Rural infrastructure three-dimensional modeling method and system based on laser point cloud

Abstract

The invention provides a rural infrastructure three-dimensional modeling method and system based on laser point cloud, and the method comprises the steps: collecting point cloud data at a rural infrastructure site through employing a standing laser scanner, and obtaining the space information of the infrastructure; preprocessing the collected point cloud data; the point cloud data is automatically segmented and processed by using a track extraction technology; performing circle fitting and automatic fine denoising on the segmented point cloud data; correcting the scanning intensity value of the point cloud based on a distance-intensity model; extracting and smoothing the central axis of the rural infrastructure, and providing a basis for the construction of a three-dimensional model; a lining plane image of the rural infrastructure is formed through a three-dimensional point cloud technology, and a basis is provided for texture mapping; rapid browsing and storage of the three-dimensional texture model of the rural infrastructure are realized, and subsequent use and analysis are facilitated. According to the invention, the acquisition and preprocessing process of the point cloud data can be optimized, the data quality is ensured, and the modeling efficiency and precision are improved.

Description

Technical Field

[0001] The present invention relates to the field of three-dimensional modeling technology, and more specifically, to a three-dimensional modeling method and system for rural infrastructure based on laser point cloud. Background Art

[0002] For 3D modeling of rural infrastructure, existing technologies are mainly concentrated in cities and developed areas, while research and application in rural areas are relatively rare. There are a large number of houses in rural areas, which are widely distributed, and the terrain is complex and changeable, which brings considerable challenges to low-cost and high-efficiency information collection and 3D modeling. In addition, the level of digitization in rural areas is relatively low, and there is a lack of sufficient technical support and talent reserves, which also limits the application and development of existing technologies in rural areas.

[0003] In the process of implementing the embodiments of the present invention, the inventors found that there are at least the following problems or defects in the prior art: Although China's real-life 3D market is in the early stages of full development, and the real-life 3D industry chain involves a complete industrial architecture from the infrastructure layer to the application layer, including satellite mapping, tilted shadow 3D models, laser point clouds and other data layer technologies, rural areas still face some challenges in the process of informatization. For example, due to the lagging development of information infrastructure and weak information capabilities of farmers in rural areas, the digital divide phenomenon continues to expand and regenerate. In addition, the digital transformation of agriculture faces holistic and systematic obstacles to the entire industry chain from factor supply to agricultural production to market sales. Especially in remote western areas, the construction of rural digital infrastructure is relatively lagging, and modern logistics and intelligent warehousing layout are relatively lacking, which has become a bottleneck affecting economic and social development. Summary of the invention

[0004] The invention provides a method and system for three-dimensional modeling of rural infrastructure based on laser point cloud.

[0005] In a first aspect of the present invention, a method for three-dimensional modeling of rural infrastructure based on laser point cloud is provided, comprising:

[0006] A stand-mounted laser scanner is used to collect point cloud data at the rural infrastructure site to obtain the spatial information of the infrastructure, where each point cloud data point P i The spatial coordinates of can be expressed as P i =(x i ,y i ,z i ), where x i ,y i ,z i Point P i Coordinate values ​​in three-dimensional space;

[0007] Preprocess the collected point cloud data, including manual stitching and manual preliminary denoising, to improve data quality;

[0008] Using trajectory extraction technology, point cloud data is automatically segmented and processed to prepare for subsequent modeling;

[0009] Perform circle fitting and automatic fine denoising on the segmented point cloud data to ensure the accuracy of the point cloud data;

[0010] The scanning intensity value of the point cloud is corrected based on the distance-intensity model to eliminate errors in the scanning process;

[0011] Extract and smooth the central axis of rural infrastructure to provide a basis for the construction of 3D models;

[0012] The lining plane image of rural infrastructure is formed by stereoscopic point cloud technology to provide a basis for texture mapping;

[0013] Realize quick browsing and saving of 3D texture models of rural infrastructure for easy subsequent use and analysis.

[0014] Furthermore, the point cloud data preprocessing step further includes:

[0015] Use image registration technology to accurately align point cloud data to ensure data consistency;

[0016] The algorithm automatically optimizes the stitching results to improve stitching accuracy and efficiency.

[0017] Furthermore, the trajectory extraction and point cloud automatic segmentation steps include:

[0018] Analyze the moving trajectory of the laser scanner and use the sensor data for spatial positioning to ensure the accuracy of the point cloud data. The moving trajectory of the laser scanner can be represented as a series of points T = {t1, t2, ..., t m}, t j =(x j ,y i ,z j ,θ j ), where x j ,y j ,z j is the spatial coordinate of the trajectory point, θ j is the azimuth of the trajectory point;

[0019] Machine learning algorithms are used to predict and correct trajectories and improve the accuracy of automatic segmentation.

[0020] Furthermore, the circle fitting and automatic fine denoising steps include:

[0021] The least squares method is used to fit the point cloud data into a circle to determine the boundary of the segmented point cloud. The circle fitting can be achieved by the following equation:

[0022]

[0023] where c x ,c y are the coordinates of the center of the circle, r is the radius of the circle;

[0024] Adaptive filtering technology is used for automatic fine denoising, and denoising parameters are dynamically adjusted to adapt to the point cloud characteristics of different areas.

[0025] Furthermore, the distance-intensity model-based correction step includes:

[0026] Analyze the reflection intensity of the point cloud and use statistical methods to eliminate intensity deviations caused by equipment errors or environmental factors;

[0027] The scanning intensity value of the point cloud is adjusted to achieve a more accurate representation of the point cloud data, wherein the scanning intensity value I of the point cloud data can be corrected according to the distance d, and the correction formula is:

[0028] I corrected =I·e -k·d

[0029] Where k is the correction factor.

[0030] Furthermore, the step of extracting and smoothing the central axis includes:

[0031] Use edge detection algorithms to extract the central axis of the infrastructure;

[0032] The central axis is smoothed by curvature analysis to achieve a smoother central axis representation, wherein the smoothing of the central axis can be achieved by applying a polynomial fit, for example using the quadratic equation: y = ax 2 +bx+c, where a, b, c are the coefficients of the quadratic equation.

[0033] Furthermore, the step of forming a lining plane image from a three-dimensional point cloud comprises:

[0034] Adopt multi-view stereo matching technology to improve the accuracy and detail expression of plane images;

[0035] The point cloud data is projected onto a two-dimensional plane to generate an accurate lining plane image, wherein the projection of the point cloud data can be achieved by the following formula: (u, v) = Project(P), where u, v are coordinates on the two-dimensional plane, and P is a point in three-dimensional space.

[0036] Furthermore, the step of realizing quick browsing and saving of the three-dimensional texture model includes:

[0037] Analyze the geometric and physical properties of the generated 3D models to evaluate the quality and performance of the models;

[0038] Integrate 3D models with Geographic Information System (GIS) data to accurately locate and analyze models in a geographic spatial environment;

[0039] Texture mapping of 3D models can be achieved by the following formula:

[0040] T(u,v)=TextureMap(P)

[0041] Among them, T(u,v) is the result of texture mapping, and P is a point on the three-dimensional model.

[0042] Furthermore, the geometric and physical property analysis of the three-dimensional model further includes:

[0043] Evaluate the geometric accuracy of the model to ensure consistency between the model and the actual infrastructure;

[0044] Analyze the physical properties of the model, including material characteristics and durability, to provide reference for the practical application of the model.

[0045] In a second aspect of the present invention, a rural infrastructure three-dimensional modeling system based on laser point cloud is provided, comprising:

[0046] Point cloud data acquisition module, used to collect point cloud data using a stand-mounted laser scanner at rural infrastructure sites;

[0047] Data processing module, used for point cloud data preprocessing, trajectory extraction, automatic segmentation, circle fitting, automatic fine denoising and correction;

[0048] Central axis extraction and smoothing module, used to extract and smooth the central axis of rural infrastructure;

[0049] 3D modeling module, used to form lining plane images of rural infrastructure and build 3D texture models;

[0050] The model browsing and saving module is used to realize the quick browsing, saving and integrated analysis of 3D models with GIS data.

[0051] According to the above-mentioned embodiments of the present invention, at least the following beneficial effects are achieved: the three-dimensional modeling method and system of rural infrastructure based on laser point cloud of the present invention can effectively improve the accuracy and practicality of the three-dimensional model of rural infrastructure through a series of precise data processing and modeling technologies. This method can not only optimize the acquisition and preprocessing process of point cloud data to ensure data quality, but also improve the efficiency and accuracy of modeling through the application of automation technology and algorithms. Furthermore, the invention can enhance the processing capability of point cloud data and the automation level of modeling by comprehensively applying advanced technologies such as image registration technology, machine learning algorithm, least squares method, adaptive filtering technology, polynomial fitting, etc. The integration of these technologies can not only improve the accuracy of data processing, but also reduce the dependence on professionals, making three-dimensional modeling more convenient and efficient. Through this method, important data support and decision-making reference can be provided for the digital transformation and information management of rural areas, which will help promote the sustainable development of rural areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, in which:

[0053] Figure 1 A schematic diagram of a process of a rural infrastructure three-dimensional modeling method based on laser point cloud provided by an embodiment of the present invention;

[0054] Figure 2 A schematic diagram of the structure of a rural infrastructure three-dimensional modeling system based on laser point cloud provided by an embodiment of the present invention;

[0055] Figure 3 The schematic diagram schematically shows the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0056] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0057] Those skilled in the art know that the embodiments of the present invention can be implemented as a system, device, apparatus, method or computer program product. Therefore, the present invention can be specifically implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0058] It should be noted that any number of elements in the drawings is for illustration rather than limitation, and any naming is only for distinction and does not have any limiting meaning.

[0059] Reference below Figure 1 , Figure 1 The following is a flow chart of a method for three-dimensional modeling of rural infrastructure based on laser point cloud provided by an embodiment of the present invention. Figure 1 As shown, a rural infrastructure 3D modeling method 100 based on laser point cloud includes:

[0060] Step 101, using a stand-mounted laser scanner to collect point cloud data at the rural infrastructure site to obtain spatial information of the infrastructure, where each point cloud data point P i The spatial coordinates of can be expressed as P i =(x i ,y i ,z i ), where x i ,y i ,z i Point P i Coordinate values ​​in three-dimensional space;

[0061] Step 102, preprocessing the collected point cloud data, including manual stitching and manual preliminary denoising, to improve data quality;

[0062] Step 103, using trajectory extraction technology to automatically segment and process point cloud data to prepare for subsequent modeling;

[0063] Step 104, performing circle fitting and automatic fine denoising on the segmented point cloud data to ensure the accuracy of the point cloud data;

[0064] Step 105, correcting the scanning intensity value of the point cloud using a distance-intensity model to eliminate errors in the scanning process;

[0065] Step 106, extracting and smoothing the central axis of the rural infrastructure to provide a basis for building a three-dimensional model;

[0066] Step 107, forming a lining plane image of the rural infrastructure through a stereoscopic point cloud technology to provide a basis for texture mapping;

[0067] Step 108, realizes the rapid browsing and saving of the three-dimensional texture model of rural infrastructure, so as to facilitate subsequent use and analysis.

[0068] It should be noted that the embodiments of the present invention involve the use of a stand-mounted laser scanner to collect on-site data of rural infrastructure, which is the basis for building a three-dimensional model. Through laser scanning, every detail of the infrastructure can be fully captured to ensure the accuracy and authenticity of the model.

[0069] Specifically, when using a stand-mounted laser scanner for data collection, each point cloud data point P i The spatial coordinate P i =(x i ,y i ,z i ), where x i ,y i ,z i They represent the horizontal position, vertical position and height of a point in three-dimensional space. These coordinate values ​​are calculated based on the flight time of the laser beam after it is reflected back from the laser scanner. During implementation, the appropriate scanning resolution and coverage can be set according to the site environment and the size of the target object to ensure the quality and integrity of the data.

[0070] Preferably, when collecting point cloud data, multiple scanning sites can be selected to obtain a better field of view and coverage. In addition, in order to improve data quality, laser scanners with different field of view angles and resolutions can be used, or the collected raw data can be optimized through software algorithms, such as removing outliers through point cloud filtering, or reducing occlusion and blind spots through the fusion of multi-site data. During the data collection process, global positioning system (GPS) and inertial measurement unit (IMU) data can also be combined to achieve higher-precision spatial positioning.

[0071] In some embodiments, the point cloud data preprocessing step further comprises:

[0072] Use image registration technology to accurately align point cloud data to ensure data consistency;

[0073] The algorithm automatically optimizes the stitching results to improve stitching accuracy and efficiency.

[0074] It should be noted that the point cloud data preprocessing step includes the use of image registration technology to accurately align the point cloud data. This technology adjusts the position and direction of the point cloud data by identifying common features or control points in the point cloud to ensure the spatial consistency of multiple sets of data. In addition, the algorithm automatically optimizes the stitching results, which can reduce manual operation errors and improve the accuracy and efficiency of stitching. In terms of parameter setting, you can select appropriate registration algorithms and thresholds based on the characteristics of the point cloud data, such as the iterative closest point (ICP) algorithm, and set a reasonable number of iterations and convergence criteria.

[0075] Preferably, the preprocessing of the point cloud data can also include a denoising operation, which can be achieved by setting specific filters and parameters, such as a Gaussian filter or a median filter, to remove noise and outliers in the point cloud. In addition, a multi-scale analysis method can be used to repeat the registration and optimization process at different scales to improve the accuracy and robustness of the stitching. An alternative could be to use automatic feature extraction and registration methods based on deep learning, which can adaptively identify and align point cloud data and reduce reliance on manual parameter settings.

[0076] In some embodiments, the trajectory extraction and point cloud automatic segmentation steps include:

[0077] Analyze the moving trajectory of the laser scanner and use the sensor data for spatial positioning to ensure the accuracy of the point cloud data. The moving trajectory of the laser scanner can be represented as a series of points T = {t1, t2, ..., t m}, t j =(x j ,y j ,z j ,θ j ), where x j ,y j ,z j is the spatial coordinate of the trajectory point, θ j is the azimuth of the trajectory point;

[0078] Machine learning algorithms are used to predict and correct trajectories and improve the accuracy of automatic segmentation.

[0079] It should be noted that the trajectory extraction and point cloud automatic segmentation steps involve analyzing the movement trajectory of the laser scanner, which is usually achieved by recording the position and azimuth of the laser scanner in space. j =(x j ,y j ,z j ,θ j x in j ,y j ,z j represents the spatial coordinates of the trajectory point, and θ j Represents the azimuth of the trajectory point. These parameters can be obtained through the built-in GPS and IMU system of the laser scanner. In the automatic segmentation process, the threshold of the segmentation can be set, such as triggering a new segmentation by an angle change or a distance change.

[0080] Preferably, in order to improve the accuracy of automatic segmentation, machine learning algorithms can be used to predict and correct the trajectory. This may include training a model to identify and predict patterns or anomalies in the trajectory, and automatically adjusting the segmentation parameters to adapt to different scanning conditions and environmental characteristics. For example, algorithms such as support vector machines (SVM) or random forests can be used to analyze the trajectory data and optimize the segmentation results. In addition, adaptive algorithms can be introduced to dynamically adjust the parameters of trajectory extraction to adapt to the complex and changing rural infrastructure environment. Alternatives may include the use of image-based trajectory analysis methods to assist in the extraction and correction of trajectories by analyzing on-site pictures taken by laser scanners.

[0081] In some embodiments, the circle fitting and automatic fine denoising steps include:

[0082] The least squares method is used to fit the point cloud data into a circle to determine the boundary of the segmented point cloud. The circle fitting can be achieved by the following equation:

[0083]

[0084] where c x ,c y are the coordinates of the center of the circle, r is the radius of the circle;

[0085] Adaptive filtering technology is used for automatic fine denoising, and denoising parameters are dynamically adjusted to adapt to the point cloud characteristics of different areas.

[0086] It should be noted that the circle fitting step uses the least squares method to process the point cloud data, which is a mathematical optimization technique that finds the best function match for the data by minimizing the sum of squares of errors. In this process, the boundaries of the segmented point cloud are determined, and the equation used is In, c x ,c y is the coordinate of the center of the circle, r is the radius of the circle, and x j ,y j is the coordinate of the point in the point cloud. The automated fine denoising step uses adaptive filtering technology to dynamically adjust the denoising parameters according to the local characteristics of the point cloud data to achieve the best denoising effect.

[0087] Preferably, a robust regression method can be introduced in the circle fitting process to improve the resistance to outliers, such as using M-estimation to replace the square loss function in the least squares method. In automated fine denoising, the scale parameter of the filter can be adaptively adjusted according to the local density and curvature changes of the point cloud data, or local principal component analysis (PCA) based on the point cloud data can be used to identify and remove noise. In addition, deep learning technology can be combined to train a denoising network to automatically identify and remove noise points in the point cloud, thereby improving the intelligence level of denoising. Alternatives may include the use of a multi-scale denoising method based on wavelet transform, which is able to analyze and remove noise at different scales and is suitable for processing point cloud data with different characteristics.

[0088] In some embodiments, the distance-intensity model-based correction step includes:

[0089] Analyze the reflection intensity of the point cloud and use statistical methods to eliminate intensity deviations caused by equipment errors or environmental factors;

[0090] The scanning intensity value of the point cloud is adjusted to achieve a more accurate representation of the point cloud data, wherein the scanning intensity value I of the point cloud data can be corrected according to the distance d, and the correction formula is:

[0091] I corrected =I·e -k·d

[0092] where k is the correction factor.

[0093] It should be noted that the correction step based on the distance-intensity model involves analyzing the reflection intensity of the point cloud and using statistical methods to eliminate the intensity deviation caused by equipment errors or environmental factors. The scanning intensity value I of the point cloud data can be corrected according to the distance d. The correction formula is: I corrected =I·e -k·d , where k is the correction factor, and its value can be adjusted according to the actual scanning equipment and environmental conditions. This correction process ensures that the point cloud data can maintain consistency and accuracy at different scanning distances.

[0094] Preferably, the correction process may further include performing local statistical analysis on the point cloud data to identify and compensate for locally varying error patterns. For example, the point cloud data may be processed in blocks, statistical parameters may be calculated for the data within each block, and the correction coefficient k may be adjusted based on these parameters. In addition, a machine learning algorithm may be introduced to predict and adjust the correction coefficient to accommodate different scanning conditions and environmental changes. Alternatives may include using a correction method based on a physical model, which may simulate and correct the intensity values ​​of the point cloud data based on the physical principles of laser-matter interaction.

[0095] In some embodiments, the step of extracting and smoothing the central axis includes:

[0096] Use edge detection algorithms to extract the central axis of the infrastructure;

[0097] The central axis is smoothed by curvature analysis to achieve a smoother central axis representation, wherein the smoothing of the central axis can be achieved by applying a polynomial fit, for example using the quadratic equation: y = ax 2 +bx+c, where a, b, c are the coefficients of the quadratic equation.

[0098] It should be noted that in the step of extracting the central axis of the infrastructure, edge detection algorithms are used to identify the edges of structures in the point cloud data. These edges usually mark the outlines of buildings or infrastructure. Through this edge information, the position of the central axis can be determined. Smoothing is to eliminate noise and irregularities caused by data collection or preprocessing, making the central axis smoother and more continuous. Polynomial fitting is a commonly used smoothing technique, in which the quadratic equation y = ax 2 +bx+c is used to best fit the data points of the central axis, and the coefficients a, b, and c are optimized based on the data points.

[0099] Preferably, the algorithm for extracting the central axis can combine image processing and computer vision technology to improve recognition accuracy. For example, the Canny operator can be used for edge detection, and the edge can be refined by non-maximum suppression and hysteresis threshold processing. In the smoothing process, in addition to using quadratic equations, high-order polynomials or spline curves can also be considered to achieve more complex curve fitting. In addition, machine learning-based smoothing algorithms can be introduced, which can automatically learn smoothing patterns in the data and adaptively adjust the fitting parameters.

[0100] More specifically, the extraction process of the central axis can further include using a region growing algorithm to connect adjacent edge points to form a continuous central axis. In the smoothing process, a local weighted scattered point smoothing (LOWESS) method can be used, which can perform weighted fitting on the point cloud data of the local area to better adapt to the actual shape of the central axis. Alternative solutions may include using an optimization method based on a genetic algorithm to find the optimal fitting curve, which can search for the best solution in a larger parameter space and improve the adaptability and robustness of the model.

[0101] In some embodiments, the step of forming a lining plane image from a three-dimensional point cloud comprises:

[0102] Adopt multi-view stereo matching technology to improve the accuracy and detail expression of plane images;

[0103] The point cloud data is projected onto a two-dimensional plane to generate an accurate lining plane image, wherein the projection of the point cloud data can be achieved by the following formula: (u, v) = Project(P), where u, v are coordinates on the two-dimensional plane, and P is a point in three-dimensional space.

[0104] It should be noted that in the step of forming a lining plane image from a three-dimensional point cloud, multi-view stereo matching technology is used, which is a method that uses multiple views acquired from different angles to improve the accuracy and detail of a plane image. The process of projecting point cloud data onto a two-dimensional plane involves calculating the corresponding coordinates u, V on the two-dimensional plane for each three-dimensional point P, which is usually achieved by finding the perpendicular projection of the point in three-dimensional space to the imaging plane. In this process, it may be necessary to set projection parameters, such as the size and position of the imaging plane, and determine which point cloud data are selected as candidates for projection.

[0105] Preferably, the generation process of the lining plane image can be combined with the technology of photogrammetry, such as constructing a digital surface model (DSM) of the point cloud for more sophisticated projection processing. In the projection process, perspective projection or orthographic projection technology can be used to select the most appropriate projection method according to the needs of the application scenario. In addition, an adaptive algorithm can be introduced to optimize the projection parameters to obtain the best image quality and detail performance.

[0106] More specifically, the projection processing of point cloud data can include gridding the original point cloud, creating a regular grid system, and projecting the point cloud data onto these grids to form a more uniform and accurate two-dimensional image. In the texture mapping stage, image-based mapping technology can be used to combine the collected texture information with the point cloud model to generate a realistic three-dimensional model. Alternatives may include using deep learning-based image generation technology, such as generative adversarial networks (GANs), to automatically generate high-quality two-dimensional plane images.

[0107] In some embodiments, the step of implementing quick browsing and saving of the three-dimensional texture model includes:

[0108] Analyze the geometric and physical properties of the generated 3D models to evaluate the quality and performance of the models;

[0109] Integrate 3D models with Geographic Information System (GIS) data to accurately locate and analyze models in a geographic spatial environment;

[0110] Texture mapping of 3D models can be achieved by the following formula:

[0111] T(u,v)=TextureMap(P)

[0112] Among them, T(u,v) is the result of texture mapping, and P is a point on the three-dimensional model.

[0113] It should be noted that in the steps of realizing the quick browsing and saving of the 3D texture model, the geometric and physical properties of the generated 3D model are first analyzed, which involves evaluating the geometric accuracy of the model and ensuring the consistency of the model with the actual infrastructure. In addition, it also includes analyzing the physical properties of the model, such as material properties and durability, to provide a reference for the practical application of the model. The 3D model is integrated with the geographic information system (GIS) data to realize the accurate positioning and analysis of the model in the geographic space environment, which is usually achieved through the import and analysis functions of the GIS software. The texture mapping of the 3D model is achieved through T(u,v)=TextureMap(P), where T(u,v) is the result of texture mapping and P is a point on the 3D model. This process involves the technology of mapping the texture image to the surface of the model.

[0114] Preferably, geometric and physical property analysis can be performed using automated software tools that can automatically identify and measure the size and shape of the model and compare it to the design specifications. GIS integration can be achieved by developing specific plug-ins or using APIs to manipulate and analyze 3D models directly in the GIS platform. Texture mapping can be enhanced by parametric mapping techniques, which can automatically adjust the scale and orientation of the texture based on the geometric characteristics of the model.

[0115] More specifically, the quick browse feature can be implemented by developing a user-friendly interface that allows users to view and navigate the 3D model in an interactive way. The saving step can include exporting the model to multiple formats to suit different usage scenarios and needs. In addition, a version control system can be introduced to manage the iteration and update of the model. Alternative solutions may include using cloud-based 3D model storage and sharing services, allowing users to access and collaborate on models from any location.

[0116] In some embodiments, the geometric and physical property analysis of the three-dimensional model further comprises:

[0117] Evaluate the geometric accuracy of the model to ensure consistency between the model and the actual infrastructure;

[0118] Analyze the physical properties of the model, including material characteristics and durability, to provide reference for the practical application of the model.

[0119] It should be noted that in the geometric and physical property analysis steps, the geometric accuracy of the model is evaluated by comparing the size, shape and position of the model with the measurement data of the actual infrastructure. This usually involves using 3D laser scanning technology or photogrammetry methods to obtain actual data and compare and analyze it with the model data. Physical property analysis includes evaluating the material properties used in the model, such as strength, durability, etc., which can be done through material testing and laboratory analysis. In addition, it can also include evaluation of the thermal performance, acoustic performance and other aspects of the model.

[0120] Preferably, the evaluation of geometric accuracy can be achieved through automated comparison software that can automatically identify differences and generate detailed deviation reports. The analysis of physical properties can be combined with expert systems, using material databases and physical models to predict the performance of the model under different environmental conditions. In addition, finite element analysis (FEA) software can be used to simulate the response of the model under stress, temperature changes, etc. to evaluate its structural integrity and durability.

[0121] More specifically, the geometric and physical property analysis steps can include the establishment of a comprehensive evaluation system that is able to integrate multiple analysis tools and methods to provide a comprehensive model evaluation report. For example, a user interface can be developed that allows users to select different analysis parameters and methods, as well as input specific environmental conditions. Alternatives may include the use of artificial intelligence-based analysis tools that are able to learn and simulate the expert's analysis process to provide more intelligent and automated evaluation results.

[0122] The above-mentioned embodiments of the present invention have the following beneficial effects: the present invention adopts laser point cloud technology for data collection, combined with automated processing and intelligent algorithms, which can effectively improve the speed and quality of model construction. Specifically, the method can ensure the high accuracy of point cloud data through steps such as fine preprocessing, trajectory extraction, circle fitting, automated denoising, and correction based on distance-intensity model. Furthermore, through the extraction and smoothing of the central axis, the lining plane image generated by the stereoscopic point cloud technology, and the three-dimensional texture model integrated with GIS data, not only can the model be quickly browsed and saved, but also strong decision-making support can be provided for the planning, construction and management of rural infrastructure. In addition, through the application of automation technology and algorithms, not only can the labor cost be reduced and the modeling efficiency be improved, but also a scientific basis can be provided for the infrastructure construction in rural areas through accurate three-dimensional models, and the economic development and social progress of rural areas can be promoted. At the same time, the technology can also provide important data support for the digital transformation and information management of rural areas, which is helpful to promote the sustainable development of rural areas and the process of urban-rural integration.

[0123] like Figure 2As shown, in some embodiments, a rural infrastructure three-dimensional modeling system 200 based on laser point cloud includes:

[0124] Point cloud data acquisition module 201, for collecting point cloud data using a stand-mounted laser scanner at a rural infrastructure site;

[0125] Data processing module 202, used for preprocessing, trajectory extraction, automatic segmentation, circle fitting, automatic fine denoising and correction of point cloud data;

[0126] A central axis extraction and smoothing module 203, used to extract and smooth the central axis of rural infrastructure;

[0127] A three-dimensional modeling module 204 is used to form a lining plane image of the rural infrastructure and construct a three-dimensional texture model;

[0128] The model browsing and saving module 205 is used to realize the quick browsing and saving of the three-dimensional model and the integrated analysis with GIS data.

[0129] It can be understood that the modules recorded in the rural infrastructure three-dimensional modeling system 200 based on laser point cloud are similar to those in the reference Figure 1 Therefore, the steps, features and beneficial effects described above for the rural infrastructure 3D modeling method based on laser point cloud are also applicable to the rural infrastructure 3D modeling system 200 based on laser point cloud and the modules contained therein, and will not be repeated here.

[0130] Reference below Figure 3 , which shows a schematic diagram of a structure 300 of an electronic device suitable for implementing some embodiments of the present invention. The electronic devices in some embodiments of the present invention may include, but are not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0131] like Figure 3As shown, the electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0132] Typically, the following devices may be connected to the I / O interface 305: input devices 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 308 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 309. The communication devices 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0133] Furthermore, the storage medium of the embodiment of the present application stores program instructions that can implement all the above methods, wherein the program instructions can be stored in the above storage medium in the form of a software product, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or terminal devices such as a computer, a server, a mobile phone, and a tablet.

[0134] The above descriptions are only some preferred embodiments of the present invention and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present invention is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the embodiments of the present invention to form a technical solution.

Claims

1. A three-dimensional modeling method for rural infrastructure based on laser point cloud, characterized in that: The following steps are involved: A stand-mounted laser scanner is used to collect point cloud data at the rural infrastructure site to obtain the spatial information of the infrastructure, where each point cloud data point P i The spatial coordinates of can be expressed as P i =(x i ,y i ,z i ), where x i ,y i ,z i Point P i Coordinate values ​​in three-dimensional space; Preprocess the collected point cloud data, including manual stitching and manual preliminary denoising, to improve data quality; Using trajectory extraction technology, point cloud data is automatically segmented and processed to prepare for subsequent modeling; Perform circle fitting and automatic fine denoising on the segmented point cloud data to ensure the accuracy of the point cloud data; The scanning intensity value of the point cloud is corrected based on the distance-intensity model to eliminate errors in the scanning process; Extract and smooth the central axis of rural infrastructure to provide a basis for the construction of 3D models; The lining plane image of rural infrastructure is formed by stereoscopic point cloud technology to provide a basis for texture mapping; Realize quick browsing and saving of 3D texture models of rural infrastructure for easy subsequent use and analysis.

2. The method for three-dimensional modeling of rural infrastructure based on laser point cloud according to claim 1, characterized in that: The point cloud data preprocessing step further includes: Use image registration technology to accurately align point cloud data to ensure data consistency; The algorithm automatically optimizes the stitching results to improve stitching accuracy and efficiency.

3. The rural infrastructure three-dimensional modeling method based on laser point cloud according to claim 1 or 2, characterized in that: The trajectory extraction and point cloud automatic segmentation steps include: Analyze the moving trajectory of the laser scanner and use the sensor data for spatial positioning to ensure the accuracy of the point cloud data. The moving trajectory of the laser scanner can be represented as a series of points T = {t1, t2, ..., t m }, t j =(x j ,y j ,z j ,θ j ), where x j ,y j ,z j is the spatial coordinate of the trajectory point, θ j is the azimuth of the trajectory point; Machine learning algorithms are used to predict and correct trajectories and improve the accuracy of automatic segmentation.

4. The method for three-dimensional modeling of rural infrastructure based on laser point cloud according to claim 3 is characterized in that: The circle fitting and automatic fine denoising steps include: The least squares method is used to fit the point cloud data into a circle to determine the boundary of the segmented point cloud. The circle fitting can be achieved by the following equation: Among them, c x ,c y are the coordinates of the center of the circle, r is the radius of the circle; Adaptive filtering technology is used for automatic fine denoising, and denoising parameters are dynamically adjusted to adapt to the point cloud characteristics of different areas.

5. The method for three-dimensional modeling of rural infrastructure based on laser point cloud according to claim 4, characterized in that: The distance-intensity model-based correction step comprises: Analyze the reflection intensity of the point cloud and use statistical methods to eliminate intensity deviations caused by equipment errors or environmental factors; The scanning intensity value of the point cloud is adjusted to achieve a more accurate representation of the point cloud data, wherein the scanning intensity value I of the point cloud data can be corrected according to the distance d, and the correction formula is: I corrected =I·e -k·d ; Where k is the correction factor.

6. The method for three-dimensional modeling of rural infrastructure based on laser point cloud according to claim 5, characterized in that: The step of extracting and smoothing the central axis includes: Use edge detection algorithms to extract the central axis of the infrastructure; The central axis is smoothed by curvature analysis to achieve a smoother central axis representation, wherein the smoothing of the central axis can be achieved by applying a polynomial fit, for example using the quadratic equation: y = ax 2 +bx+c, where a, b, c are the coefficients of the quadratic equation.

7. The method for three-dimensional modeling of rural infrastructure based on laser point cloud according to claim 6, characterized in that: The step of forming a lining plane image from a three-dimensional point cloud comprises: Adopt multi-view stereo matching technology to improve the accuracy and detail expression of plane images; The point cloud data is projected onto a two-dimensional plane to generate an accurate lining plane image, wherein the projection of the point cloud data can be achieved by the following formula: (u, v) = Project(P), where u, v are coordinates on the two-dimensional plane, and P is a point in three-dimensional space.

8. The method for three-dimensional modeling of rural infrastructure based on laser point cloud according to claim 7, characterized in that: The steps of realizing quick browsing and saving of the three-dimensional texture model include: Analyze the geometric and physical properties of the generated 3D models to evaluate the quality and performance of the models; Integrate 3D models with Geographic Information System (GIS) data to accurately locate and analyze models in a geographic spatial environment; Texture mapping of 3D models can be achieved by the following formula: T(u,v)=TextureMap(P); Among them, T(u,v) is the result of texture mapping, and P is a point on the three-dimensional model.

9. The method for three-dimensional modeling of rural infrastructure based on laser point cloud according to claim 8, characterized in that: The geometric and physical property analysis of the three-dimensional model further includes: Evaluate the geometric accuracy of the model to ensure consistency between the model and the actual infrastructure; Analyze the physical properties of the model, including material characteristics and durability, to provide reference for the practical application of the model.

10. A rural infrastructure 3D modeling system based on laser point cloud, characterized in that: include: Point cloud data acquisition module, used to collect point cloud data using a stand-mounted laser scanner at rural infrastructure sites; Data processing module, used for point cloud data preprocessing, trajectory extraction, automatic segmentation, circle fitting, automatic fine denoising and correction; Central axis extraction and smoothing module, used to extract and smooth the central axis of rural infrastructure; 3D modeling module, used to form lining plane images of rural infrastructure and build 3D texture models; The model browsing and saving module is used to realize the quick browsing, saving and integrated analysis of 3D models with GIS data.

Images