Method and system for establishing Chinese chestnut germplasm 3D model

Through 3D scanning and data processing technology, a high-quality three-dimensional model of chestnut germplasm was generated, which solved the problem of difficulty in displaying chestnut germplasm in the existing technology, provided a detailed three-dimensional morphology and structure description, and provided research and application. Reliable data support.

CN118967972AInactive Publication Date: 2024-11-15SHANDONG INST OF POMOLOGY
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202410995414.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-24
Publication Date
2024-11-15
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The prior art is difficult to record and display chestnut germplasm through three-dimensional intuitive recording and display, and the text records are stiff, and two-dimensional pictures are difficult to show the full picture of the germplasm, and there is a lack of description of three-dimensional models.

Method used

A 3D scanner uses a 3D scanner to collect point cloud data from multiple angles, pre-process and remove noise and discrete points. Through micronization, differential mapping, fusion deduplication, gridization and texture enhancement, high-quality chestnut germplasm three-dimensional Model.

Benefits of technology

Accurate capture of the morphology and structure of chestnut germplasm is achieved, and a high-quality and realistic three-dimensional model is generated, providing a reliable data foundation for subsequent research and application.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118967972A_ABST
    Figure CN118967972A_ABST
Patent Text Reader

Abstract

The invention discloses a Chinese chestnut germplasm 3D model establishment method and system, and relates to the technical field of Chinese chestnut germplasm model.The method comprises the steps that multiple pieces of point cloud data of the same Chinese chestnut germplasm are collected and preprocessed, the preprocessed multiple pieces of point cloud data are subjected to infinitesimal processing, and a top point cloud model and a side point cloud model of the Chinese chestnut germplasm are extracted; carrying out monolayer mapping operation on infinitesimal elements in the top point cloud model and the side point cloud model based on differential mapping, and carrying out monolayer mapping on multiple layers of point clouds; and performing fusion duplicate removal, gridding and grid fairing on the processed top point cloud model and the processed side point cloud model, and performing texture enhancement to obtain a Chinese chestnut germplasm three-dimensional model. Through the steps of multi-angle scanning, detailed data processing and step-by-step optimization, the form, structure and details of the Chinese chestnut germplasm can be accurately captured, and finally the three-dimensional model with high quality and strong sense of reality is generated. And a comprehensive data basis is provided, and reliable support is provided for subsequent research and application.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of chestnut germplasm modeling, and more particularly to a method and system for establishing a 3D chestnut germplasm model. Background Art

[0002] At present, the records and descriptions of chestnut germplasm resources are still mainly in the form of text or text + pictures, lacking a three-dimensional and intuitive description method. Among them, the text records are relatively rigid and stereotyped, and it is difficult to show the true form of the important organs of chestnut. The picture records are more intuitive, but they are only two-dimensional pictures, which are difficult to show the full picture of the germplasm.

[0003] In recent years, with the continuous development of laser scanning technology and computer technology, computer graphics three-dimensional real-time point cloud models have been widely used in many fields.

[0004] Therefore, how to more intuitively restore and display the original appearance of chestnut germplasm based on 3D models is an urgent problem that technicians in this field need to solve. Summary of the invention

[0005] In view of this, the present invention provides a method and system for establishing a 3D model of chestnut germplasm to solve the problems existing in the background technology.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] On the one hand, a method for establishing a 3D model of chestnut germplasm is provided, comprising:

[0008] Use 3D scanner to scan multiple times from different angles to obtain multiple point cloud data of the same chestnut germplasm;

[0009] Preprocess the collected point cloud data, including removing discrete points and restoring noise points to the correct position;

[0010] The preprocessed multiple point cloud data are microelemented to extract the top point cloud model and the side point cloud model of the chestnut germplasm, wherein the side point cloud model refers to the remaining point cloud excluding the top point cloud model;

[0011] Based on differential mapping, the microelements in the top point cloud model and the side point cloud model are respectively subjected to single-layer mapping operations to convert the multi-layer point cloud into a single layer;

[0012] The processed top point cloud model and side point cloud model are fused, deduplicated, meshed, mesh smoothed, and texture enhanced to obtain a three-dimensional model of chestnut germplasm.

[0013] Optionally, the preprocessing includes:

[0014] First, local division is performed to divide each point cloud data into multiple local patches. Each local patch is centered on a central point and contains all points within a range of r from the center.

[0015] Secondly, each local patch is input into the quaternion space transfer network to learn a four-dimensional vector. Through this four-dimensional vector, the original point cloud is transferred to a new posture state, and then the dimension of the posture state is changed through a fully connected layer. The multi-layer perceptron network is used to extract the features of each point cloud, and the feature vector of each point cloud is inversely transformed to restore it to the original posture. After obtaining the processed point-by-point feature vector, the feature vector of each point in a patch is fused to obtain the feature vector of a local patch.

[0016] Finally, the fused feature vector is regressed using several fully connected layers to obtain the final result.

[0017] Optionally, the extraction includes: extracting the highest position point in each point cloud data, extracting a local patch where the highest position point is higher than a position threshold, thereby constituting a top point cloud model of the chestnut germplasm.

[0018] Optionally, the differential mapping specifically includes:

[0019] Find the 8 micro-elements adjacent to the current micro-element.

[0020] Extract the highest position points of 8 microelements and fit a plane based on the highest position points;

[0021] Map all point clouds in the current microelement onto the fitted plane.

[0022] Optionally, it also includes repairing holes of micro-elements in the top point cloud model and the side point cloud model based on a threshold differential repair method, specifically:

[0023] Detect whether there are holes in the microelement and the number of point clouds in the microelement. If the number of point clouds is lower than the threshold, it is determined that there are holes in the microelement.

[0024] Point clouds are uniformly inserted into micro-elements with holes until the number of point clouds in the micro-elements reaches a data threshold.

[0025] Optionally, the specific process of obtaining the three-dimensional model of chestnut germplasm is:

[0026] The processed top point cloud model and the side point cloud model are fused and deduplicated to form a fused chestnut germplasm three-dimensional point cloud model;

[0027] Performing Poisson reconstruction on the fused three-dimensional point cloud model of chestnut germplasm to generate a grid model of chestnut germplasm;

[0028] Performing grid smoothing on the chestnut germplasm grid model;

[0029] The smoothed chestnut germplasm mesh model is texture enhanced to obtain an optimized three-dimensional model of the chestnut germplasm.

[0030] Optionally, the preprocessing further includes:

[0031] Normal calculation is performed on multiple point cloud data of the same chestnut germplasm: a KD tree is generated based on the point cloud to establish the topological relationship of the point cloud; the neighborhood of any point cloud is found, and the neighborhood points of the point cloud are reduced to a two-dimensional plane by principal component analysis, and the normal of the two-dimensional plane is set as the normal of the point cloud, and then the normal of the point cloud is oriented according to the position of the 3D scanner to complete the normal calculation;

[0032] The distance between the point cloud data and each point in the neighborhood is traversed and averaged, and it is determined whether the average exceeds a preset threshold. If so, the point cloud is determined to be a noise point and removed.

[0033] On the other hand, a system for establishing a 3D model of chestnut germplasm is provided, comprising:

[0034] The point cloud data acquisition module uses a 3D scanner to perform multiple scans from different angles to obtain multiple point cloud data for the same chestnut germplasm;

[0035] The data preprocessing module preprocesses the collected multiple point cloud data, including removing discrete points and restoring noise points to the correct position;

[0036] The model extraction module performs microelementization on the preprocessed multiple point cloud data, and extracts the top point cloud model and the side point cloud model of the chestnut germplasm. The side point cloud model refers to the remaining point cloud excluding the top point cloud model.

[0037] The model processing module performs single-layer mapping operations on the microelements in the top point cloud model and the side point cloud model based on differential mapping, and converts the multi-layer point cloud into a single layer;

[0038] The model output module fuses and removes duplicates, meshes, and smoothes the processed top point cloud model and side point cloud model, and performs texture enhancement to obtain a three-dimensional model of chestnut germplasm.

[0039] Through the above technical scheme, it can be known that compared with the prior art, the present invention discloses a method and system for establishing a 3D model of chestnut germplasm, using a 3D scanner to perform multiple scans from multiple different angles to obtain multiple point cloud data for the same chestnut germplasm; preprocessing the collected multiple point cloud data, including removing discrete points and restoring noise points to the correct position; micro-elementing the preprocessed multiple point cloud data, extracting the top point cloud model and the side point cloud model of the chestnut germplasm, the side point cloud model refers to the remaining point cloud except the top point cloud model; based on differential mapping, the micro-element in the top point cloud model and the side point cloud model is respectively single-layered, and the multi-layer point cloud is single-layered; the processed top point cloud model and the side point cloud model are fused, deduplicated, gridded, and grid-smoothed, and texture enhanced to obtain a three-dimensional model of chestnut germplasm. The present invention can accurately capture the morphology, structure and details of chestnut germplasm through multi-angle scanning, meticulous data processing and step-by-step optimization steps, and finally generate a high-quality, realistic three-dimensional model. It also provides a comprehensive data foundation, providing reliable support for subsequent research and applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0041] Figure 1 A flow chart of the method provided by the present invention;

[0042] Figure 2 This is a schematic diagram of the system structure provided by the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] On the one hand, the embodiment of the present invention discloses a method for establishing a 3D model of chestnut germplasm, such as Figure 1 As shown, including:

[0045] S1. Use a 3D scanner to perform multiple scans from multiple different angles to obtain multiple point cloud data for the same chestnut germplasm;

[0046] In the actual scanning scenario, due to the limited scanning angle of the scanner itself, a single scan cannot obtain the complete point cloud data of the chestnut germplasm surface. Multiple scanning stations are required to perform multiple scans at multiple different angles, so that multiple point cloud data for the same chestnut germplasm are obtained; that is, for each chestnut germplasm, leaves, fruit branches, thorn buds, and nuts are collected, and multiple point cloud data for the same chestnut germplasm are obtained using a 3D scanner.

[0047] S2. Preprocessing the collected multiple point cloud data, including removing discrete points and restoring noise points to the correct position;

[0048] In the original process of point cloud collection, due to the influence of the collection environment (light, water vapor in the air, equipment jitter), scanning equipment (equipment failure, low scanning accuracy) and chestnut germplasm (small bumps on the surface), there are usually some noise points in the scanned point cloud. These noise points not only increase the data volume of the point cloud, but also affect the accuracy of the subsequent point cloud processing steps, thus affecting the final modeling effect. Therefore, the first step of the processing work is to remove these noise points.

[0049] The initial input of this method is a frame of point cloud data. Before the point cloud data is input into the network, it needs to be locally divided first, dividing the entire point cloud into multiple local patches. Each local patch is centered on a central point and contains all points within a range of r from the center. In this way, the point cloud data can be divided into multiple smaller blocks, which is convenient for network learning.

[0050] Next, the quaternion space transfer network is used to learn a four-dimensional vector, through which the original point cloud is transferred to a new posture state to ensure that the model is independent of affine changes in a specific space. Then, a multi-layer perceptron network is used to extract the features of each point, and a similar inverse transformation is performed on the feature vector of each point to reach the original posture. In order to ensure that the model can adapt to the disorder of the point cloud data, a symmetric function is used to fuse the features of each point and merge the features of each point into a feature vector of a local patch.

[0051] After obtaining the fusion result, several fully connected layers are used to process the fused local feature vectors to obtain the final result of the network output. It should be noted that since point cloud data often contains outliers and noise points, the number of channels in the last regression layer can be changed to adapt to the required output size and enhance the robustness of the model.

[0052] The points to be removed by the present invention include two types: one is called outlier points, which are points far away from the main part of the point cloud and can be directly removed; the other is called noise points, which are actually formed by the deviation of the correct points. Therefore, removing them does not mean deleting them, but applying an inverse deviation vector to them to restore them to the correct position. The whole denoising process:

[0053]

[0054] in Represents the point cloud result obtained after denoising. represents the set of outliers, P′ represents the original noise point cloud, and p i ′ represents a point in the original noise point cloud, d i Represents the offset vector of each point learned. The specific process is as follows:

[0055] (1) Local division. In order to grasp the local features of point cloud data, it is necessary to divide the entire point cloud into multiple patches. A local patch is p i Point is the center, including all points within the range of r from the center, denoted as p i r , assuming there are M points.

[0056] (2) Point-by-point feature extraction. The initial input is a set of all point cloud data for a frame. Each point cloud is represented by a 1-dimensional vector, where N represents the number of points and 3 corresponds to the XYZ coordinates. First, the data is divided into n patches according to the previous step. Each time a patch is input, a pre-trained quaternion space transfer network (STN1) is used to output the parameters of a 4D vector (rotation) of pose transformation: Through this operation, the model is guaranteed to be independent of affine changes in a specific space, and the original point cloud is migrated to a new pose state that is conducive to network learning: Then, a fully connected layer (FNN1) is used to change the dimension for feature extraction, and then a multi-layer perceptron network is used to extract and process the features of each point: g = MLP (p i r ′), and finally, a similar inverse transformation is performed on the feature vector point by point to reach the original position, so a four-element space transfer network (STN2) and a fully connected layer (FNN2) are also introduced: h = (FNN2 · STN2) (g), note that there are M g and h here, because they are p i r The eigenvectors of the M points in , but the dimensions of g and h are different.

[0057] (3) Feature fusion. After obtaining the processed point-by-point feature vectors, the network sums the feature vectors of each point in a patch (i.e., implemented through a symmetric function), and thus merges them into a feature vector of a patch. Regardless of the order of the input points, the commutative law of addition ensures that the merged feature vectors are the same, which ensures that the model can adapt to the disorder of point cloud data, and at the same time it retains more information: H has the same dimension as h.

[0058] (4) Regression module. After obtaining the result of feature fusion, the fused feature vector is processed by several fully connected layers (FNN3) to obtain the final result of the network output. The number of channels of the last regression layer (i.e., fully connected layer) can be changed to adapt to the required output size. For outlier determination, the number of channels of the regression layer is set to 1, indicating the probability that the point is an outlier. The process description function g is You can set the threshold o by yourself, that is, o i >o, determine p i ′ is an outlier point, which is removed, and the set of all outliers is recorded as For noise point determination, the number of channels in the regression layer is set to 3, which indicates the inverse offset vector to be applied to the point. The process description function f is The corrected point is

[0059] S3, performing microelementization on the preprocessed multiple point cloud data, extracting the top point cloud model and the side point cloud model of the chestnut germplasm, wherein the side point cloud model refers to the remaining point cloud excluding the top point cloud model;

[0060] S4, based on differential mapping, performing single-layer mapping operations on the microelements in the top point cloud model and the side point cloud model respectively, and converting the multi-layer point cloud into a single layer;

[0061] S5. The processed top point cloud model and the side point cloud model are fused, deduplicated, meshed, meshed, and smoothed, and texture enhanced to obtain a three-dimensional model of the chestnut germplasm.

[0062] In a specific embodiment, the extraction includes: extracting the highest position point in each point cloud data, extracting the local patch where the highest position point is higher than the position threshold, that is, constituting the top point cloud model of the chestnut germplasm.

[0063] In a specific embodiment, the differential mapping specifically includes:

[0064] Find the 8 micro-elements adjacent to the current micro-element.

[0065] Extract the highest position points of 8 microelements and fit a plane based on the highest position points;

[0066] Map all point clouds in the current microelement onto the fitted plane.

[0067] In a specific embodiment, it also includes repairing holes of micro-elements in the top point cloud model and the side point cloud model respectively based on a threshold differential repair method, specifically:

[0068] Detect whether there are holes in the microelement and the number of point clouds in the microelement. If the number of point clouds is lower than the threshold, it is determined that there are holes in the microelement.

[0069] Point clouds are uniformly inserted into micro-elements with holes until the number of point clouds in the micro-element reaches the data threshold.

[0070] In a specific embodiment, the specific process of obtaining the three-dimensional model of chestnut germplasm is as follows:

[0071] The processed top point cloud model and the side point cloud model are fused and deduplicated to form a fused chestnut germplasm three-dimensional point cloud model;

[0072] Perform Poisson reconstruction on the fused chestnut germplasm 3D point cloud model to generate a chestnut germplasm grid model;

[0073] Specifically, the method includes: using an octree structure to store point sets, defining an octree according to the position of the sampling point set, and then subdividing the octree so that each sampling point falls on a leaf node with a depth of D; setting a spatial function F for each node of the octree, and the linear sum of all node functions F can represent a vector field V, and the basis function F uses an n-dimensional convolution of a box filter; in the case of uniform sampling, assuming that the divided blocks are constants, using cubic strip interpolation to approximate the gradient of the indicator function through the vector field V; using the Laplace matrix iteration to find the solution to the Poisson equation; finally, estimating the position of the sampling point, and then using its average value to extract the isosurface, and then using the marching cube algorithm to obtain the isosurface.

[0074] Grid smoothing was performed on the chestnut germplasm grid model;

[0075] Specifically, the modulus of the Laplacian coordinate δ of the reconstructed mesh model is reduced, the direction remains unchanged, and the Laplacian mesh reconstruction is performed to smooth the mesh.

[0076] The smoothed chestnut germplasm mesh model was texture enhanced to obtain an optimized three-dimensional model of the chestnut germplasm.

[0077] In a specific embodiment, the preprocessing may further include another method:

[0078] Normal calculation was performed on multiple point cloud data of the same chestnut germplasm: a KD tree was generated based on the point cloud to establish the topological relationship of the point cloud; the neighborhood of any point cloud was found, and the principal component analysis was performed on the neighborhood points of the point cloud to reduce the dimension to a two-dimensional plane, and the normal of the two-dimensional plane was set as the normal of the point cloud. The normal of the point cloud was then oriented according to the position of the 3D scanner to complete the normal calculation;

[0079] The distance between the point cloud data and each point in the neighborhood is traversed and averaged to determine whether the average exceeds the preset threshold. If so, the point cloud is determined to be a noise point and removed.

[0080] On the other hand, a system for establishing a 3D model of chestnut germplasm is provided, such as Figure 2 As shown, including:

[0081] The point cloud data acquisition module uses a 3D scanner to perform multiple scans from different angles to obtain multiple point cloud data for the same chestnut germplasm;

[0082] The data preprocessing module preprocesses the collected multiple point cloud data, including removing discrete points and restoring noise points to the correct position;

[0083] The model extraction module performs microelementization on the preprocessed multiple point cloud data, and extracts the top point cloud model and the side point cloud model of the chestnut germplasm. The side point cloud model refers to the remaining point cloud excluding the top point cloud model.

[0084] The model processing module performs single-layer mapping operations on the microelements in the top point cloud model and the side point cloud model based on differential mapping, and converts the multi-layer point cloud into a single layer;

[0085] The model output module fuses and removes duplicates, meshes, and smoothes the processed top point cloud model and side point cloud model, and performs texture enhancement to obtain a three-dimensional model of chestnut germplasm.

[0086] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0087] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for establishing a 3D model of chestnut germplasm, characterized in that: include: Use 3D scanner to scan multiple times from different angles to obtain multiple point cloud data of the same chestnut germplasm; Preprocess the collected point cloud data, including removing discrete points and restoring noise points to the correct position; The preprocessed multiple point cloud data are microelemented to extract the top point cloud model and the side point cloud model of the chestnut germplasm, wherein the side point cloud model refers to the remaining point cloud excluding the top point cloud model; Based on differential mapping, the microelements in the top point cloud model and the side point cloud model are respectively subjected to single-layer mapping operations to convert the multi-layer point cloud into a single layer; The processed top point cloud model and side point cloud model are fused, deduplicated, meshed, mesh smoothed, and texture enhanced to obtain a three-dimensional model of chestnut germplasm.

2. The method for establishing a 3D model of chestnut germplasm according to claim 1, characterized in that: The pre-processing comprises: First, local division is performed to divide each point cloud data into multiple local patches. Each local patch is centered on a central point and contains all points within a range of r from the center. Secondly, each local patch is input into the quaternion space transfer network to learn a four-dimensional vector. Through this four-dimensional vector, the original point cloud is transferred to a new posture state, and then the dimension of the posture state is changed through a fully connected layer. The multi-layer perceptron network is used to extract the features of each point cloud, and the feature vector of each point cloud is inversely transformed to restore it to the original posture. After obtaining the processed point-by-point feature vector, the feature vector of each point in a patch is fused to obtain the feature vector of a local patch. Finally, the fused feature vector is regressed using several fully connected layers to obtain the final result.

3. The method for establishing a 3D model of chestnut germplasm according to claim 2, characterized in that: The extraction includes: extracting the highest position point in each point cloud data, extracting the local patch where the highest position point is higher than the position threshold, that is, constituting the top point cloud model of the chestnut germplasm.

4. The method for establishing a 3D model of chestnut germplasm according to claim 1, characterized in that: The differential mapping specifically includes: Find the 8 micro-elements adjacent to the current micro-element. Extract the highest position points of 8 microelements and fit a plane based on the highest position points; Map all point clouds in the current microelement onto the fitted plane.

5. The method for establishing a 3D model of chestnut germplasm according to claim 1, characterized in that: It also includes repairing holes of micro-elements in the top point cloud model and the side point cloud model based on the threshold differential repair method, specifically: Detect whether there are holes in the micro-element and the number of point clouds in the micro-element. If the number of point clouds is lower than the threshold, it is determined that there are holes in the micro-element. Point clouds are uniformly inserted into micro-elements with holes until the number of point clouds in the micro-elements reaches a data threshold.

6. The method for establishing a 3D model of chestnut germplasm according to claim 1, characterized in that: The specific process of obtaining the three-dimensional model of chestnut germplasm is as follows: The processed top point cloud model and the side point cloud model are fused and deduplicated to form a fused chestnut germplasm three-dimensional point cloud model; Performing Poisson reconstruction on the fused three-dimensional point cloud model of chestnut germplasm to generate a grid model of chestnut germplasm; Performing grid smoothing on the chestnut germplasm grid model; The smoothed chestnut germplasm mesh model is texture enhanced to obtain an optimized three-dimensional model of the chestnut germplasm.

7. The method for establishing a 3D model of chestnut germplasm according to claim 1, characterized in that: The pre-processing further comprises: Normal calculation is performed on multiple point cloud data of the same chestnut germplasm: a KD tree is generated based on the point cloud to establish the topological relationship of the point cloud; the neighborhood of any point cloud is found, and the neighborhood points of the point cloud are reduced to a two-dimensional plane by principal component analysis, and the normal of the two-dimensional plane is set as the normal of the point cloud, and then the normal of the point cloud is oriented according to the position of the 3D scanner to complete the normal calculation; The distance between the point cloud data and each point in the neighborhood is traversed and averaged, and it is determined whether the average exceeds a preset threshold. If so, the point cloud is determined to be a noise point and removed.

8. A system for establishing a 3D model of chestnut germplasm, characterized in that: A method for establishing a 3D model of chestnut germplasm according to any one of claims 1 to 7, comprising: The point cloud data acquisition module uses a 3D scanner to perform multiple scans from different angles to obtain multiple point cloud data for the same chestnut germplasm; The data preprocessing module preprocesses the collected multiple point cloud data, including removing discrete points and restoring noise points to the correct position; The model extraction module performs microelementization on the preprocessed multiple point cloud data, and extracts the top point cloud model and the side point cloud model of the chestnut germplasm. The side point cloud model refers to the remaining point cloud excluding the top point cloud model. The model processing module performs single-layer mapping operations on the microelements in the top point cloud model and the side point cloud model based on differential mapping, and converts the multi-layer point cloud into a single layer; The model output module fuses and removes duplicates, meshes, and smoothes the processed top point cloud model and side point cloud model, and performs texture enhancement to obtain a three-dimensional model of the chestnut germplasm.

Citation Information

Patent Citations

  • Scattered image based crop fruit three-dimensional point cloud extracting system

    CN108198230A

  • A complex automobile point cloud model reconstruction method

    CN109840938A

  • Three-dimensional face reconstruction method and system

    CN110363858A

  • Deep learning-based point cloud data three-dimensional reconstruction method and system

    CN116258835A