A method and system for assessing the overall confidence level of geometric shape fusion

Abstract

This invention provides a method and system for evaluating the overall confidence level of geometric shape fusion. The method includes: obtaining the spatial correlation between geometric shapes before and after fusion through clustering; adjusting the spatial correlation based on the relative positional relationship of the geometric shapes before and after fusion; establishing a corresponding probability distribution model based on the measurement errors of the geometric shapes before and after fusion; estimating the weight of each shape point before fusion based on the probability distribution model; fusing the confidence values ​​of each shape point according to the weights; and evaluating the overall confidence level of the fusion. This method focuses on solving the problem of how to estimate the overall error distribution after fusion algorithm processing and obtain the overall evaluation index, i.e., the overall confidence level, based on the evaluation index before fusion and the various elements of map fragments in the traditional mapping fusion process.

Description

Technical Field

[0001] This invention relates to the field of computer vision and mapping, and more specifically, to a method and system for evaluating the overall confidence level of geometric shape fusion. Background Technology

[0002] Relative position estimation of geometric shapes refers to quantifying the measurement error between various generated map elements, such as linear and polygonal elements, and existing ground truth values ​​based on the sample probability distribution of spatial relative positions after processing a large amount of local fragmented map data through the construction of real-world geometric maps and logical topology layers and big data fusion algorithms, using a large amount of input local fragmented map data. Confidence level refers to a theoretical and systematic method for evaluating the reliability of the relative position estimation error. This evaluation method provides timely feedback to the fusion algorithm in the existing cloud-based global mapping, serving as a key feedback on the positional accuracy of each element in the mapping algorithm. Since map construction requires the registration and fusion of a large number of geometric shapes, the confidence levels of the various production stages that provide error feedback cannot be effectively fused. Therefore, solving the fusion problem is not only about building a dynamic flow and interaction of confidence levels between modules, but also about systematically improving the accuracy of the entire mapping production process. Summary of the Invention

[0003] This invention addresses the technical problems existing in the prior art by providing a method and system for evaluating the overall confidence level of geometric shape fusion.

[0004] According to a first aspect of the present invention, a method for evaluating the overall confidence level of geometric shape fusion is provided, comprising:

[0005] Spatial relationships between geometric shapes before and after fusion are obtained through clustering;

[0006] The spatial relationship is adjusted based on the relative positional relationship of the geometric shapes before and after fusion;

[0007] Establish a corresponding probability distribution model based on the measurement errors of the geometric shapes before and after fusion;

[0008] The weights of each shape point before fusion are estimated based on the probability distribution model described above.

[0009] The confidence values ​​of each point are fused according to the weights, and the overall confidence value of the fusion is evaluated.

[0010] According to a second aspect of the present invention, a system for evaluating the overall confidence level of geometric shape fusion is provided, comprising:

[0011] The acquisition module is used to obtain the spatial relationship between the geometric shapes before and after fusion through clustering;

[0012] An adjustment module is used to adjust the spatial relationship based on the relative positional relationship of the geometric shapes before and after fusion;

[0013] A module is established to create a corresponding probability distribution model based on the measurement errors of the geometric shapes before and after fusion;

[0014] The estimation module is used to estimate the weight of each shape point before fusion based on the probability distribution model.

[0015] The evaluation module is used to evaluate the overall confidence value of the fused points based on the weighted fusion confidence values.

[0016] According to a third aspect of the present invention, an electronic device is provided, including a memory and a processor, the processor being configured to execute a computer management program stored in the memory to implement a method for evaluating the overall confidence level of geometric shape fusion.

[0017] According to a fourth aspect of the present invention, a computer-readable storage medium is provided having a computer management class program stored thereon, wherein the computer management class program, when executed by a processor, implements the steps of a method for evaluating the overall confidence level of geometric shape fusion.

[0018] This invention provides a method and system for evaluating the overall confidence level of geometric shape fusion. This method focuses on solving the problem of how to estimate the overall error distribution after fusion algorithm processing and obtain the overall evaluation index, i.e., the overall confidence level, based on the evaluation index before fusion and the various elements of map fragments in the traditional map fusion process. Attached Figure Description

[0019] Figure 1 A flowchart of an overall confidence assessment method for geometric shape fusion provided by the present invention;

[0020] Figure 2 A data flow diagram for a confidence fusion method;

[0021] Figure 3 This is a schematic diagram of the overall confidence assessment system for geometric shape fusion provided by the present invention;

[0022] Figure 4 A schematic diagram of the hardware structure of a possible electronic device provided by the present invention;

[0023] Figure 5 This is a schematic diagram of the hardware structure of a possible computer-readable storage medium provided by the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. In addition, the technical features of the various embodiments or individual embodiments provided by the present invention can be arbitrarily combined with each other to form feasible technical solutions. Such combinations are not constrained by the order of steps and / or structural composition patterns, but must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0025] Figure 1 A flowchart of a method for evaluating the overall confidence level of geometric shape fusion provided by this invention is shown below. Figure 1 As shown, the method includes:

[0026] S1 obtains the spatial relationship between the geometric shapes before and after fusion through clustering.

[0027] As an example, the step of obtaining the spatial relationship between geometric shapes before and after fusion by clustering includes: calculating the weighted distance between two points based on the three-dimensional coordinates of the points on the multiple geometric shapes before fusion and the three-dimensional coordinates of the points on the geometric shapes after fusion, and then clustering the points on the multiple geometric shapes before fusion and the geometric shapes after fusion to obtain the many-to-one spatial relationship between the geometric shapes before fusion and the geometric shapes after fusion.

[0028] Understandably, the 3D coordinates of the measured shape points before and after fusion are clustered using a distance correlation algorithm: shape points that have a generational relationship with the fused points are aligned in a many-to-one manner. Algorithm frameworks that can be used, but are not limited to, include DBSCAN / K-means, with the aim of associating and clustering linear features, areal features, etc., in space.

[0029] In the necessary distance calculation of the above clustering algorithm, the quaternion matrix between two points is used. Eigenvalue decomposition is used to solve for the average quaternion. Let:

[0030]

[0031] The average quaternion can be solved using a large number of existing matrix eigenvalue decomposition algorithms (such as Homotopy / MRRR, etc.).

[0032] Among them, by utilizing the three-dimensional coordinate system in the field of computer vision mapping, quaternions and angle measurements are introduced into the Euclidean distance calculation in the clustering process to establish a weighted distance:

[0033]

[0034] Where, x i With x j These represent the three-dimensional coordinates (x, y, z) of a point on the geometry before and after merging, respectively. The average quaternion is the quaternion matrix of the coordinates of two points: The eigenvectors obtained by eigenvalue decomposition. u is the difference between two ordinary quaternions, where q... k k∈{1,2} is the quaternion of point 1 and point 2. Finally, the l-2 norm is calculated to obtain the three dimensions of the feature vector, and then the new weighted distance can be obtained.

[0035] By calculating the weighted distance between the shape points on the geometric shape before fusion and the shape points on the geometric shape after fusion, the shape points on the geometric shape before fusion and the geometric shape after fusion are clustered, thereby obtaining the spatial relationship between the geometric shapes before and after fusion, that is, the correspondence between multiple geometric shapes before fusion and one geometric shape after fusion.

[0036] S2, adjust the spatial relationship based on the relative positional relationship of the geometric shapes before and after fusion.

[0037] Understandably, for geometric shapes within clusters, some spatially related shapes before and after merging may not be usable as subsequent data for various reasons. This invention introduces angle threshold filtering based on actual conditions:

[0038]

[0039] Where, q i q j These are the quaternions of the shape points on the geometry before and after fusion, respectively. When the angle is greater than a set threshold, the spatial association between the two geometry types is deleted; otherwise, the spatial association between the two geometry types with an angle less than the set threshold is retained. θ is a pre-set angle threshold that can be adjusted according to the actual situation. In this invention, the mapping standard of 10 degrees is used as the value of θ.

[0040] S3. Establish a corresponding probability distribution model based on the measurement errors of the geometric shapes before and after fusion.

[0041] Understandably, considering that the error during the mapping process is a positive distance scalar, a scalar-coefficientd Beta distribution is used, with the following probability density function:

[0042]

[0043] In the mapping application of this invention, but not limited to other cases, considering that the error is positive, the value of (a, c) is set to (0, max(d)). i The method of calculation for (α, β) is explained in the following implementation. Be(α, β) is a Beta function of (α, β), and its expression is:

[0044]

[0045] Where t is an auxiliary variable that does not actually occur in practice. This expression can be derived from its relationship with the Gamma function. The method for determining the values ​​of (α, β) will be explained in the subsequent implementation steps.

[0046] Specifically, considering the range of distance error (0, c), c = max(d i Substituting ) > 0 into the scalar coefficient-based Beta distribution of formula (4), we get:

[0047]

[0048] Here, Be(α, β) remains a Beta function. To construct the fusion probability model in the next step (step S4), (α, β) is further hyperparameterized using the density kernel method. Specifically, let α + β = τ and for the constant c, we set... Then our kernel density function k(x; μ, τ) is:

[0049]

[0050] Therefore, the fusion probability model f(x; G) can be rewritten as:

[0051] f(x;G)=∫k(x;μ,τ)dG(μ,τ),G~DP(α,G0)) (8);

[0052] Where DP(α, G0) is a Dirichlet distribution, α is its parameter, and its prior distribution is set as a Gamma distribution Ga(α, G0). α b α Where G0 is the fundamental distribution in the Dirichlet process, it is generally assumed to be independent, i.e.: G0(μ, τ) = G 01 (μ)G 02 (τ). Where G 01 With G 02 Let G be the independent prior distribution of (μ, τ). 01(μ)~Unif(0,max(d i )) is uniformly distributed, G 02 ~InvGa(α) τ ,β τ ) is an inverse Gamma distribution, and because (α) τ ,β τ Within a reasonable range, it is not sensitive to the initial value; here we let α... τ =2, This is the average error of the corresponding pre-fusion shape points among the current fused shape points.

[0053] S4. Estimate the weight of each point before fusion based on the probability distribution model.

[0054] In the fused map, since its confidence value is fused into a multi-class probability distribution fusion and is directly related to the inter-class error, the Dirichlet mixture model is adopted.

[0055] Specifically, following the fusion probability model f(x; G) in step S3, the specific Bayesian joint posterior distribution is derived. Suppose that a fused geometric shape is composed of K pre-fusion shapes, then the fusion distribution of the error X is:

[0056]

[0057] Where, k(x) i μ i , τ i Let φ be the kernel distribution function described in step S3. For ease of understanding, let φ be... i =(μ i , τ i For any i = 1, 2, ..., n, the posterior inference to be solved is:

[0058] p(G,φ,α,β|X)=p(G|φ,α,β)p(φ,α,β|X) (10);

[0059] Where G, φ, α, β|X is a typical Dirichlet process. The posterior distribution update process is illustrated here as an example, but is not limited to sampling using commonly used Bayesian algorithms, including Markov Monte Carlo (MCMC) or variational inferences. The Dirichlet process used for sampling G, φ, α, β|X is as follows: Among them G * basic distribution Dirac delta measures for μ and τ:

[0060]

[0061] This example uses the Metropolis-Hastings algorithm, but is not limited to other algorithms. First, samples are drawn from p(φ, α, β|X), and then the samples φ, α, β|X are used for discretization and truncation approximation. Discretization Recorded as Then the weighted sum of the Dirac delta measures is:

[0062]

[0063] ω jd The solution method is a typical stick-breaking method. Where ω 1d =z 1d Starting from j=2, then z md From Beta(1, α) d +n) sampling. Since the model is not sensitive to the initial values ​​here, a relatively small constant such as α can be chosen. d =2 is sufficient.

[0064] S5. Based on the weighted fusion of the confidence values ​​of each point, evaluate the overall confidence value of the fusion.

[0065] Understandably, the overall error distribution can be discretized using a weighting algorithm, i.e., by utilizing the predicted posterior distribution in the Dirichlet process:

[0066]

[0067] Using analogy, its probability accumulation function (CDF) can be obtained as follows: Where K(x; μ) jd , τ jd F can be obtained by integrating the kernel density function, which belongs to the probability accumulation function of the conventional scalar coefficient-based Beta distribution. d (x) represents the overall confidence value after the geometric shapes are merged.

[0068] See Figure 2 This is a schematic diagram of the overall confidence assessment method for geometric shape fusion provided by the present invention. Based on the three-dimensional coordinates of each point on the geometric shape before fusion and the three-dimensional related coordinates of each point on the geometric shape after fusion, a spatial relationship is formed between the geometric shapes before and after fusion. The spatial relationship is adjusted based on the angular sizes of the geometric shapes before and after fusion. Then, the weights of the points on the geometric shape before fusion are calculated using a Dirichlet mixture model, and the overall confidence value is calculated by fusing the confidence values ​​of each point using these weights.

[0069] Figure 3 A structural diagram of an overall confidence assessment system for geometric shape fusion provided in an embodiment of the present invention is shown below. Figure 3 As shown, a system for evaluating the overall confidence level of geometric shape fusion includes an acquisition module 301, an adjustment module 302, a building module 303, an estimation module 304, and an evaluation module 305, wherein:

[0070] The acquisition module 301 is used to obtain the spatial relationship between the geometric shapes before and after fusion through clustering;

[0071] The adjustment module 302 is used to adjust the spatial association relationship according to the relative positional relationship of the geometric shapes before and after fusion;

[0072] Module 303 is established to build a corresponding probability distribution model based on the measurement errors of the geometric shapes before and after fusion;

[0073] Estimation module 304 is used to estimate the weight of each shape point before fusion based on the probability distribution model;

[0074] Evaluation module 305 is used to evaluate the overall confidence value of the fusion based on the confidence values ​​of each point according to the weights.

[0075] It is understood that the overall confidence assessment system for geometric shape fusion provided by the present invention corresponds to the overall confidence assessment method for geometric shape fusion provided in the foregoing embodiments. The relevant technical features of the overall confidence assessment system for geometric shape fusion can be referred to the relevant technical features of the overall confidence assessment method for geometric shape fusion, and will not be repeated here.

[0076] Please see Figure 4 , Figure 4 This is a schematic diagram illustrating an embodiment of the electronic device provided in this invention. For example... Figure 4 As shown, an embodiment of the present invention provides an electronic device 400, including a memory 410, a processor 420, and a computer program 411 stored in the memory 410 and executable on the processor 420. When the processor 420 executes the computer program 411, it implements the steps of the overall confidence assessment method for geometric shape fusion.

[0077] Please see Figure 5 , Figure 5 This is a schematic diagram illustrating an embodiment of a computer-readable storage medium provided by the present invention. (See diagram below.) Figure 5 As shown, this embodiment provides a computer-readable storage medium 500 on which a computer program 511 is stored. When the computer program 511 is executed by a processor, it implements the steps of the overall confidence assessment method for geometric shape fusion.

[0078] This invention provides a method and system for evaluating the overall confidence of geometric shape fusion. By incorporating weights, it overcomes the negative impact of spatial index symmetry in traditional geometric shape clustering algorithms based on distance between shapes, and further simplifies computational complexity, thereby improving the efficiency of the overall evaluation algorithm and ensuring the high efficiency of high-precision map building algorithms. In subsequent weight calculations, a more accurate distribution density function is introduced for the support range of the distance error random variable, enhancing the rationality of the evaluation of high-precision map building algorithms and further improving the security of high-precision maps.

[0079] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0080] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0081] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0082] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0083] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0084] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0085] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A geometric fusion overall confidence evaluation method, applied to map construction, for quantifying the measurement error between the generated various map elements and the existing true value, characterized in that, The method comprises the following steps: obtaining the spatial correlation between the geometric shapes before and after fusion through clustering; adjusting the spatial correlation according to the relative position relationship between the geometric shapes before and after fusion; establishing a corresponding probability distribution model according to the measurement error of the geometric shapes before and after fusion; estimating the weight of each point before fusion based on the probability distribution model; evaluating the overall confidence value of fusion according to the confidence value of each point fused by weight; the step of establishing a corresponding probability distribution model according to the measurement error of the geometric shapes before and after fusion comprises: Considering the range of distance error values The probability density function of the scalarized Beta distribution is: ； wherein is the Beta function, which is expressed as: is the Beta function, which is expressed as: ； wherein is an auxiliary variable, this expression is based on the relation with the Gamma function is solved for Let Further, we use the density kernel method to parameterize it, which is: And for the normalized Let Then the kernel density function is: ； Fusion probability model For: ； wherein, is a Dirichlet distribution, is its parameter, and its prior distribution is set as a Gamma distribution , is the base distribution in the Dirichlet process, and is assumed to be independent, i.e., wherein and is their respective independent prior distributions, is a uniform distribution, is the maximum distance between the shape points before fusion and the shape points after fusion with spatial correlation, is an inverse Gamma distribution, because is not sensitive to the initial value within a reasonable range, and let , is the average of the errors of the corresponding shape points before fusion in the current shape points after fusion. the step of estimating the weight of each point before fusion based on the probability distribution model comprises: Fusion probability model The specific Bayesian joint posterior distribution is derived, assuming that a certain geometric object after fusion is composed of individual objects before fusion, then the fusion distribution of the error is ； where is the kernel distribution function, and let The posterior inference to be solved is then: ； For a typical Dirichlet Process, used for sampling The Dirichlet Process of where The base distribution , is The Dirac delta measure of ； First, samples are taken from for discretization denoted by The Dirac delta measure weighted sum is then​​ ； The weights of each point are determined using a typical stick-breaking method. ,from Beginning , Available from Sampling, constant ; the step of evaluating the overall confidence value of fusion according to the confidence value of each point fused by weight comprises: discretizing the overall error distribution, that is, obtaining the predictive posterior distribution in the Dirichlet process: ； Using the analogy method, the probability cumulative function (CDF) of the distribution is obtained as where is obtained by integrating the kernel density function, which is the probability cumulative function of the conventional scalar-coefficient Beta distribution, i.e., the overall confidence.

2. The method of claim 1, wherein, the step of obtaining the spatial correlation between the geometric shapes before and after fusion through clustering comprises: based on the three-dimensional coordinates of the points on the multiple geometric shapes before fusion and the three-dimensional coordinates of the points on the geometric shape after fusion, the weighted distance between two points is calculated, the points on the multiple geometric shapes before fusion and the points on the geometric shape after fusion are clustered, and then the one-to-many spatial correlation between the geometric shapes before fusion and the geometric shape after fusion is obtained.

3. The method of claim 2, wherein, the step of calculating the weighted distance between two points comprises: ； wherein, and are the three-dimensional coordinates of the shape points on the pre-fusion geometry and the post-fusion geometry, respectively , is the average quaternion, i.e. the eigenvector obtained by performing eigen decomposition on the quaternion matrix of the two points coordinates: and is the difference between the two ordinary quaternions, wherein is the quaternion of point 1 and point 2, and finally the three-dimensional norm of the eigenvector is calculated to obtain the new weighted distance; where the matrix of quaternions between two points is The average quaternion is solved by eigen decomposition, let: ； The matrix eigenvalue decomposition algorithm is used to solve .

4. The method of claim 1, wherein, the step of adjusting the spatial correlation according to the relative position relationship between the geometric shapes before and after fusion comprises: calculating the angle between the points on the geometric shapes before fusion and the points on the geometric shape after fusion with spatial correlation: ； wherein, , Qi and Qj are quaternions of the shape points i on the geometry before fusion and j on the geometry after fusion, respectively. when the angle is greater than a set threshold, the spatial correlation of the two geometric shapes is deleted.

5. A geometric fusion of confidence evaluation system for quantifying the measurement error between the generated various map elements and the existing true value when applied to map construction, characterized in that, The method comprises the following steps: an acquisition module is configured to obtain the spatial correlation between the geometric shapes before and after fusion through clustering; an adjustment module is configured to adjust the spatial correlation according to the relative position relationship between the geometric shapes before and after fusion; an establishment module is configured to establish a corresponding probability distribution model according to the measurement error of the geometric shapes before and after fusion; an estimation module is configured to estimate the weight of each point before fusion based on the probability distribution model; an evaluation module is configured to evaluate the overall confidence value of fusion according to the confidence value of each point fused by weight; the step of establishing a corresponding probability distribution model according to the measurement error of the geometric shapes before and after fusion comprises: Considering the range of distance error values The probability density function of the scalarized Beta distribution is: ； wherein is the Beta function, whose expression is: ​ ； where is an auxiliary variable, this expression is based on the relation with the Gamma function is solved for Let Further, we use the density kernel method to parameterize it, which is: And for the normalized Let Then the kernel density function is: ； Fusion probability model For: ； where, is Dirichlet distribution, is its parameter, and its prior distribution is set as Gamma distribution , is the base distribution in Dirichlet process, and it is assumed to be independent, i.e., where and are their respective independent prior distributions, is uniform distribution, is the maximum distance between the shape points before fusion and the shape points after fusion with spatial correlation, is inverse Gamma distribution, because the initial value is not sensitive within a reasonable range, and let , is the average of the errors of the corresponding shape points before fusion in the current shape points after fusion; the step of estimating the weight of each point before fusion based on the probability distribution model comprises: Fusion probability model based on The specific Bayesian joint posterior distribution is derived, assuming that a certain geometric object after fusion is composed of individual objects before fusion, then the fusion distribution of the error is: ； where is the kernel distribution function, and let The posterior inference to be solved is then ； For a typical Dirichlet Process, used for sampling The Dirichlet Process of where The base distribution , is The Dirac delta measure of ； First, samples are drawn from for the discretization of the truncated approximation denoted by The Dirac delta measure weighted sum is then​​ ； The weights of each point are determined using a typical stick-breaking method. ,from Beginning , Available from Sampling, constant ; the step of evaluating the overall confidence value of fusion according to the confidence value of each point fused by weight comprises: discretizing the overall error distribution, that is, obtaining the predictive posterior distribution in the Dirichlet process: ； Using the analogy method, the probability cumulative function (CDF) of the distribution is obtained as where is obtained by integrating the kernel density function and belongs to the probability cumulative function of the conventional scalar coefficient Beta distribution, i.e., the overall confidence.

6. An electronic device, comprising: The method comprises the following steps:

7. A computer-readable storage medium, characterized in that, a memory and a processor are provided, and the processor is configured to execute a computer management program stored in the memory to realize the steps of the overall confidence evaluation method of geometric shape fusion according to any one of claims 1-4. A computer management program is stored on the memory, and the computer management program is executed by the processor to realize the steps of the overall confidence evaluation method of geometric shape fusion according to any one of claims 1-4.

Images