteristics of space target imaging. The structure-aware gradient loss maintains the edge and contour features of the spacecraft surface by extracting high-frequency features, while the multi-exposure joint optimization is used to solve the problem of multi-exposure imaging.

[0040] Furthermore, after completing one iterative optimization, an optical image is randomly selected from the three data sets, and the optimized three-dimensional Gaussian distribution is used for three-dimensional reconstruction, and then two-dimensional projection is performed, the loss function is calculated, and then the three-dimensional Gaussian distribution is optimized. Steps S120 to S140 are repeated until the number of iterations reaches a preset number, and the optimization is completed to obtain the final three-dimensional Gaussian distribution. In step S150, the optimized final three-dimensional Gaussian distribution is used to generate a three-dimensional model of the space target, and the three-dimensional model can accurately represent the structure of the space target.

[0041] like Figure 2 Shown is a schematic diagram of the process of this method.

[0042] In the above-mentioned edge-enhanced multi-exposure space target three-dimensional reconstruction method, the problem of uneven illumination caused by the optical image of the space target in the cosmic environment is addressed. The traditional method is often prone to overfitting a certain image when processing such multi-exposure images, and it is difficult to fully utilize the benefits of the multi-exposure imaging mechanism. This method introduces multi-exposure constraints on the basis of 3DGS, innovatively designs low, medium and high three-level exposure constraint mechanisms, and fully utilizes the complementary information of different exposures to improve the quality of scene reconstruction. At the same time, the Sobel edge feature extraction technology is integrated to further enhance the ability to represent scene structure details. The effectiveness of this method is also proved by experiments. The experimental results show that, as shown in Tables 1-3, this method has achieved a relatively significant improvement in the three key indicators of PSNR, SSIM and LPIPS. Among them, this method has achieved a 33.8% improvement in the LPIPS indicator, and the LPIPS value has decreased by 33.8% compared with the baseline. This fully proves that this method can overcome the challenges brought by multi-exposure and achieve better visual reconstruction quality.

[0043] Table 1. PSNR (the larger the better) comparison values

[0044] Table 2. SSIM (larger is better) comparison values

[0045] Table 3. LPIPS (smaller is better) comparison values

[0046] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0047] In one embodiment, Figure 3 As shown, a device for three-dimensional reconstruction of space targets with edge enhancement and multiple exposures is provided, comprising: a multi-exposure level data acquisition module 200, a to-be-reconstructed data selection module 210, a three-dimensional Gaussian data generation module 220, a high-frequency feature image extraction module 230, an optimization iteration module 240 and a space target three-dimensional reconstruction module 250, wherein: The multi-exposure level data acquisition module 200 is used to acquire multiple optical images of the same space target at different exposure levels and multiple different postures, and form a data set of optical images at the same exposure level to obtain multiple data sets; The to-be-reconstructed data selection module 210 is used to select any one optical image from each of the data sets as a group of to-be-reconstructed data, wherein the spatial targets in each optical image in the to-be-reconstructed data present different postures; A three-dimensional Gaussian data generating module 220 is used to reconstruct the space target according to a set of data to be reconstructed under a randomly initialized three-dimensional Gaussian distribution by using a 3DGS method to obtain three-dimensional Gaussian data; A high-frequency feature image extraction module 230 is used to perform two-dimensional projection based on the three-dimensional Gaussian data according to different postures of the space target in the data to be reconstructed, obtain a corresponding plurality of rendered images, and use a Sobel operator to respectively extract high-frequency feature images of the plurality of rendered images and the corresponding optical image; An optimization iteration module 240 is used to calculate a structure-aware gradient loss function using the high-frequency feature images corresponding to the rendered image and the optical image, and optimize the three-dimensional Gaussian distribution in combination with a multi-exposure joint optimization strategy, and complete an optimization iteration, and after performing a preset number of iterative optimizations on the three-dimensional Gaussian distribution, obtain a final optimized three-dimensional Gaussian distribution; The space target three-dimensional reconstruction module 250 is used to perform three-dimensional reconstruction based on the point cloud converted into the final optimized three-dimensional Gaussian distribution to obtain a three-dimensional reconstructed model of the space target.

[0048] For the specific definition of the edge-enhanced multi-exposure three-dimensional reconstruction device for space targets, please refer to the definition of the edge-enhanced multi-exposure three-dimensional reconstruction method for space targets in the above text, which will not be repeated here. Each module in the above-mentioned edge-enhanced multi-exposure three-dimensional reconstruction device for space targets can be fully or partially implemented by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above-mentioned modules.

[0049] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for three-dimensional reconstruction of a space target with edge enhancement and multiple exposures is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0050] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0051] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented: Acquire multiple optical images of the same space target at different exposure levels and multiple poses, and form a data set of optical images at the same exposure level to obtain multiple data sets; Selecting an optical image from each of the data sets as a group of data to be reconstructed, wherein the spatial target in each optical image in the data to be reconstructed presents a different posture; The space target is reconstructed according to a set of data to be reconstructed under a randomly initialized three-dimensional Gaussian distribution by using a 3DGS method to obtain three-dimensional Gaussian data; Based on the three-dimensional Gaussian data, two-dimensional projection is performed according to different postures of the space target in the data to be reconstructed to obtain a corresponding plurality of rendered images, and a Sobel operator is used to respectively extract high-frequency feature images of the plurality of rendered images and the corresponding optical images; Using the high-frequency feature image corresponding to the rendered image and the optical image, a structure-aware gradient loss function is calculated, and the three-dimensional Gaussian distribution is optimized in combination with a multi-exposure joint optimization strategy, and an optimization iteration is completed. After the three-dimensional Gaussian distribution is iteratively optimized for a preset number of times, a final optimized three-dimensional Gaussian distribution is obtained; The three-dimensional Gaussian distribution after the final optimization is converted into a point cloud for three-dimensional reconstruction to obtain a three-dimensional reconstruction model of the space target.

[0052] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: Acquire multiple optical images of the same space target at different exposure levels and multiple poses, and form a data set of optical images at the same exposure level to obtain multiple data sets; Selecting an optical image from each of the data sets as a group of data to be reconstructed, wherein the spatial target in each optical image in the data to be reconstructed presents a different posture; The space target is reconstructed according to a set of data to be reconstructed under a randomly initialized three-dimensional Gaussian distribution by using a 3DGS method to obtain three-dimensional Gaussian data; Based on the three-dimensional Gaussian data, two-dimensional projection is performed according to different postures of the space target in the data to be reconstructed to obtain a corresponding plurality of rendered images, and a Sobel operator is used to respectively extract high-frequency feature images of the plurality of rendered images and the corresponding optical images; Using the high-frequency feature image corresponding to the rendered image and the optical image, a structure-aware gradient loss function is calculated, and the three-dimensional Gaussian distribution is optimized in combination with a multi-exposure joint optimization strategy, and an optimization iteration is completed. After the three-dimensional Gaussian distribution is iteratively optimized for a preset number of times, a final optimized three-dimensional Gaussian distribution is obtained; The three-dimensional Gaussian distribution after the final optimization is converted into a point cloud for three-dimensional reconstruction to obtain a three-dimensional reconstruction model of the space target.

[0053] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0054] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0055] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the method of the present application shall be subject to the attached claims.

Claims

1. A method for three-dimensional reconstruction of space objects using edge-enhanced multi-exposure, characterized in that: The method comprises: Acquire multiple optical images of the same space target at different exposure levels and multiple poses, and form a data set of optical images at the same exposure level to obtain multiple data sets; Selecting an optical image from each of the data sets as a group of data to be reconstructed, wherein the spatial target in each optical image in the data to be reconstructed presents a different posture; The space target is reconstructed according to a set of data to be reconstructed under a randomly initialized three-dimensional Gaussian distribution by using a 3DGS method to obtain three-dimensional Gaussian data; Based on the three-dimensional Gaussian data, two-dimensional projection is performed according to different postures of the space target in the data to be reconstructed to obtain a corresponding plurality of rendered images, and a Sobel operator is used to respectively extract high-frequency feature images of the plurality of rendered images and the corresponding optical images; Using the high-frequency feature image corresponding to the rendered image and the optical image, a structure-aware gradient loss function is calculated, and the three-dimensional Gaussian distribution is optimized in combination with a multi-exposure joint optimization strategy, and an optimization iteration is completed. After the three-dimensional Gaussian distribution is iteratively optimized for a preset number of times, a final optimized three-dimensional Gaussian distribution is obtained; The three-dimensional Gaussian distribution after the final optimization is converted into a point cloud for three-dimensional reconstruction to obtain a three-dimensional reconstruction model of the space target.

2. The edge-enhanced multi-exposure space target 3D reconstruction method according to claim 1, characterized in that: Multiple optical images of the same space target include three exposure levels: high, medium and low.

3. The edge-enhanced multi-exposure space target 3D reconstruction method according to claim 2, characterized in that: When obtaining multiple two-dimensional rendered images based on three-dimensional Gaussian data: According to the spatial target posture in each optical image in the data to be reconstructed, a corresponding transformation matrix is ​​given; Different transformation matrices are used to perform two-dimensional projection according to the three-dimensional Gaussian data to obtain a rendering image corresponding to each optical image in the data to be reconstructed.

4. The edge-enhanced multi-exposure space target 3D reconstruction method according to claim 3, characterized in that: The structure-aware gradient loss function is expressed as: In the above formula, and represent the rendered image and the optical image respectively, and They are high-frequency feature extraction in the horizontal and vertical directions respectively. express Norm.

5. The edge-enhanced multi-exposure space target 3D reconstruction method according to claim 4, characterized in that: The multi-exposure joint optimization strategy is expressed as: In the above formula, Optical images representing three different exposure levels, represents the mean absolute error at each exposure level, represents the structural similarity loss at each exposure level, Represents the weight coefficient used to balance the two losses.

6. An edge-enhanced multi-exposure space target three-dimensional reconstruction device, characterized in that: The device comprises: A multi-exposure level data acquisition module is used to acquire multiple optical images of the same space target at different exposure levels and multiple different postures, and to form a data set of optical images at the same exposure level to obtain multiple data sets; A module for selecting data to be reconstructed, used for selecting an optical image from each of the data sets as a group of data to be reconstructed, wherein the spatial targets in each optical image in the data to be reconstructed present different postures; A three-dimensional Gaussian data generation module is used to reconstruct the space target according to a set of data to be reconstructed under a randomly initialized three-dimensional Gaussian distribution using a 3DGS method to obtain three-dimensional Gaussian data; A high-frequency feature image extraction module is used to perform two-dimensional projection based on the three-dimensional Gaussian data according to different postures of the space target in the data to be reconstructed, obtain a corresponding plurality of rendered images, and use a Sobel operator to respectively extract high-frequency feature images of the plurality of rendered images and the corresponding optical image; An optimization iteration module, used to calculate a structure-aware gradient loss function using the high-frequency feature images corresponding to the rendered image and the optical image, and optimize the three-dimensional Gaussian distribution in combination with a multi-exposure joint optimization strategy, and complete an optimization iteration, and after performing a preset number of iterative optimizations on the three-dimensional Gaussian distribution, obtain a final optimized three-dimensional Gaussian distribution; The space target three-dimensional reconstruction module is used to perform three-dimensional reconstruction based on the point cloud converted into the final optimized three-dimensional Gaussian distribution to obtain a three-dimensional reconstruction model of the space target.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

Images