Asteroid pose tracking method and system based on monocular vision and three-dimensional contour

Abstract

The invention relates to the technical field of computer vision, and discloses an asteroid pose tracking method and system based on monocular vision and a three-dimensional contour. The method specifically comprises the following steps: extracting and obtaining a three-dimensional contour section from all three-dimensional line segments on the surface of an asteroid three-dimensional model, and establishing a matching relationship between control points of the three-dimensional contour section and image edge points; and then, by taking an angle difference between a back projection line of a minimized image edge point and a projection plane of a corresponding three-dimensional contour section as a target, establishing an objective function of an asteroid pose, and obtaining a maximum likelihood estimation value of the asteroid pose by optimizing and solving the objective function. The pose optimization method is carried out in a three-dimensional space, and has good robustness for model scale, target distance change, background noise and shielding phenomena. And finally, the maximum likelihood estimation value is corrected by using a second-order autoregression volume Kalman model to obtain the maximum posterior probability estimation value of the asteroid, so that the tracking stability performance is greatly improved.

Description

Technical Field

[0001] The present application relates to the field of computer vision technology, and in particular to a method and system for tracking an asteroid posture based on monocular vision and three-dimensional contours. Background Art

[0002] Computer vision-based six-degree-of-freedom pose tracking technology is a core technology in fields such as space exploration and robotic navigation. Its goal is to determine the three-dimensional position and pose of a target object in real time using camera sensors. Monocular cameras, with their low price, good real-time performance, high accuracy, and ease of deployment, play a key role in visual navigation. In deep space exploration scenarios, asteroid surfaces lack regular geometric structures and significant texture features, and are subject to interference from drastic lighting changes and self-occlusion. Traditional pose estimation methods, due to their strong reliance on structured features, are insufficiently adaptable to irregular celestial scenes.

[0003] Current mainstream technologies primarily utilize geometric feature-driven methods for tracking, also known as edge-alignment-based pose estimation methods. For example, the edge-alignment-based pose estimation framework proposed in European Patent EP3285186B1 optimizes pose by iteratively matching 3D model contours with image edges. While these edge-alignment-based methods avoid texture dependency, they employ traditional reprojection error minimization strategies, making them sensitive to model scale errors and image noise. They are also susceptible to scale uncertainty and noise interference, and are prone to falling into local optimality in occluded scenes. Balancing computational efficiency and accuracy is difficult in dynamic scenes. Summary of the Invention

[0004] This application provides an asteroid pose tracking method based on monocular vision and three-dimensional contours to address the problems in the existing technology. The traditional pose estimation method has a strong dependence on structured features, resulting in insufficient adaptability in irregular celestial scenes, and the optimization framework based on reprojection error is susceptible to scale uncertainty and noise interference, making it difficult to balance computational efficiency and accuracy in dynamic scenes.

[0005] Correspondingly, the present application also provides an asteroid posture tracking system based on monocular vision and three-dimensional contours, an electronic device, and a computer-readable storage medium to ensure the implementation and application of the above method.

[0006] In order to solve the above technical problems, the present application discloses a method for tracking an asteroid's posture based on monocular vision and three-dimensional contours, the method comprising:

[0007] Extracting a three-dimensional contour segment from all three-dimensional line segments on the surface of a preset asteroid three-dimensional model, and determining control points of the three-dimensional contour segment;

[0008] Extract image edge points from asteroid monocular images and establish a matching relationship between control points and image edge points;

[0009] The objective function of asteroid pose is established with the goal of minimizing the angular difference between the back-projected line of the image edge point and the projection plane of the corresponding 3D contour segment.

[0010] Optimize and solve the objective function to obtain the maximum likelihood estimate of the asteroid's position;

[0011] The maximum likelihood estimate is corrected using a second-order autoregressive cubature Kalman model to obtain the maximum a posteriori probability estimate of the asteroid.

[0012] Preferably, extracting a three-dimensional contour segment from all three-dimensional line segments on the surface of a preset three-dimensional model of the asteroid and determining the control points of the three-dimensional contour segment include:

[0013] Project all three-dimensional line segments on the surface of the preset asteroid three-dimensional model onto the image plane to generate corresponding two-dimensional projection line segments;

[0014] When the back-projection line of the midpoint of the two-dimensional projection line segment is tangent to the three-dimensional model of the asteroid, the corresponding three-dimensional line segment is determined as a three-dimensional contour segment;

[0015] The points on the 3D contour segment corresponding to the midpoints of the 2D projection line segments are determined as control points.

[0016] Preferably, the objective function of the asteroid pose is established with the goal of minimizing the angular difference between the back-projected line of the image edge point and the projection plane of the corresponding three-dimensional contour segment, including:

[0017] Determine the back-projection line of the edge point of the image and the normal vector of the projection plane of the three-dimensional contour segment;

[0018] Let α ε is the angle between the back-projection line and the normal vector. Minimizing the angle difference between the back-projection line and the projection plane is equivalent to minimizing |cosα ε |;

[0019] To minimize |cosα ε | is the goal, and the objective function of asteroid posture is established.

[0020] Preferably, to minimize |cosα ε | is the goal, and the objective function of asteroid posture is established, including:

[0021] Use M to estimate the pair |cosα ε |Optimize to minimize |cosα ε The sum of M estimators ρ of | is used as the target to establish the objective function:

[0022]

[0023] in, d ε is the normal vector of the projection plane of the 3D contour segment, u ε is the back projection line of the edge point of the image, σ is {cosα ε}In the initial pose p given by the volumetric Kalman model k 0 The median absolute deviation below .

[0024] Preferably, optimizing and solving the objective function to obtain a maximum likelihood estimate of the asteroid's position and pose includes:

[0025] The objective function is solved by the iterative least squares algorithm to obtain the maximum likelihood estimate of the asteroid's position.

[0026] Preferably, in the process of solving the objective function by the iterative least squares algorithm, the iterative formula used is:

[0027] p (s+1) =p (s) -(J (s)T W (s) J (s) +τI6) -1 J (s)T W (s) v (s)

[0028] Where, v (s) =[cosα1 (s) cosα2 (s) …cosα n (s) ] T ; s represents the number of iteration steps; τ is a non-negative robust factor coefficient; W (s) represents the unit matrix {…ω(cosα ε (s) / σ)…}, where ω(x) represents the weight function of the M estimator; J (s) Indicates v (s) P (s) The Jacobian matrix of .

[0029] Preferably, a second-order autoregressive cubature Kalman model is used to correct the maximum likelihood estimate to obtain a maximum a posteriori probability estimate of the asteroid, including:

[0030] Define the state variable of the volumetric Kalman model as s k =[p k T p k-1 T ]T , where p -1 =p0;

[0031] The state equation of the volumetric Kalman model is determined as:

[0032] Ω k+1 =log SO(3) (R k R k-1 -1 R k )+Γ k

[0033] t k+1 =R k R k-1 -1 (t k -t k-1 )+t k +Ψ k

[0034] Ω k =Ω k +η k

[0035] t k =t k +κ k

[0036] Where, T = exp(p ^ ) is the maximum likelihood estimate of the asteroid's position p k The rigid body motion matrix generated in log SO(3) (R) represents the rotation vector Ω obtained by inversely solving the rotation matrix R; Γ k ,Ψ k ,η k and κ k represents the prediction error vector;

[0037] The observation equation for the volumetric Kalman model is:

[0038] z k =[I6 O6]s k +ε k

[0039] Where I6 represents the sixth-order unit matrix, O6 represents the sixth-order zero matrix, ε k represents the observation noise vector, z k is the observed quantity;

[0040] The maximum a posteriori probability estimate of asteroids is obtained from solving the volumetric Kalman model.

[0041] The present application also discloses an asteroid posture tracking system based on monocular vision and three-dimensional profile, the system comprising:

[0042] A 3D contour extraction module is used to extract 3D contour segments from all 3D line segments on the surface of a preset asteroid 3D model and determine the control points of the 3D contour segments;

[0043] Image edge extraction module, used to extract image edge points from asteroid monocular images and establish a matching relationship between control points and image edge points;

[0044] The pose estimation module is used to establish an objective function for the asteroid pose with the goal of minimizing the angle difference between the back-projection line of the image edge point and the projection plane of the corresponding three-dimensional contour segment;

[0045] The pose estimation module is also used to optimize and solve the objective function to obtain the maximum likelihood estimate of the asteroid's pose;

[0046] The pose correction module is used to correct the maximum likelihood estimate using a second-order autoregressive cubature Kalman model to obtain the maximum a posteriori probability estimate of the asteroid.

[0047] The present application also discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, one or more methods described in the present application are implemented.

[0048] The present application also discloses a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, one or more methods described in the present application are implemented.

[0049] In this application, a three-dimensional contour segment is extracted from all three-dimensional line segments on the surface of the asteroid three-dimensional model, and the control points of the three-dimensional contour segment are determined. The image edge points are extracted from the asteroid monocular image, and a matching relationship between the control points and the image edge points is established. Then, with the goal of minimizing the angular difference between the back-projection line of the image edge point and the projection plane of the corresponding three-dimensional contour segment, an objective function of the asteroid posture is established, and the maximum likelihood estimate of the asteroid posture is obtained by optimizing and solving the objective function. The above-mentioned posture optimization method is performed in three-dimensional space and has good robust performance against model scale, target distance changes, background noise and occlusion. Finally, the maximum likelihood estimate is corrected using a second-order autoregressive cubature Kalman model to obtain the maximum posterior probability estimate of the asteroid, which greatly improves the tracking stability.

[0050] In addition, the acquisition of the three-dimensional contour in this application does not rely on expensive acceleration units (such as GPU and FPGA), has low economic cost, high efficiency, and high economic practicality.

[0051] Additional aspects and advantages of the present application will be given in the following description, which will become apparent from the following description, or will be understood through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0053] Figure 1 Flowchart of the asteroid posture tracking method based on monocular vision and three-dimensional contours provided in an embodiment of the present application;

[0054] Figure 2 A schematic diagram of a posture optimization algorithm provided in an embodiment of the present application;

[0055] Figure 3 A schematic diagram of the structure of the asteroid posture tracking system based on monocular vision and three-dimensional contours provided in an embodiment of the present application;

[0056] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0057] The following describes embodiments of the present application in detail. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application and are not to be construed as limiting the present application.

[0058] It will be understood by those skilled in the art that, unless expressly stated otherwise, the singular forms "a", "an", "said" and "the" used herein may also include the plural forms. It should be further understood that the term "comprising" used in the specification of this application refers to the presence of features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, "connected" or "coupled" as used herein may include wireless connections or wireless couplings. The term "and / or" used herein includes all or any units and all combinations of one or more associated listed items.

[0059] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those skilled in the art in the art to which the present invention pertains. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with their meanings in the context of the prior art and, unless specifically defined as herein, will not be interpreted in an idealized or overly formal sense.

[0060] The solution provided in the embodiments of the present application can be executed by any electronic device, such as a terminal device or a server, wherein the server can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services. The terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, etc., but is not limited to this. The terminal and the server can be directly or indirectly connected via wired or wireless communication, and this application does not limit this. With respect to the technical problems existing in the prior art, the asteroid posture tracking method and system based on monocular vision and three-dimensional contours provided in this application are intended to solve at least one of the technical problems of the prior art.

[0061] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0062] The present application embodiment provides a possible implementation method, such as Figure 1 As shown, a flowchart of an asteroid posture tracking method based on monocular vision and three-dimensional contours is provided. The scheme can be executed by any electronic device, and optionally, can be executed on the server side or terminal device.

[0063] like Figure 1 As shown in , the method may include the following steps:

[0064] Step 101: extracting 3D contour segments from all 3D line segments on the surface of a preset 3D asteroid model, and determining control points of the 3D contour segments;

[0065] The asteroid's 3D model is a mesh model, which includes vertices, edges formed by two vertices, and faces composed of three or more vertices. In the embodiment of the present application, the edges are treated as 3D line segments, and 3D line segments that meet preset conditions are selected as 3D contour segments.

[0066] Step 102: extract image edge points from the asteroid monocular image and establish a matching relationship between the control points and the image edge points.

[0067] Step 103: Establish an objective function for the asteroid's posture with the goal of minimizing the angle difference between the back-projection line of the image edge point and the projection plane of the corresponding three-dimensional contour segment.

[0068] Step 104: Optimize and solve the objective function to obtain the maximum likelihood estimate of the asteroid's position and posture.

[0069] In the embodiment of the present application, the geometric correlation between 3D contours and 2D edges is combined with the angular residual optimization strategy of the back-projection line and the projection plane to improve the robustness to model scale, target distance changes, background noise and occlusion phenomena.

[0070] Step 105: Use a second-order autoregressive cubature Kalman model to correct the maximum likelihood estimate to obtain a maximum a posteriori probability estimate of the asteroid.

[0071] Using the maximum a posteriori probability estimate as the final estimate of the asteroid's pose can further improve the accuracy of pose estimation and thus improve the tracking stability.

[0072] In an embodiment of the present application, a three-dimensional contour segment is extracted from all three-dimensional line segments on the surface of the asteroid three-dimensional model, and the control points of the three-dimensional contour segment are determined. The image edge points are extracted from the asteroid monocular image, and a matching relationship between the control points and the image edge points is established. Then, with the goal of minimizing the angular difference between the back-projection line of the image edge point and the projection plane of the corresponding three-dimensional contour segment, an objective function of the asteroid posture is established, and the maximum likelihood estimate of the asteroid posture is obtained by optimizing and solving the objective function. The above-mentioned posture optimization method is performed in three-dimensional space and has good robust performance against model scale, target distance changes, background noise and occlusion. Finally, the maximum likelihood estimate is corrected using a second-order autoregressive cubature Kalman model to obtain the maximum posterior probability estimate of the asteroid, which greatly improves the tracking stability.

[0073] In an optional embodiment, extracting a 3D contour segment from all 3D line segments on the surface of a preset 3D model of the asteroid and determining the control points of the 3D contour segment includes:

[0074] Project all three-dimensional line segments on the surface of the preset asteroid three-dimensional model onto the image plane to generate corresponding two-dimensional projection line segments;

[0075] When the back-projection line of the midpoint of the two-dimensional projection line segment is tangent to the three-dimensional model of the asteroid, the corresponding three-dimensional line segment is determined as a three-dimensional contour segment;

[0076] The points on the 3D contour segment corresponding to the midpoints of the 2D projection line segments are determined as control points.

[0077] In the embodiment of the present application, the six-degree-of-freedom posture p is first defined using the SO(3) group, which can be expressed as [Ω T t T ] T , where Ω represents the three-dimensional rotation vector and t represents the three-dimensional translation vector.

[0078] Given an asteroid 3D model, in order to extract 3D contour segments from the asteroid 3D model, firstly, all 3D line segments on the surface of the 3D mesh model are passed through the initial pose p k 0 (provided by the state equation of the volume Kalman model shown in equation (5) below) is projected onto the image plane to generate a large number of two-dimensional projection line segments. Since the three-dimensional line segments of the asteroid grid model are usually short, in the embodiment of the present application, it is only necessary to judge whether the three-dimensional line segment can be regarded as a three-dimensional contour segment by the projection characteristics of the midpoint of the line segment. Specifically, for a three-dimensional line segment, if the back projection line of the midpoint of its two-dimensional projection line segment is tangent to the grid model, then the three-dimensional line segment will be regarded as a three-dimensional contour segment. At this time, the three-dimensional point on the line segment corresponding to the midpoint of the projection line segment will be defined as a control point. All three-dimensional contour segments and their corresponding control points are respectively marked as {L c} and {M c}.

[0079] The three-dimensional contour extraction by the above method does not rely on expensive acceleration units (such as GPU and FPGA), and has low economic cost and high efficiency.

[0080] In an optional embodiment, an objective function for the asteroid pose is established with the goal of minimizing the angular difference between the back-projected line of the image edge point and the projection plane of the corresponding three-dimensional contour segment, including:

[0081] Determine the back-projection line of the edge point of the image and the normal vector of the projection plane of the three-dimensional contour segment;

[0082] Let α ε is the angle between the back-projection line and the normal vector. Minimizing the angle difference between the back-projection line and the projection plane is equivalent to minimizing |cosα ε |;

[0083] To minimize |cosα ε | is the goal, and the objective function of asteroid posture is established.

[0084] The projection plane of the three-dimensional contour segment is a plane determined by the three-dimensional contour segment and the optical center of the camera.

[0085] In the embodiment of the present application, real-time pose estimation is performed based on the SO(3) group, as follows:

[0086] First, the image edge is extracted from the asteroid monocular image using the Canny operator. The image edge is composed of a series of image edge points. In this embodiment, the control point {M c} and the edge points of the image.

[0087] Figure 2 The visualization process of constructing the pose optimization objective function is shown. At time k, assuming that n 3D contour segments {L ε ,ε=1,2,…,n}, where each 3D contour segment L ε The matched two-dimensional image edge point is {x ε ,ε=1,2,…,n}, then each image edge point x ε The back projection line u ε It can be expressed as:

[0088]

[0089] Where K represents the camera intrinsic parameter matrix, Represents x ε A homogeneous vector of .

[0090] Let C ε 1 and C ε 2 is the three-dimensional contour segment L ε The endpoint of L ε The normal vector d of the projection plane ε It can be calculated by the following formula:

[0091]

[0092] in, And φ ε =C ε 2 -C ε 1 .

[0093] Let α ε for u ε and d ε The angle between them is minimized ε and π ε The angle difference between them is equivalent to minimizing |cosα ε |, therefore, in the embodiment of the present application, the method of minimizing |cosα ε | is the goal, and the objective function of asteroid posture is established.

[0094] In an alternative embodiment, to minimize |cosα ε | is the goal, and the objective function of asteroid posture is established, including:

[0095] In order to weaken the influence of outliers, use M estimators to |cosα ε |Optimize to minimize |cosα ε The sum of M estimators ρ of | is used as the target to establish the objective function:

[0096]

[0097] in, d ε is the normal vector of the projection plane of the 3D contour segment, u ε is the back projection line of the edge point of the image, σ is {cosα ε}In the initial pose p given by the volumetric Kalman model k 0 The median absolute deviation under M estimator ρ can use Tukey estimator.

[0098] In an optional embodiment, optimizing and solving the objective function to obtain a maximum likelihood estimate of the asteroid's position and pose includes:

[0099] The objective function is solved by the iterative least squares algorithm to obtain the maximum likelihood estimate of the asteroid's position.

[0100] In an optional embodiment, in the process of solving the objective function by an iterative least squares algorithm, the iterative formula used is:

[0101] p (s+1) =p (s) -(J (s)T W (s) J (s) +τI6) -1 J (s)T W (s) v (s) (4)

[0102] Where, v (s) =[cosα1 (s) cosα2 (s) …cosα n (s) ] T ; s represents the number of iteration steps; τ is a non-negative robust factor coefficient; W (s) represents the unit matrix {…ω(cosα ε (s) / σ)…}, where ω(x) represents the weight function of the M estimator; J (s) Indicates v(s) P (s) The Jacobian matrix of .

[0103] The maximum likelihood estimate of the pose parameters at the current k moments can be obtained through this optimization algorithm.

[0104] In the embodiment of the present application, the above-mentioned posture optimization algorithm is performed in three-dimensional space, and has good robustness to model scale, target distance changes, background noise and occlusion phenomena.

[0105] In an optional embodiment, a second-order autoregressive cubature Kalman model is used to correct the maximum likelihood estimate to obtain a maximum a posteriori probability estimate of the asteroid, including:

[0106] Define the state variable of the SCKF as s k =[p k T p k-1 T ] T , where p -1 =p0;

[0107] The state equation of the volumetric Kalman model is determined as:

[0108]

[0109] Where, T = exp(p ^ ) is the maximum likelihood estimate of the asteroid's position p k The rigid body motion matrix generated in log SO(3) (R) represents the rotation vector Ω obtained by inversely solving the rotation matrix R; Γ k ,Ψ k ,η k and κ k represents the prediction error vector;

[0110] The observation equation for the volumetric Kalman model is:

[0111] z k =[I6 O6]s k +ε k (6)

[0112] Where I6 represents the sixth-order unit matrix, O6 represents the sixth-order zero matrix, ε k represents the observation noise vector, z k is the observed quantity;

[0113] The maximum a posteriori probability estimate of asteroids is obtained from solving the volumetric Kalman model.

[0114] In the embodiment of the present application, the current k-time pose vector p can be calculated from the SCKF k The maximum a posteriori probability estimate is the final pose result in the embodiment of the present application. The introduction of the second-order autoregressive cubature Kalman model makes the final tracking result meet the maximum a posteriori probability meaning, greatly improving the tracking stability.

[0115] Based on the same principle as the method provided in the embodiment of the present application, the embodiment of the present application also provides an asteroid posture tracking system based on monocular vision and three-dimensional contours, such as Figure 3 As shown, the system includes:

[0116] A 3D contour extraction module 301 is used to extract 3D contour segments from all 3D line segments on the surface of a preset 3D asteroid model and determine the control points of the 3D contour segments;

[0117] An image edge extraction module 302 is used to extract image edge points from the asteroid monocular image and establish a matching relationship between the control points and the image edge points;

[0118] The pose estimation module 303 is used to establish an objective function for the asteroid pose with the goal of minimizing the angle difference between the back-projection line of the image edge point and the projection plane of the corresponding three-dimensional contour segment;

[0119] The pose estimation module 303 is also used to optimize and solve the objective function to obtain the maximum likelihood estimate of the asteroid's pose;

[0120] The pose correction module 304 is used to correct the maximum likelihood estimate using a second-order autoregressive cubature Kalman model to obtain a maximum a posteriori probability estimate of the asteroid.

[0121] In an embodiment of the present application, a three-dimensional contour segment is extracted from all three-dimensional line segments on the surface of the asteroid three-dimensional model, and the control points of the three-dimensional contour segment are determined. The image edge points are extracted from the asteroid monocular image, and a matching relationship between the control points and the image edge points is established. Then, with the goal of minimizing the angular difference between the back-projection line of the image edge point and the projection plane of the corresponding three-dimensional contour segment, an objective function of the asteroid posture is established, and the maximum likelihood estimate of the asteroid posture is obtained by optimizing and solving the objective function. The above-mentioned posture optimization method is performed in three-dimensional space and has good robust performance against model scale, target distance changes, background noise and occlusion. Finally, the maximum likelihood estimate is corrected using a second-order autoregressive cubature Kalman model to obtain the maximum posterior probability estimate of the asteroid, which greatly improves the tracking stability.

[0122] The asteroid posture tracking system based on monocular vision and three-dimensional contours provided in the embodiment of the present application can achieve Figure 1 and Figure 2 To avoid repetition, the various processes implemented in the method embodiment will not be described again here.

[0123] The asteroid posture tracking system based on monocular vision and three-dimensional contours in the embodiment of the present application can execute the asteroid posture tracking method based on monocular vision and three-dimensional contours provided in the embodiment of the present application, and its implementation principle is similar. The actions performed by each module and unit in the asteroid posture tracking system based on monocular vision and three-dimensional contours in each embodiment of the present application correspond to the steps in the asteroid posture tracking method based on monocular vision and three-dimensional contours in each embodiment of the present application. For the detailed functional description of each module of the asteroid posture tracking system based on monocular vision and three-dimensional contours, please refer to the description of the corresponding asteroid posture tracking method based on monocular vision and three-dimensional contours shown in the previous text, which will not be repeated here.

[0124] Based on the same principles as the methods shown in the embodiments of this application, the embodiments of this application also provide an electronic device, which may include but is not limited to: a processor and a memory; the memory is configured to store a computer program; and the processor is configured to execute the asteroid pose tracking method based on monocular vision and 3D contours shown in any optional embodiment of this application by calling the computer program. Compared with the prior art, the asteroid pose tracking method based on monocular vision and 3D contours provided in this application extracts 3D contour segments from all 3D line segments on the surface of the asteroid's 3D model, determines the control points of the 3D contour segments, extracts image edge points from the asteroid's monocular image, and establishes a matching relationship between the control points and the image edge points. Then, with the goal of minimizing the angular difference between the back-projection line of the image edge points and the projection plane of the corresponding 3D contour segment, an objective function for the asteroid pose is established, and a maximum likelihood estimate of the asteroid pose is obtained by optimizing and solving this objective function. This pose optimization method is performed in 3D space and is robust to model scale, target distance variations, background noise, and occlusion. Finally, the maximum likelihood estimate is corrected using a second-order autoregressive cubature Kalman model to obtain the maximum a posteriori probability estimate of the asteroid, which greatly improves the tracking stability.

[0125] In an optional embodiment, an electronic device is also provided, such as Figure 4 As shown, Figure 4 The electronic device 400 shown may be a server, including a processor 401 and a memory 403. The processor 401 and the memory 403 are connected, for example, via a bus 402. Optionally, the electronic device 400 may further include a transceiver 404. It should be noted that in actual applications, the number of transceivers 404 is not limited to one, and the structure of the electronic device 400 does not constitute a limitation on the embodiments of the present application.

[0126] Processor 401 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 401 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0127] The bus 402 may include a path for transmitting information between the above components. The bus 402 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 402 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0128] The memory 403 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0129] The memory 403 is used to store application code for executing the solution of the present application, and the execution is controlled by the processor 401. The processor 401 is used to execute the application code stored in the memory 403 to implement the content shown in the above method embodiment.

[0130] Among them, electronic devices include but are not limited to: mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 4 The electronic device shown is merely an example and should not limit the functions and scope of use of the embodiments of the present application.

[0131] The server provided in this application can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. It can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The terminal can be a smartphone, tablet computer, laptop computer, desktop computer, smart speaker, smart watch, etc., but is not limited to these. The terminal and the server can be directly or indirectly connected through wired or wireless communication, and this application does not limit this.

[0132] An embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer-readable storage medium is run on a computer, the computer can execute the corresponding contents of the aforementioned method embodiment.

[0133] It should be understood that although the steps in the flowcharts of the accompanying drawings are shown in sequence as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the flowcharts of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0134] It should be noted that the computer-readable storage medium mentioned above in this application may also be a computer-readable signal medium or a combination of a computer-readable storage medium and a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof.

[0135] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0136] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device executes the method shown in the above embodiment.

[0137] According to one aspect of the present application, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to implement the asteroid pose tracking method and system based on monocular vision and three-dimensional profiles provided in the various optional implementations described above.

[0138] Computer program code for performing the operations of the present application may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0139] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0140] The modules described in the embodiments of this application may be implemented in software or hardware. In some cases, the module name does not limit the module itself. For example, the 3D contour extraction module may also be described as "a 3D contour extraction module for extracting 3D contour segments from all 3D line segments on the surface of a preset 3D asteroid model and determining the control points of the 3D contour segments."

[0141] The above description is merely a preferred embodiment of the present application and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this application is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but also encompasses other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the above-mentioned disclosed concepts. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A method for tracking an asteroid's posture based on monocular vision and three-dimensional contours, characterized in that: The method comprises: Extracting a three-dimensional contour segment from all three-dimensional line segments on the surface of a preset three-dimensional asteroid model, and determining control points of the three-dimensional contour segment; Extracting image edge points from the asteroid monocular image, and establishing a matching relationship between the control points and the image edge points; Establishing an objective function for the asteroid pose with the goal of minimizing the angle difference between the back-projection line of the image edge point and the projection plane of the corresponding three-dimensional contour segment; Optimizing and solving the objective function to obtain a maximum likelihood estimate of the asteroid's position and posture; The maximum likelihood estimate is corrected using a second-order autoregressive cubature Kalman model to obtain a maximum a posteriori probability estimate of the asteroid.

2. The asteroid posture tracking method based on monocular vision and three-dimensional profile according to claim 1, characterized in that: The step of extracting a three-dimensional contour segment from all three-dimensional line segments on the surface of a preset three-dimensional asteroid model and determining control points of the three-dimensional contour segment comprises: Project all three-dimensional line segments on the surface of the preset asteroid three-dimensional model onto the image plane to generate corresponding two-dimensional projection line segments; When a back-projection line of a midpoint of the two-dimensional projection line segment is tangent to the three-dimensional model of the asteroid, determining the corresponding three-dimensional line segment as a three-dimensional contour segment; A point on the three-dimensional contour segment corresponding to the midpoint of the two-dimensional projection line segment is determined as the control point.

3. The asteroid posture tracking method based on monocular vision and three-dimensional profile according to claim 1, characterized in that: The objective function of the asteroid pose is established with the goal of minimizing the angle difference between the back-projection line of the image edge point and the projection plane of the corresponding three-dimensional contour segment, including: Determining a back-projection line of the image edge point and a normal vector of the projection plane of the three-dimensional contour segment; Let α ε is the angle between the back-projection line and the normal vector, minimizing the angle difference between the back-projection line and the projection plane is equivalent to minimizing |cosα ε |; To minimize |cosα ε | is the goal, and the objective function of asteroid posture is established.

4. The asteroid posture tracking method based on monocular vision and three-dimensional profile according to claim 3, characterized in that: To minimize |cosα ε | is the goal, and the objective function of asteroid posture is established, including: Use M to estimate the pair |cosα ε |Optimize to minimize |cosα ε The sum of M estimators of |ρ is the target to establish the objective function: in, d ε is the normal vector of the projection plane of the 3D contour segment, u ε is the back projection line of the edge point of the image, σ is {cosα ε The initial pose p given by the volumetric Kalman model k 0 The median absolute deviation below .

5. The asteroid posture tracking method based on monocular vision and three-dimensional profile according to claim 4, characterized in that: The optimizing and solving the objective function to obtain a maximum likelihood estimate of the asteroid's position and posture includes: The objective function is solved by an iterative least squares algorithm to obtain a maximum likelihood estimate of the asteroid's position.

6. The asteroid posture tracking method based on monocular vision and three-dimensional profile according to claim 5, characterized in that: In the process of solving the objective function by the iterative least squares algorithm, the iterative formula used is: p (s+1) =p (s) -(J (s)T IN (s) J (s) +τI6) -1 J (s)T IN (s) v (s) Where, v (s) =[cosα1 (s) cosα2 (s) …cosα n (s) ] T ; s represents the number of iteration steps; τ is a non-negative robust factor coefficient; W (s) represents the unit matrix {…ω(cosα ε (s) / σ)…}, where ω(x) represents the weight function of the M estimator; J (s) Indicates v (s) P (s) The Jacobian matrix of .

7. The asteroid posture tracking method based on monocular vision and 3D profile according to claim 1, characterized in that: The method uses a second-order autoregressive cubature Kalman model to correct the maximum likelihood estimate to obtain a maximum a posteriori probability estimate of the asteroid, including: Define the state variable of the volumetric Kalman model as s k =[p k T p k-1 T ] T , where p -1 =p0; The state equation of the volumetric Kalman model is determined as: Oh k+1 =log SO(3) (R k R k-1 -1 R k )+C k t k+1 =R k R k-1 -1 (t k -t k-1 )+t k +Ψ k Oh k =Oh k +n k t k =t k +k k Where, T = exp(p ^ ) is the maximum likelihood estimate of the asteroid's position p k The rigid body motion matrix generated in log SO(3) (R) represents the rotation vector Ω obtained by inversely solving the rotation matrix R; Γ k ,Ψ k ,η k and κ k represents the prediction error vector; The observation equation for the volumetric Kalman model is: With k =[I6 O6]s k +ε k Where I6 represents the sixth-order unit matrix, O6 represents the sixth-order zero matrix, ε k represents the observation noise vector, z k is the observed quantity; The maximum a posteriori probability estimate of the asteroid is obtained by solving the volumetric Kalman model.

8. An asteroid posture tracking system based on monocular vision and three-dimensional contours, characterized in that: The system comprises: A 3D contour extraction module is used to extract 3D contour segments from all 3D line segments on the surface of a preset 3D asteroid model and determine the control points of the 3D contour segments; An image edge extraction module is used to extract image edge points from the asteroid monocular image and establish a matching relationship between the control points and the image edge points; A pose estimation module is configured to establish an objective function for the asteroid pose with the goal of minimizing the angle difference between the back-projection line of the image edge point and the projection plane of the corresponding three-dimensional contour segment; The pose estimation module is also used to optimize and solve the objective function to obtain a maximum likelihood estimate of the asteroid's pose; The posture correction module is used to correct the maximum likelihood estimate using a second-order autoregressive cubature Kalman model to obtain a maximum a posteriori probability estimate of the asteroid.

9. An electronic device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the method according to any one of claims 1 to 7 is implemented when the processor executes the program.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images