Multi-source fusion point cloud data driven three-dimensional streetscape generation and optimization method

Abstract

The invention discloses a multi-source fusion point cloud data driven three-dimensional streetscape generation and optimization method, which relates to the technical field of three-dimensional streetscape generation, and comprises the following steps of: according to a space range of input large-range point cloud data, combining a three-dimensional coordinate system distribution characteristic of the input large-range point cloud data; dividing the whole registration area into a plurality of uniformly distributed subunit areas with the same area by adopting a structured space division algorithm; for each subunit region, geometric structure features of internal point clouds of the subunit region are obtained, and a geometric feature descriptor containing multiple dimensions is constructed; according to the method, subunit structure complexity identification is realized through a heterogeneity evaluation mechanism based on geometric features and a support vector machine model, the upper limit of the number of matching points is dynamically regulated and controlled, the registration precision of a high-heterogeneous region is improved, and the mismatching risk of a low-heterogeneous region is reduced, so that the structure restoration and continuity of a streetscape model are optimized, and the robustness of the streetscape model is improved. And the stability and intelligence of the three-dimensional modeling system are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of three-dimensional street view generation, and particularly to a method for generating and optimizing a three-dimensional street view driven by multi-source fusion point cloud data. Background Art

[0002] The three-dimensional street view driven by multi-source fusion point cloud data refers to integrating the point cloud information generated from data sources of various sensors (such as lidar, stereo cameras, unmanned aerial vehicle aerial survey systems, mobile measurement platforms, etc.), and using data fusion technology to perform unified registration, accuracy improvement, and structural complementation processing on spatial information from different sources, with different accuracies, and different perspectives, so as to construct a three-dimensional street view model with high accuracy and high restoration degree. This method not only improves the integrity and detail expressiveness of three-dimensional modeling, but also enables more realistic and data-driven street view simulation and scene reconstruction in the fields of smart cities, autonomous driving, security monitoring, etc., effectively supporting the in-depth application of urban-level spatial information.

[0003] The existing technologies have the following deficiencies: During the unified registration process of large-scale point cloud data, due to significant differences in the spatial expressions of regional data from different sources, different accuracies, and different perspectives, the local geometric heterogeneity levels of each region vary greatly. Existing registration technologies usually adopt a unified parameter strategy and cannot intelligently adjust the selection strategy of matching points according to the point cloud structure complexity of different regions. In this context, if the number of selected matching points for high-heterogeneity regions with rich geometric structures is insufficient, key feature information cannot be effectively extracted and utilized, resulting in a lack of fine-structure constraints in the registration process, prone to local rigid transformation errors, and further spreading to other regions through the global optimization process, ultimately leading to problems such as misaligned building contours, distorted structural boundaries, and distorted street view models, seriously affecting the accuracy of navigation path inference, spatial recognition, and engineering analysis; conversely, if the number of selected matching points for low-heterogeneity regions with simple geometric structures is excessive, redundant matching pairs or mismatching relationships are likely to be introduced, forming interfering registration-driven solutions, thereby causing non-structural geometric offsets and superimposing errors in multi-region fusion, resulting in a decline in the spatial consistency of the overall point cloud model and topological structure breakage, seriously restricting the stability and reliability of subsequent high-precision surveying and mapping modeling and intelligent recognition tasks.

[0004] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure, and thus it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The object of the present invention is to provide a multi-source fusion point cloud data-driven three-dimensional street view generation and optimization method. By introducing a heterogeneity evaluation mechanism constructed based on the geometric features of sub-unit regions and combining with a trained support vector machine model, it realizes the quantitative recognition and classification judgment of the local structure complexity, and dynamically adjusts the upper limit of the number of matching points accordingly, so that high-heterogeneity regions obtain more geometric constraints, and low-heterogeneity regions avoid redundant matching interference, thereby effectively avoiding the registration error diffusion and structural misalignment problems caused by the unified parameter strategy, improving the overall quality and stability of the street view model in terms of morphological restoration, boundary alignment, and topological continuity, and providing a more robust, efficient, and intelligent technical support for application scenarios such as three-dimensional urban reconstruction, autonomous driving map construction, and precision mapping, so as to solve the problems in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions: A multi-source fusion point cloud data-driven three-dimensional street view generation and optimization method, comprising the following steps: According to the spatial range of the input large-scale point cloud data and combining its three-dimensional coordinate system distribution characteristics, a structured spatial division algorithm is used to divide the overall registration area into multiple sub-unit regions with equal area and uniform distribution, and the wide-area point cloud with non-uniform geometric features is disassembled into cells that can be locally modeled and independently analyzed; For each sub-unit region, obtain the geometric structure features of the internal point cloud, construct a geometric feature descriptor including multiple dimensions, and provide multi-source input variables for subsequent heterogeneity analysis; Through feature engineering techniques, key indicators reflecting the high heterogeneity of the sub-unit region are extracted from the obtained geometric structure features. After in-depth analysis of the extracted key indicators, the geometric heterogeneity level of each sub-unit region is quantified based on the analysis results; The analyzed indicators are used as feature vectors and input into a support vector machine model that has been trained in advance with a large number of labeled samples to intelligently predict the degree of geometric heterogeneity corresponding to the sub-unit region; Based on the degree of heterogeneity of the sub-unit region predicted by the support vector machine model, an adaptive mapping relationship between the degree of heterogeneity and the registration parameters is constructed, and the corresponding upper limit of the number of matching points is dynamically allocated for sub-unit regions with different heterogeneity levels to achieve region-specific matching point numbers; After setting the range of the number of matching points based on the regulation strategy, the matching point control parameter and other registration parameters (such as matching radius, tolerance threshold) are jointly input into the subsequent point cloud registration module to achieve the adaptive registration of the sub-unit region.

[0007] Preferably, the overall registration area is divided into multiple sub-unit areas with equal areas and uniform distributions. The specific division strategy adopted is a recursive space segmentation method based on octree, a fixed-resolution grid division (VoxelGrid), or a region-growing type of space adaptive segmentation algorithm.

[0008] Preferably, the overall registration area is divided into multiple sub-unit areas with equal areas and uniform distributions by using the recursive space segmentation method based on octree. The specific steps are as follows: First, an overall bounding cube is established for the input large-range point cloud data, which is used as the root node to represent the entire spatial region. Then, the cube is recursively divided into eight equal parts according to the three-dimensional coordinate range (X, Y, Z) of the point cloud. Each division evenly divides the parent node space into eight sub-cube regions, which are respectively added to the octree structure as child nodes. In each division stage, it is judged whether the number of point clouds in each sub-cube meets the minimum segmentation threshold. If it exceeds the threshold, continue to recursively subdivide until the preset minimum voxel size or the lower limit condition of the number of points is met. The finally generated leaf nodes are multiple sub-unit areas with equal areas (volumes) and uniform spatial distributions. These areas not only retain the spatial continuity of the original point cloud but also provide a structured spatial basis for subsequent local geometric feature extraction and adaptive registration strategies.

[0009] Preferably, key indicators reflecting the high heterogeneity of the sub-unit area are extracted from the obtained geometric structure features through feature engineering techniques. Among them, the extracted key indicators include the degree of dispersion of the distribution of the principal curvature values of all points in the sub-unit area and the degree of spatial gradient change of the point density within the sub-unit area. After in-depth analysis of the extracted key indicators, a curvature dispersion reference value and a density gradient reference value are respectively generated, and the geometric heterogeneity level of each sub-unit area is quantified based on the curvature dispersion reference value and the density gradient reference value.

[0010] Preferably, the analyzed curvature dispersion reference value and density gradient reference value are used as feature vectors and input into a support vector machine model that has been trained in advance with a large number of labeled samples. Through the support vector machine model, a heterogeneity level coefficient is generated, and the geometric heterogeneity degree corresponding to the sub-unit area is intelligently predicted based on the heterogeneity level coefficient.

[0011] Preferably, a corresponding range of the number of matching points is dynamically allocated for sub-unit areas with different heterogeneity levels. The specific steps are as follows: Obtain the heterogeneity level coefficient of each sub-unit area predicted by the support vector machine model, and construct a matching point adjustment factor for controlling the allocation intensity of the number of matching points based on the heterogeneity level coefficient. The matching point adjustment factor is calculated in the form of exponential mapping, and the calculation expression is: , where: represents the matching point adjustment factor driven by heterogeneity, and its value range is in ; is the maximum amplification coefficient, which controls the upper limit of the matching point adjustment factor and is used to adjust the resource inclination degree of the high-heterogeneity area, ; is the non-linear response adjustment factor, which is used to control the response speed of the matching point adjustment factor curve to the heterogeneity level coefficient; is the heterogeneity level coefficient of this sub-unit area; Based on the constructed matching point adjustment factor , dynamically allocate the upper limit of the final number of matching points for this sub-unit area. The dynamic allocation expression is as follows: , where: represents the upper limit of the number of matching points allocated to the current sub-unit area; represents the minimum lower limit of the number of matching points set by the system for the low-heterogeneity area (such as flat ground); represents the maximum upper limit of the number of matching points, which is applicable to the high-heterogeneity area (such as complex buildings or edge mixing areas).

[0012] Preferably, the specific steps for generating the curvature dispersion reference value by deeply analyzing the dispersion degree of the main curvature value distribution of all points in the sub-unit area are as follows: First, for all points in the sub-unit area, use the k-nearest neighbor method (k is generally 10-30) to calculate the main curvature value of each point , and sort the main curvatures of all points in this sub-unit area in spatial order or density gradient to construct a curvature sequence . Then, introduce the relative jump ratio factor, which is defined as the logarithmic ratio form of the change between adjacent curvature points, and is used to highlight the non-linear mutation characteristics between points with large jumps. The calculation expression of the relative jump ratio factor is: , where: is the main curvature value of the i th point; is a small constant used to avoid the denominator being 0 (for example ); is the relative jump ratio factor, representing the logarithmic jump amplitude between the i th curvature point and the th curvature point; N is the total number of points in the sub-unit area; Constructing a jump ratio sequence based on a relative jump ratio factor , after obtaining the jump ratio sequence, the weight of the locally discrete prominent region is strengthened by constructing a rate of change enhancement function, and finally a curvature discrete reference value is generated. The generation expression of the curvature discrete reference value is: , where: is the curvature discrete reference value; represents the local rate of change of the curvature jump ratio sequence within the neighborhood of the i -th curvature point, and is used to capture the trend fluctuation information of the degree of curvature change in the sub-unit region; is the hyperbolic tangent function, which is used to compress excessive jumps and strengthen the response in the medium and high curvature change intervals; is the discrete amplification factor (such as ), which is used to non-linearly enhance the overall heterogeneity trend.

[0013] Preferably, the specific steps for generating a density gradient reference value by deeply analyzing the degree of spatial gradient change of the point density within the sub-unit region are as follows: Divide the sub-unit region into equally spaced voxel grids, and assume that the side length of each voxel is l . For each voxel, calculate the sum of the density gradient transition intensities with its 6 face-adjacent voxels. The calculation expression is: , where: is the local density transition intensity between the voxel and its 6 face-adjacent voxels; is the index set of the 6 face-adjacent voxels adjacent to the voxel ; d represents the index of a face-adjacent voxel adjacent to the current voxel , and is used to enumerate the adjacent voxels in all 6 directions (up and down, front and back, left and right) in contact with the voxel ; , respectively represent the number of points in the current voxel and the adjacent voxel; is the non-linear enhancement index, which is used to enhance the response of high-density jumps (such as taking values from 1.5 to 2.5); After obtaining the local density transition intensities of all voxels, a structure-sensitive non-linear weighted aggregation method is used to generate the density gradient reference value of the entire sub-unit region. The generated expression is as follows: , where: is the density gradient reference value, which is used to characterize the heterogeneity level of the sub-unit region; w represents the voxel index within the sub-unit region; M is the total number of voxels within the sub-unit region; is a non-linear aggregation weight factor, which is used to further suppress weak gradients and highlight significant gradients. ; It is used to compress large values and enhance small values, strengthening the discrimination ability of the density gradient reference value for high-variation regions.

[0014] In the above technical solution, the technical effects and advantages provided by the present invention are as follows: By introducing a heterogeneity evaluation mechanism constructed based on the geometric features of subunit regions and combining the trained support vector machine model, the present invention realizes the quantitative recognition and classification judgment of local structure complexity, and dynamically adjusts the upper limit of the number of matching points accordingly, so that high-heterogeneity regions obtain more geometric constraints, and low-heterogeneity regions avoid redundant matching interference, thereby effectively avoiding the registration error diffusion and structural misalignment problems caused by the unified parameter strategy, improving the overall quality and stability of the street view model in terms of morphological restoration, boundary alignment, and topological continuity, and providing more robust, efficient, and intelligent technical support for application scenarios such as 3D urban reconstruction, autonomous driving map construction, and precision mapping. Description of the Drawings

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required to be used in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0016] Figure 1 It is the method flow chart of the multi-source fusion point cloud data-driven 3D street view generation and optimization method of the present invention. Detailed Embodiments

[0017] Now, the exemplary embodiments will be described more comprehensively with reference to the drawings. However, the exemplary embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these exemplary embodiments are provided so that the present disclosure will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art.

[0018] The present invention provides a Figure 1 multi-source fusion point cloud data-driven 3D street view generation and optimization method as shown, including the following steps: According to the spatial range of the input large-scale point cloud data and combining its three-dimensional coordinate system distribution characteristics, a structured spatial partitioning algorithm is used to divide the overall registration area into multiple subunit regions with equal area and uniform distribution, disassembling the wide-area point cloud with non-uniform geometric features into cells that can be locally modeled and independently analyzed; Common partitioning strategies include the recursive space partitioning method based on Octree, the fixed-resolution grid partitioning (Voxel Grid), or the region-growing type of space adaptive segmentation algorithm. The core goal of this process is to decompose the wide-area point cloud with heterogeneous geometric features into cells that can be locally modeled and analyzed independently, providing an operational basis and spatial boundary conditions for subsequent determination of the heterogeneity level and assignment of registration parameters for each subunit.

[0019] The recursive space partitioning method based on Octree divides the overall registration area into multiple sub-unit areas with equal area and uniform distribution. The specific steps are as follows: First, establish an overall bounding cube for the input large-range point cloud data, which serves as the root node representing the entire spatial area; Then, recursively divide the cube into eight equal parts according to the three-dimensional coordinate range (X, Y, Z) of the point cloud. Each division evenly divides the parent node space into eight sub-cube areas, which are respectively added to the Octree structure as child nodes; At each division stage, judge whether the number of point clouds in each sub-cube meets the minimum division threshold. If it exceeds the threshold, continue recursive subdivision until the preset minimum voxel size or the lower limit of the number of points condition is met; The finally generated leaf nodes are multiple sub-unit areas with equal area (volume) and uniform spatial distribution. These areas not only retain the spatial continuity of the original point cloud but also provide a structured spatial basis for subsequent local geometric feature extraction and adaptive registration strategies.

[0020] For each sub-unit area, obtain the geometric structure features of the internal point cloud, and construct a geometric feature descriptor containing multiple dimensions to provide multi-source input variables for subsequent heterogeneity analysis; To obtain the geometric structure features of the internal point cloud of each sub-unit area, local neighborhood analysis needs to be performed on all the point clouds in this area. The specific method is: for each point in the sub-unit, use the fixed-radius or k-nearest neighbor search method to construct its local point set, and calculate its spatial distribution characteristics and geometric properties, such as the principal curvature direction, normal vector, point density, etc.; then statistically summarize the local features of all points, such as calculating the mean value, standard deviation, or histogram distribution, and finally form a global descriptor reflecting the structural features of the entire sub-unit. This process can be realized by algorithms such as PCA (Principal Component Analysis), normal vector statistics, entropy measurement, or in-voxel structure distribution, ensuring that the geometric characteristics of each sub-unit area are fully expressed.

[0021] The geometric structure features of the sub-unit region usually include the following indexes in multiple dimensions: point density, that is, the number of points per unit volume, reflecting the density of spatial distribution; principal curvature values and their statistics, such as mean value, extreme value, variance, used to measure the surface undulation degree; variance of normal vector direction, used to characterize the variation amplitude of surface normal, reflecting the structural consistency or complexity; point cloud information entropy, quantifying the order degree or randomness of the spatial arrangement of the point cloud; proportion of edge points, which can be extracted by detecting normal mutation or gradient mutation, used to identify the structural contour features; voxel uniformity index, measuring the spatial uniformity or aggregation of the point cloud within the sub-unit.

[0022] The above indexes are combined to form a multi-dimensional feature vector for further evaluation of the heterogeneity level and regulation of matching point parameters.

[0023] By using feature engineering techniques, key indexes reflecting the high heterogeneity of the sub-unit region are extracted from the obtained geometric structure features. After in-depth analysis of the extracted key indexes, the geometric heterogeneity level of each sub-unit region is quantified based on the analysis results; By using feature engineering techniques, key indexes reflecting the high heterogeneity of the sub-unit region are extracted from the obtained geometric structure features. Among them, the extracted key indexes include the dispersion degree of the distribution of the principal curvature values of all points in the sub-unit region and the spatial gradient change degree of the point density within the sub-unit region. After in-depth analysis of the extracted key indexes, a curvature dispersion reference value and a density gradient reference value are respectively generated, and the geometric heterogeneity level of each sub-unit region is quantified based on the curvature dispersion reference value and the density gradient reference value.

[0024] For each sub-unit region, if the dispersion degree of the distribution of the principal curvature values of all points in the sub-unit region is higher, it usually indicates that the geometric heterogeneity level of this region is higher. This is because the principal curvature reflects the bending degree of the local morphology of the point cloud surface, and a high dispersion degree indicates that there are various different geometric feature forms in this region, such as the transition from flat to strongly convex, from concave to edge mutation; this drastic curvature change usually means complex structure composition, large detail differences or coexistence of multiple configurations, which are common in building facades, carved surfaces, natural vegetation, damaged components or structural junction areas, etc. Therefore, the dispersion of the principal curvature not only quantifies the severity of geometric undulation, but also is an important sensitive index characterizing the internal heterogeneity of the local sub-unit region.

[0025] The specific steps for generating the curvature dispersion reference value after in-depth analysis of the dispersion degree of the distribution of the principal curvature values of all points in the sub-unit region are as follows: First, for all points within the sub-unit region, the principal curvature value of each point is calculated using the k-nearest neighbor method (k is generally 10 - 30) and the principal curvatures of all points within the sub-unit region are sorted in spatial order or density gradient to construct a curvature sequence Then, a relative jump ratio factor is introduced, defined in the form of the logarithmic ratio of the changes between adjacent curvature points, which is used to highlight the non-linear mutation characteristics between points with drastic jumps. The calculation expression of the relative jump ratio factor is: where: is the principal curvature value of the i -th point; is a small constant used to avoid a zero denominator (e.g., ); is the relative jump ratio factor, representing the logarithmic jump amplitude between the i -th curvature point and the -th curvature point; N is the total number of points within the subunit region; This ratio sequence is used to capture the density and intensity of curvature mutation events in the subunit region. It not only considers the local deformation amplitude but also effectively avoids the interference of extreme values, and is the basic feature flow for constructing the heterogeneity characterization.

[0026] Constructing a jump ratio sequence based on the relative jump ratio factor After obtaining the jump ratio sequence, the weight of the region highlighting local discreteness is strengthened by constructing a rate-of-change enhancement function, and finally a curvature discrete reference value is generated. The generation expression of the curvature discrete reference value is: where: is the curvature discrete reference value; represents the local rate of change of the curvature jump ratio sequence in the neighborhood of the i -th curvature point, which is used to capture the trend fluctuation information of the degree of drastic curvature change in the subunit region; is the hyperbolic tangent function, which is used to compress excessive jumps and strengthen the response in the medium and high curvature change intervals; is the discrete amplification factor (e.g., ), which is used to non-linearly enhance the overall heterogeneity trend; From the curvature discrete reference value, it can be seen that the larger the value of the curvature discrete reference value generated after in-depth analysis of the discreteness of the distribution of the principal curvature values of all points in the subunit region, the higher the geometric heterogeneity level of the subunit region; conversely, the lower the geometric heterogeneity level of the subunit region. The reason is that the curvature discrete reference value essentially measures the fluctuation intensity of the curvature jump ratio in the spatial sequence within the region, which is manifested as the density of drastic morphological changes of geometric surfaces at different positions within a local range. When the curvature discrete reference value is large, it indicates that there are a large number of local geometric units with structural mutations, normal discontinuities, complex morphologies or mixed multi-configurations in the region, which is a typical region with high heterogeneity expression; while if the curvature discrete reference value is small, it means that the curvature change in the region is gentle, the structure is unified, the surface continuity is high, the overall configuration is regular, and the geometric heterogeneity level is naturally low.

[0027] For each sub - unit area, the higher the degree of change in the spatial gradient of the point density within the sub - unit area, the higher the level of geometric heterogeneity of the sub - unit area is usually indicated. This is because the spatial gradient of the point density essentially reflects the degree of non - uniformity of the distribution of the point cloud in space, that is, the rate of change of the number of points from one local position to another local position. In areas with uniform or regular structures, such as flat ground, walls, or open spaces, the point cloud density is usually relatively consistent and the gradient is close to zero; while in areas with complex geometric structures or abrupt boundary changes, such as building corners, structural junctions, occlusion edges, or stacks of anisotropic components, the point density changes sharply with shape undulations, occlusion projections, or scanning perspective changes, forming a significant density gradient, thus revealing the geometric discontinuity and heterogeneity within the area. Therefore, the degree of change in the spatial gradient of the point density can be used as an important auxiliary indicator for identifying high - heterogeneity areas.

[0028] The specific steps for generating the density gradient reference value after in - depth analysis of the degree of change in the spatial gradient of the point density within the sub - unit area are as follows: Divide the sub - unit area into voxel grids with equal spacing, and assume that the side length of each voxel is l . For each voxel, calculate the sum of the density gradient transition intensities with its 6 - face adjacent voxels. The calculation formula is: where: is the local density transition intensity between voxel and its 6 - face adjacent voxels; is the index set of the 6 - face adjacent voxels adjacent to voxel ; d represents the index of a face - adjacent voxel adjacent to the current voxel , which is used to enumerate all adjacent voxels in the 6 directions (up - down, front - back, left - right) in contact with voxel ; , respectively represent the number of points in the current voxel and the adjacent voxel; is a non - linear enhancement index used to enhance the response to high - density jumps (such as taking values from 1.5 to 2.5); This step constructs a local density transition field, emphasizing the characteristics of rapid spatial changes in density and avoiding the interference of smooth areas on the characterization of structural heterogeneity.

[0029] After obtaining the local density transition intensities of all voxels, a structure - sensitive non - linear weighted aggregation method is used to generate the density gradient reference value for the entire sub - unit area. The generated formula is as follows: where: is the density gradient reference value, which is used to characterize the heterogeneity level of the sub - unit area;w represents the voxel index within the subunit region; M is the total number of voxels in the subunit area; is a nonlinear aggregation weight factor, which is used to further suppress weak gradients and highlight significant gradients. ; It is used to compress large values, enhance small values, and strengthen the ability of density gradient reference values ​​to distinguish high-variability areas; Through nonlinear transformation and aggregation, the contribution of regions with low structural differences is suppressed, and regions with significant structural mutations are highlighted, thus achieving highly sensitive quantitative characterization of the heterogeneity level of subunits.

[0030] It can be seen from the density gradient reference value that the larger the performance value of the density gradient reference value generated after in-depth analysis of the spatial gradient variation of the point density in the sub-unit area, the higher the level of geometric heterogeneity of the sub-unit area, and vice versa. The reason is that the density gradient reference value describes the spatial variation of the point density inside the sub-unit by performing enhanced nonlinear accumulation of the point density difference between each voxel and its adjacent voxels; when a region contains obvious structural transitions, occluded edges, interlaced objects or mutation surfaces, its point cloud density will show sharp changes in different spatial positions, resulting in a large local transition intensity, thereby pushing the overall density gradient reference value up; while in areas with smooth geometric shapes and uniform distribution of point density, the density gradient is weak, and the density gradient reference value also decreases accordingly. Therefore, the high or low density gradient reference value can effectively reflect the continuity or mutation degree of the structure in the region, and is an important measurement basis for geometric heterogeneity.

[0031] The analyzed indicators are input as feature vectors into a support vector machine model that has been trained with a large number of labeled samples in advance, and the degree of geometric heterogeneity corresponding to the sub-unit area is intelligently predicted; The analyzed curvature discrete reference value and density gradient reference value are input as feature vectors into the support vector machine model that has been trained in advance with a large number of labeled samples. The heterogeneity level coefficient is generated by the support vector machine model, and the degree of geometric heterogeneity corresponding to the sub-unit area is intelligently predicted based on the heterogeneity level coefficient.

[0032] A trained support vector machine (SVM) model refers to a set of model parameters obtained through learning and optimization using the support vector machine (SVM) algorithm based on a large amount of known input features and label data, which can be used for classification or regression tasks. During the training phase, the model constructs an optimal hyperplane to separate or fit input samples with different output categories (or numerical levels). In the prediction phase, new, unlabeled input features (such as curvature discrete reference values and density gradient reference values) can be substituted into the model to obtain corresponding output results (such as geometric heterogeneity levels or continuous heterogeneity coefficients), enabling intelligent identification and evaluation of the heterogeneity levels in local subunit regions.

[0033] In this application context, the trained SVM model is a classification or regression model designed for the geometric heterogeneity discrimination task of large-scale point cloud data. Its training process includes multiple stages such as sample collection, feature extraction, label construction, model learning, and optimization. Specifically, a large number of representative point cloud samples are first selected from subunit regions with different scenarios and structural complexities, and a multi-dimensional feature vector containing curvature discrete reference values and density gradient reference values is constructed for each sample. Subsequently, based on domain knowledge or expert evaluation results, a true "geometric heterogeneity level" label is assigned to each sample point cloud. For example, it can be divided into three levels: "high heterogeneity", "medium heterogeneity", and "low heterogeneity", or directly quantified as a continuous value (such as a heterogeneity coefficient) to form a standard training sample set. Then, these labeled feature samples are input into the SVM learning framework, and the optimal classification boundary or regression function is found through the principle of structural risk minimization.

[0034] The basic idea of the support vector machine is to find a hyperplane with the maximum margin in the feature space to separate samples of different classes, or to find an optimal function expression that fits within a tolerance error range in a regression problem. By introducing kernel functions (such as RBF kernel, linear kernel, polynomial kernel, etc.) during the training process, the SVM can map the input non-linear features into a high-dimensional space, making the originally inseparable samples linearly separable in the new space. In this application, due to the highly non-linear and structurally coupled characteristics of the curvature discrete reference values and density gradient reference values themselves, the Gaussian radial basis function (RBF) is usually used as the kernel function to enhance the model's ability to identify local change features.

[0035] The output of the trained SVM model includes the following key components: the support vector set, which consists of the key training samples that form the optimal boundary; the Lagrange multipliers or weight coefficients, indicating the influence degree of each support vector on the decision boundary; the bias term b, representing the offset of the hyperplane in the feature space; the kernel function type and parameter configuration for mapping new samples; the error tolerance and penalty factor C, which control the balance between the model's fitting ability and generalization performance. After being cross-validated and performance-evaluated (such as accuracy, recall, F1-score, RMSE, etc.), the model is fixed for subsequent prediction tasks.

[0036] In the point cloud registration system, this trained SVM model plays a core role as a "local structure complexity recognizer". Its prediction process is as follows: for each local sub-unit area to be processed, the system first calculates its discrete reference value of curvature and reference value of density gradient, and constructs the corresponding two-dimensional feature vector; then, this vector is input into the trained SVM model, and the model judges the heterogeneity level (such as high, medium, low) to which the sample should belong through the learned decision function, or outputs a continuous heterogeneity level coefficient. It should be noted that the training effect of the SVM model highly depends on the diversity of samples and the accuracy of labels. In high-precision point cloud processing tasks, the training samples should cover various urban landform types (such as high-rise dense areas, open spaces, transition zones, structurally dense facilities, etc.), and take into account the feature performance differences brought by different sensor sources (such as LiDAR, structured light, camera fusion data, etc.); in addition, the label construction process needs to involve domain experts or be automatically labeled through an auxiliary rule system to ensure that the heterogeneity level reflects the true geometric complexity state.

[0037] In summary, the "trained support vector machine model" is not only a static classifier or regressor, but also a fundamental core module for realizing local structure cognition and parameter adaptive adjustment in the entire intelligent registration and regulation system. By introducing this model, it can effectively replace manual rules and empirical judgments, realize the regulation of intelligent matching strategies under "structure perception drive", and fundamentally solve the problems of insufficient adaptability, parameter solidification, and easy error diffusion in the current point cloud registration technology when dealing with local heterogeneity. This solution has high generalization ability and can be extended to other applications such as 3D modeling, SLAM, and map reconstruction, providing a solid intelligent recognition foundation for constructing a highly robust and high-precision multi-source point cloud fusion system.

[0038] The support vector machine model is not specifically limited here, and any machine learning model that can realize comprehensive analysis of the discrete reference value of curvature and the reference value of density gradient to generate a heterogeneity level coefficient is acceptable. To implement the technical solution of the present invention, the present invention provides a specific implementation method; the heterogeneity level coefficient The generated expression is: , where , are respectively the weight coefficients of the discrete curvature reference value and the density gradient reference value , and , are both greater than 0. The weight coefficients (i.e., and in the formula) refer to the proportional factors used to balance and regulate the relative contribution degrees of two different input indicators (the discrete curvature reference value and the density gradient reference value) in the comprehensive analysis during the process of generating the heterogeneity level coefficient. The setting of the weight coefficients reflects the system's judgment on the importance of different feature dimensions: when a certain feature is more significant in characterizing heterogeneity, its corresponding weight should be higher, and vice versa. By performing a weighted combination of these two indicators, the heterogeneity level coefficient can more accurately reflect the true geometric heterogeneity state of the local subunit region. Therefore, the weight coefficients are not only mathematical parameters but also the implicit "feature importance trade-off mechanism" in the model.

[0039] It can be seen from the heterogeneity level coefficient that the larger the performance value of the discrete curvature reference value generated by deeply analyzing the discrete degree of the principal curvature value distribution of all points in the subunit region, and the larger the performance value of the density gradient reference value generated by deeply analyzing the spatial gradient change degree of the point density within the subunit region, that is, the larger the performance value of the heterogeneity level coefficient generated by the intelligent prediction of the geometric heterogeneity degree corresponding to the subunit region by the trained support vector machine model, the higher the geometric heterogeneity level of the subunit region, and vice versa, it indicates that the geometric heterogeneity level of the subunit region is lower.

[0040] Based on the heterogeneity degree of the subunit region predicted by the support vector machine model, an adaptive mapping relationship between the heterogeneity degree and the registration parameters is constructed, and the corresponding upper limit of the number of matching points is dynamically allocated for different heterogeneity levels of the subunit region to achieve region-specific allocation of the number of matching points; Dynamically allocate the corresponding range of the number of matching points for different heterogeneity levels of the subunit region. The specific steps are as follows: Obtain the heterogeneity level coefficient of each subunit region predicted by the support vector machine model, and construct a matching point adjustment factor for controlling the allocation intensity of the number of matching points based on the heterogeneity level coefficient. This matching point adjustment factor is calculated in the form of an exponential mapping, and the calculation expression is: , where: represents the matching point adjustment factor driven by heterogeneity, and its value range is in ; is the maximum amplification factor, which controls the upper limit of the matching point adjustment factor and is used to adjust the degree of resource tilt in high heterogeneity regions. ; is the non-linear response adjustment factor, which is used to control the response speed of the matching point adjustment factor curve to the heterogeneity level coefficient; is the heterogeneity level coefficient of this sub-unit area; The core advantage of this exponential function design is that when the heterogeneity coefficient is small, the matching point adjustment factor responds slowly, avoiding excessive resource investment in low-complexity regions; while when the heterogeneity coefficient approaches 1, the matching point adjustment factor rapidly amplifies, so as to concentrate more matching points in high heterogeneity regions and achieve the control strategy of "focused tilt and smooth growth".

[0041] Based on the constructed matching point adjustment factor , the upper limit of the final number of matching points in this sub-unit area is dynamically allocated, and the expression of the dynamic allocation is as follows: , where: represents the upper limit of the number of matching points allocated to the current sub-unit area; represents the minimum lower limit of the number of matching points set by the system, which is used for low heterogeneity regions (such as flat ground); represents the maximum upper limit of the number of matching points, which is applicable to high heterogeneity regions (such as complex buildings or edge mixing areas); The formula design reflects the regulation logic of "adapting to local conditions": when the heterogeneity level is low (i.e., ), the matching point adjustment factor , and at this time ; while when the heterogeneity level is high (i.e., ), , and at this time is close to or equal to . This allocation mechanism based on the exponential regulation factor can effectively avoid mis-matching caused by "excessive matching points", and can also provide sufficient geometric constraint support in complex regions, enhancing the local adaptability and global stability of point cloud registration.

[0042] By introducing a parameter regulation mechanism driven by heterogeneity, an intelligent control strategy of "dynamically adjusting parameter configuration due to regional structural differences" is realized during the point cloud registration process, thus effectively solving problems such as insufficient accuracy, high risk of misregistration, and unreasonable allocation of computing resources in the existing unified registration strategy when dealing with spatially heterogeneous regions. Specifically, aiming at the geometric structure differences presented by different sub-unit regions in the large-scale point cloud registration task, this step first uses a support vector machine model to classify or quantitatively judge the heterogeneity degree of each region, and establishes a mapping relationship between the "heterogeneity degree" and the "registration control parameter (such as the upper limit of the number of matching points)" based on the model output results. Through this mapping mechanism, regions with high structural complexity and significant heterogeneity will obtain more support for the number of matching points to strengthen the local rigid transformation constraint and feature capture ability; while regions with single geometric features and low heterogeneity can reduce the matching point configuration to reduce the risk of redundant matching and error propagation. This strategy not only improves the registration accuracy and robustness of complex regions, but also optimizes the resource allocation efficiency of the entire system, which is the key support link for realizing high-precision and multi-scene adaptive point cloud registration. Through the implementation of this step, the adaptability of the registration system to multi-source and heterogeneous environments can be significantly enhanced, effectively ensuring the geometric accuracy and global consistency of subsequent tasks such as 3D modeling, environmental perception, and navigation positioning.

[0043] After setting the range of the number of matching points based on the regulation strategy, the matching point control parameter and other registration parameters (such as the matching radius, tolerance threshold) are jointly input into the subsequent point cloud registration module to achieve the adaptive registration of sub-unit regions; Each sub-unit region can independently execute the establishment of initial matching pairs and the estimation of local rigid transformation, and perform consistency coordination through a global optimizer (such as pose graph or BA). This strategy not only improves the local accuracy of registration, but also significantly reduces the risk of conduction and accumulation of global errors, and finally a high-precision 3D point cloud model with topological continuity, geometric accuracy, and heterogeneous compatibility can be constructed.

[0044] Through the above method for generating and optimizing 3D street scenes driven by multi-source fused point cloud data, it is possible to achieve intelligent local adaptive processing of large-scale heterogeneous point cloud data based on structural perception, significantly improving the registration accuracy and overall modeling consistency. By introducing a heterogeneity evaluation mechanism based on the geometric features of sub-unit regions and combining the trained support vector machine model, this method realizes the quantitative recognition and classification judgment of local structural complexity, and dynamically adjusts the upper limit of the number of matching points accordingly, enabling high-heterogeneity regions to obtain more geometric constraints and low-heterogeneity regions to avoid redundant matching interference, thereby effectively avoiding the registration error diffusion and structural misalignment problems caused by the unified parameter strategy, and improving the overall quality and stability of the street scene model in terms of shape restoration, boundary alignment, and topological continuity, providing a more robust, efficient, and intelligent technical support for application scenarios such as 3D city reconstruction, autonomous driving map construction, and precision surveying and mapping.

[0045] The above formulas are all dimensionless and take their numerical calculations. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0046] Only some exemplary embodiments of the present invention have been described above by way of illustration. Undoubtedly, for those of ordinary skill in the art, without departing from the spirit and scope of the present invention, the described embodiments can be modified in various different ways. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the protection scope of the claims of the present invention.

[0047] As described above, this is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for generating and optimizing 3D street scenes driven by multi-source fused point cloud data, characterized in that It includes the following steps: According to the spatial range of the input large-scale point cloud data and combining its three-dimensional coordinate system distribution characteristics, the overall registration area is divided into multiple sub-unit areas with equal area and uniform distribution by using a structured space division algorithm; For each sub-unit area, obtain the geometric structure characteristics of the internal point cloud and construct a geometric feature descriptor containing multiple dimensions; Extract the key indicators reflecting the high heterogeneity of the sub-unit area from the obtained geometric structure characteristics through feature engineering techniques. After in-depth analysis of the extracted key indicators, quantify the geometric heterogeneity level of each sub-unit area based on the analysis results; Input the analyzed indicators as feature vectors into the support vector machine model that has been trained in advance with a large number of labeled samples to intelligently predict the geometric heterogeneity degree corresponding to the sub-unit area; Based on the heterogeneity degree of the sub-unit area predicted by the support vector machine model, construct an adaptive mapping relationship between the heterogeneity degree and the registration parameters, and dynamically allocate the corresponding upper limit of the number of matching points for different heterogeneity levels of the sub-unit area to achieve area-specific matching point numbers; After setting the range of the number of matching points based on the regulation strategy, input the matching point control parameters and other registration parameters into the subsequent point cloud registration module together to achieve the adaptive registration of the sub-unit area.

2. The multi-source fusion point cloud data-driven three-dimensional street view generation and optimization method according to claim 1, wherein, Use a recursive space division method based on octree, fixed-resolution grid division, or region-growing type space adaptive segmentation algorithm to divide the overall registration area into multiple sub-unit areas with equal area and uniform distribution.

3. The multi-source fusion point cloud data-driven three-dimensional street view generation and optimization method according to claim 2, wherein The overall registration area is divided into multiple sub-unit areas with equal area and uniform distribution by using a recursive space division method based on octree. The specific steps are as follows: Establish an overall bounding cube for the input large-scale point cloud data to represent the entire spatial area as the root node; Recursively divide the cube into eight equal parts according to the three-dimensional coordinate range of the point cloud. Each division evenly divides the parent node space into eight sub-cube areas and adds them as child nodes to the octree structure respectively; At each division stage, judge whether the number of point clouds in each sub-cube meets the minimum division threshold. If it exceeds the minimum division threshold, continue recursive subdivision until the preset minimum voxel size or point number lower limit condition is met. The finally generated leaf nodes are multiple sub-unit areas with equal area and uniform spatial distribution.

4. The method for generating and optimizing 3D street scenes driven by multi-source fused point cloud data according to claim 1, wherein Extract the key indicators reflecting the high heterogeneity of the sub-unit area from the obtained geometric structure characteristics through feature engineering techniques. Among them, the extracted key indicators include the dispersion degree of the distribution of the principal curvature values of all points in the sub-unit area and the spatial gradient change degree of the point density within the sub-unit area. After in-depth analysis of the extracted key indicators, generate the curvature dispersion reference value and the density gradient reference value respectively, and quantify the geometric heterogeneity level of each sub-unit area based on the curvature dispersion reference value and the density gradient reference value.

5. The method for generating and optimizing 3D street views driven by multi-source fusion point cloud data according to claim 4, characterized in that The analyzed discrete reference value of curvature and the reference value of density gradient are used as feature vectors and input into a support vector machine model that has been trained in advance with a large number of labeled samples. The support vector machine model generates a heterogeneity level coefficient, and based on this coefficient, an intelligent prediction of the geometric heterogeneity degree corresponding to the sub-unit region is made.

6. The multi-source fusion point cloud data-driven three-dimensional street view generation and optimization method according to claim 5, characterized in that, Dynamically allocate the corresponding range of the number of matching points for sub-unit regions with different heterogeneity levels. The specific steps are as follows: Obtain the heterogeneity level coefficient of each sub-unit region predicted by the support vector machine model, and construct a matching point adjustment factor for controlling the allocation intensity of the number of matching points based on the heterogeneity level coefficient. The matching point adjustment factor is calculated in an exponential mapping manner, and the calculation expression is: , where: represents the matching point adjustment factor driven by heterogeneity, and its value range is in ; is the maximum amplification coefficient, which controls the upper limit of the matching point adjustment factor and is used to adjust the resource tilt degree in the high-heterogeneity region, ; is the non-linear response adjustment factor, which is used to control the response speed of the matching point adjustment factor curve to the heterogeneity level coefficient; is the heterogeneity level coefficient of this sub-unit region; Based on the constructed matching point adjustment factor , the upper limit of the final number of matching points in this subunit area is dynamically allocated, and the expression for dynamic allocation is as follows: , where: represents the upper limit of the number of matching points allocated to the current subunit area; represents the lower limit of the minimum number of matching points set by the system; represents the upper limit of the maximum number of matching points.

7. The method for generating and optimizing a 3D street view driven by multi-source fusion point cloud data according to claim 4, wherein The specific steps for generating the discrete reference value of curvature after deeply analyzing the discrete degree of the distribution of the principal curvature values of all points in the sub-unit region are as follows: For all points within the sub-unit region, the principal curvature value of each point is calculated using the k-nearest neighbor method and the principal curvatures of all points within the sub-unit region are sorted in spatial order or density gradient to construct a curvature sequence ; A relative jump ratio factor is introduced to highlight the non-linear mutation characteristics between points with intense jumps. The calculation expression of the relative jump ratio factor is as follows: , where: is the principal curvature value of the i -th point; is a small constant; is the relative jump ratio factor, indicating the logarithmic jump amplitude between the i -th curvature point and the -th curvature point; N is the total number of points within the subunit region; Construct a jump ratio sequence based on the relative jump ratio factor , after obtaining the jump ratio sequence, strengthen the weight of the locally discrete prominent area by constructing a rate of change enhancement function, and finally generate a curvature discrete reference value. The generation expression of the curvature discrete reference value is: , where: is the curvature discrete reference value; represents the local rate of change of the curvature jump ratio sequence within the neighborhood of the i -th curvature point, and is used to capture the trend fluctuation information of the degree of curvature change in the sub-unit area; is a discrete amplification factor, which is used for non-linear enhancement of the overall heterogeneity trend.

8. The method for generating and optimizing 3D street scenes driven by multi-source fusion point cloud data according to claim 4, characterized in that The specific steps for generating the reference value of density gradient after deeply analyzing the degree of spatial gradient change of the point density within the sub-unit region are as follows: Divide the sub-unit area into a voxel grid with equal spacing. Assume that the side length of each voxel is l . For each voxel, calculate the sum of the density gradient transition intensities with its six face-adjacent voxels. The calculation expression is: where: is the local density transition intensity between the voxel and its six face-adjacent voxels; is the index set of the six face-adjacent voxels adjacent to the voxel ; d represents the index of a face-adjacent voxel adjacent to the current voxel , and is used to enumerate all adjacent voxels in the six directions in contact with the voxel ; , respectively represent the number of points in the current voxel and the adjacent voxel; is the non-linear enhancement index; After obtaining the local density transition intensities of all voxels, a structure-sensitive non-linear weighted aggregation method is used to generate the density gradient reference value for the entire subunit region, and the generated expression is as follows: , where: is the density gradient reference value, which is used to characterize the heterogeneity level of the subunit region; w represents the voxel index within the subunit region; M is the total number of voxels within the subunit region; is the non-linear aggregation weight factor, .

Images