Multi-dimensional data comprehensive processing method and system for real estate surveying and mapping

Abstract

The invention discloses a multi-dimensional data comprehensive processing method and system for real estate surveying and mapping, and relates to the technical field of real estate surveying and mapping and evaluation, and the method comprises the steps: carrying out the comprehensive scanning of a real estate region through laser radar equipment, collecting multi-dimensional data including topographic data, spatial data and attribute data, and carrying out the preprocessing of the data, the method comprises the following steps: identifying and extracting elements of a real estate region by adopting a data segmentation algorithm, further carrying out deep analysis on spatial distribution features, attribute features and environment correlation features through an adaptive convolutional neural network and a graph convolutional network, and calculating spatial distribution feature indexes, attribute feature indexes and environment correlation feature indexes to obtain a real estate region; according to the method, real estate areas are comprehensively evaluated, the comprehensive score of each area is finally generated, the real estate areas are divided into important areas and common areas according to the scores, real-time surveying and mapping and a dynamic adjustment mechanism are combined, data acquisition frequencies and processing parameters of different areas are optimized, and intelligent data processing is achieved.

Description

Technical Field

[0001] The present invention relates to the field of real estate surveying and evaluation technology, and in particular to a multi-dimensional data comprehensive processing method and system for real estate surveying and mapping. Background Art

[0002] With the continuous advancement of urbanization, the planning, evaluation and management of real estate play an increasingly important role in urban development. Traditional real estate evaluation methods mostly rely on manual surveys and manual measurements, which cannot fully utilize the advantages of modern technology, resulting in low efficiency and accuracy of data collection. With the development of LiDAR, remote sensing technology and computer vision technology, it is possible to obtain multidimensional data of real estate areas. These data include not only terrain and building shapes, but also real estate attribute information such as area, volume, use, etc. However, how to accurately and comprehensively analyze these multidimensional data and effectively extract features related to real estate evaluation is still a major challenge facing current technology.

[0003] The prior art has the following deficiencies: Most existing real estate assessment methods use a single data source or traditional spatial analysis techniques, which have many deficiencies in the feature extraction and analysis process. First, many traditional methods rely on static models when extracting spatial distribution features, which cannot effectively deal with the precise analysis of irregular shapes and complex areas, resulting in low accuracy when dealing with complex areas. Second, existing attribute feature extraction methods usually ignore high-frequency changes in the data, resulting in inaccurate prediction results, especially when dealing with large-scale data, the decline in accuracy is particularly obvious. Furthermore, the analysis of environmental factors often remains on static models, which cannot reflect the impact of dynamic changes such as traffic flow and facility use on real estate in real time. In general, existing technologies cannot fully integrate the different characteristics of multidimensional data, resulting in a lack of comprehensiveness and timeliness in the assessment and decision support of real estate areas. Therefore, there is an urgent need for an innovative method that can effectively integrate multidimensional data, improve feature extraction accuracy, and have dynamic adaptability to cope with the complex challenges in modern urban real estate assessment. Summary of the invention

[0004] The purpose of the present invention is to provide a multi-dimensional data comprehensive processing method and system for real estate surveying and mapping to solve the problems in the above background.

[0005] The purpose of the present invention can be achieved through the following technical solutions: The multi-dimensional data comprehensive processing method for real estate surveying and mapping comprises the following steps: Use LiDAR to collect multi-dimensional data of real estate, including terrain data, spatial data and attribute data, and pre-process the collected data to reduce noise and optimize data integrity and accuracy; Identify and separate the element areas in real estate through data segmentation algorithms, including: boundary lines, building outlines, terrain features, analyze the element areas, and extract the spatial distribution characteristics, attribute characteristics, and environmental association characteristics of the element areas; Analyze the spatial distribution characteristics, attribute characteristics and environmental association characteristics of each element area respectively, and calculate the spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index in each element area according to the analysis results; Among them, spatial distribution characteristics include: location coordinates and shape, attribute characteristics include: area, volume and use, and environmental association characteristics include: surrounding facilities and road network; Comprehensively analyze the spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index, and generate a comprehensive score for each real estate element area based on the analysis results; Based on the comprehensive score, the real estate element areas are divided into important areas and ordinary areas; Combining real-time surveying and mapping with dynamic adjustment mechanisms, the data collection frequency and processing parameters of different types of real estate element areas are optimized to achieve intelligent data processing for different surveying and mapping targets.

[0006] Preferably, the identifying and separating the element areas in the real estate by using a data segmentation algorithm specifically includes: Multi-scale segmentation of real estate data collected by LiDAR, including terrain data, spatial data and attribute data, using local area feature coefficients Preprocess and optimize data; Among them, the calculation expression of the local area characteristic coefficient is: ; In the formula, Indicates The attribute value of each data point, Indicates the collected data points. represents the average attribute value of the region, is the total number of data points, is the local area characteristic coefficient, Indicates the coordinates of the area location; The region is segmented using geometric analysis methods, and the normal vector and local curvature of each data point are calculated; The calculation expression of the normal vector is: ; The calculation expression of local curvature is: ; In the formula, represents the normal vector of the data point, represents the Laplacian of the data points, represents the gradient of a data point, Represents the local curvature of a data point; According to the normal vector and local curvature of each data point, the area score of each area is calculated to identify the feature area. The calculation expression is: ; In the formula, represents the area score, and represents the preset scale factor, and and All are greater than 0; It is determined whether the area score of each area is greater than or equal to a preset threshold. If so, it is identified as a feature area; if not, it is identified as a non-feature area.

[0007] Preferably, the analyzing of the element regions to extract the spatial distribution characteristics, attribute characteristics and environmental association characteristics of the element regions specifically includes: The spatial distribution features of real estate areas are extracted through multiple convolutional layers and pooling layers of the convolutional neural network, including the geometric features of location, shape, boundary and outline, including: Use convolution kernels to perform convolution operations on real estate area data to extract building boundaries, terrain features, and building outlines; Through adaptive pooling operations, the pooling factor and convolution kernel size are dynamically adjusted, so that the convolutional neural network can effectively extract spatial features at different scales; Through the joint learning network, the spatial data and attribute data of real estate are integrated to extract the attribute characteristics of the real estate area, including: While using the convolutional layer to extract spatial features, the fully connected layer is used to map the spatial features with the attribute data to predict the attributes of the building; Regression analysis is used to predict the physical attributes of real estate areas; Through graph theory and graph convolutional networks combined with convolutional neural networks, the correlation features between real estate areas and their surrounding environments are extracted, including: The real estate area and its surrounding environment are modeled as graph data, and the graph convolutional network is used to extract the correlation features between the real estate and the surrounding environment, and the shortest path and accessibility environmental factors between the real estate and surrounding facilities are calculated; The environmental impact characteristics of real estate areas are extracted by jointly learning environmental data and the spatial features output by convolutional neural networks.

[0008] Preferably, the process of acquiring the spatial distribution characteristic index is: Extract the spatial distribution characteristics of the feature area, including location, shape, boundary and contour geometric features, and generate a set of spatial data points , where each point Represents the spatial coordinates of a feature in the property area, Indicates the number of collection points, Indicates The location coordinates of the collection points, Indicates the total number of collection points; For the extracted spatial data point set Apply the density clustering algorithm and preset the neighborhood radius as , and set the density threshold to divide the spatial data points into multiple clustering areas according to their density relationship , where each cluster area Include Feature points ; Calculate the spatial distribution characteristic parameters of each cluster area; Calculate the area of ​​the region, the calculation expression is: ; in, Represents cluster area The area of Convex hull boundary points representing the region; Calculate the regional density, the calculation expression is: ; In the formula, Indicates The regional density of the collection points, Represents cluster area The number of points within Calculate the morphological complexity, the calculation expression is: ; In the formula, Represents the fractal dimension of the region; According to the comprehensive calculation of the regional area, regional density and morphological complexity of each cluster region, the spatial distribution characteristic index of the region is calculated as follows: ; In the formula, Indicates the value of the maximum area among all cluster regions. represents the maximum density value among all cluster areas, An index representing the spatial distribution characteristics of clustering areas.

[0009] Preferably, the process of obtaining the attribute characteristic index is: Extract attribute characteristics of feature areas, including area, volume and usage; Construct attribute features into an attribute vector ; For the attribute vector Perform Haar wavelet transform to decompose the attribute features into low-frequency components and high-frequency components. Indicates The areas corresponding to the collection points include: Perform recursive Haar wavelet transform on the attribute data and transform the data vector Decomposed into multiple levels of low-frequency and high-frequency components, the calculation expression is: ; In the formula, and are two adjacent attribute data, and are the low-frequency and high-frequency components of layer 1, respectively; Repeat the decomposition process to convert the low-frequency components Continue to decompose to obtain higher-level low-frequency and high-frequency components, and the calculation expression is: ; In the formula, Indicates The low-frequency components of the layer, Indicates The high-frequency components of the layer, Indicates the number of layers; According to the low-frequency and high-frequency components obtained after Haar wavelet transform, the attribute characteristic index of the real estate element area is calculated. The calculation expression is: ; In the formula, Represents the attribute characteristic index within the real estate element area, and is the preset weight coefficient, and and The sum of is 1.

[0010] Preferably, the process of obtaining the environment-related characteristic index is: Extracting environmental related features of the real estate element area, wherein the environmental related features include: surrounding facilities and road network; Constructing the environment feature vector , Indicates The area corresponding to each collection point; Perform Fourier transform on the environmental feature vector to convert the time domain signal into the frequency domain signal. The calculation expression is: ; in, Indicates the frequency domain signal at The value of the frequency point, Indicates the frequency point, Indicates the number of environmental characteristic data, Indicates the total amount of environmental characteristic data, represents the imaginary unit, Represents the first Environmental characteristic data; Calculate the amplitude of the frequency component after Fourier transformation. The calculation expression is: ; in, For the The amplitude of the frequency component; Use the mean calculation expression to calculate the average frequency of all frequency components, and use the difference between the maximum frequency value and the minimum frequency value to calculate the frequency range value; Calculate the environmental association characteristic index, the calculation expression is: ; In the formula, represents the average frequency of all frequency components, Indicates the frequency range, Represents the environmental association characteristic index.

[0011] Preferably, the spatial distribution characteristic index, the attribute characteristic index and the environmental association characteristic index are comprehensively analyzed, and a comprehensive score of each real estate element area is generated according to the analysis results, specifically including: The spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index of the real estate element area are obtained, and the spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index are normalized and calculated to obtain a comprehensive score of the real estate element area.

[0012] Preferably, the real estate element area is divided into important areas and ordinary areas according to the comprehensive score, specifically including: Determine whether the comprehensive score of each real estate element area is greater than or equal to the preset threshold. If so, it means that the corresponding area is an important area and is marked as an important area. If not, it means that the corresponding area is an ordinary area and is marked as an ordinary area.

[0013] Preferably, the combination of real-time surveying and dynamic adjustment mechanism optimizes the data collection frequency and processing parameters of different types of real estate element areas to achieve intelligent data processing for different surveying and mapping targets, specifically including: According to the spatial distribution characteristics, attribute characteristics and environmental association characteristics of different real estate element areas, a classification model is established to classify the real estate element areas into multiple categories, and the classification basis is not limited to the type, area, form and surrounding environment of the real estate; Dynamically adjust the frequency of data collection based on the classification results of each real estate element area; Optimize data processing parameters based on the classification results and real-time surveying and mapping data of each real estate element area; Based on the feedback data from real-time surveying and mapping, the data collection frequency and processing parameters of different categories of areas are continuously optimized through machine learning algorithms; Adjust the model parameters based on historical data and surveying accuracy feedback; Through adaptive mechanisms, the characteristic change trends of different regions are evaluated and the computational complexity is dynamically adjusted; Optimize the computing performance of the entire system based on the dynamically adjusted data acquisition frequency and processing parameters; Generate a real-time feedback mechanism to automatically update the feature classification of real estate areas based on new collected data and processing results.

[0014] The multi-dimensional data comprehensive processing system for real estate surveying and mapping includes: A data acquisition module, which uses a laser radar to collect multi-dimensional data of real estate, including terrain data, spatial data, and attribute data, and pre-processes the collected data to reduce noise and optimize the integrity and accuracy of the data; The element area recognition module uses a data segmentation algorithm to identify and separate element areas in real estate, including: boundary lines, building outlines, and terrain features. A feature extraction module, wherein the feature extraction module analyzes the element region and extracts the spatial distribution characteristics, attribute characteristics and environmental association characteristics of the element region; A characteristic index calculation module, wherein the characteristic index calculation module analyzes the spatial distribution characteristics, attribute characteristics and environmental association characteristics of each element area respectively, and calculates the spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index in each element area according to the analysis results; A comprehensive analysis module, which performs a comprehensive analysis on the spatial distribution characteristic index, the attribute characteristic index and the environmental association characteristic index, and generates a comprehensive score for each real estate element area according to the analysis results; A region division module, wherein the region division module divides the real estate element region into an important region and a common region according to the comprehensive score; A dynamic adjustment module combines real-time surveying and mapping with a dynamic adjustment mechanism to optimize the data collection frequency and processing parameters of different categories of real estate element areas, thereby realizing intelligent data processing for different surveying and mapping targets.

[0015] Beneficial effects of the present invention: (1) The present invention uses LiDAR technology to accurately collect multi-dimensional data of real estate, including terrain data, spatial data and attribute data, and combines advanced deep learning algorithms such as adaptive convolutional neural networks and graph convolutional networks to improve the extraction accuracy of spatial distribution, attribute characteristics and environmental correlation characteristics of real estate areas. When faced with buildings and areas with complex and irregular shapes, the present invention uses a dynamic adjustment mechanism to automatically optimize the size of the convolution kernel and the selection of pooling factors according to the feature complexity of the area, which can more accurately capture subtle spatial and geometric changes and improve the flexibility and adaptability of feature extraction. Traditional methods usually use fixed convolution kernels and pooling factors, which are difficult to adapt to the huge changes in the morphology and structure of real estate areas. The present invention overcomes this problem through a dynamic mechanism, ensuring that buildings and areas of different sizes and shapes can obtain accurate feature extraction.

[0016] Furthermore, the present invention uses a multi-level joint learning network to deeply integrate the spatial data and attribute data of real estate, breaking through the limitation of separate processing of spatial features and attribute features in traditional methods, and enhancing the collaborative ability of the model in multimodal data analysis. The convolution layer first extracts the spatial features of the real estate area, such as geometric features such as building shape, outline, and boundary line, and then deeply maps these spatial features with the physical attributes of the real estate through the fully connected layer to perform attribute prediction. Through the regression analysis method, the network weights are adaptively adjusted during the training process, which improves the accuracy of the model's prediction of attributes for known and unknown areas.

[0017] (2) The present invention comprehensively analyzes the spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index of real estate areas, deeply explores the intrinsic characteristics of real estate areas from multiple dimensions, and improves the scientificity and accuracy of evaluation and decision-making. First, the calculation of the spatial distribution characteristic index not only takes into account the area, density and morphological complexity of the area, but also comprehensively quantifies the geometric morphology of the real estate area through density clustering algorithm and normal vector calculation, and realizes accurate modeling of complex morphology and different scale areas. This method can automatically identify and extract the spatial distribution patterns of different areas, providing a solid foundation for subsequent analysis. Secondly, the construction of the attribute characteristic index uses the Haar wavelet transform technology to perform multi-level decomposition of the physical attribute data of real estate, accurately captures low-frequency and high-frequency changes, breaks through the limitation of ignoring details in traditional methods, ensures accurate prediction of attributes such as building area, volume, and use, and improves the flexibility and accuracy of data processing. Finally, the design of the environmental association characteristic index combines the Fourier transform to perform frequency domain analysis on the dynamic changes of the surrounding environment, which can effectively capture the impact of time-varying factors such as traffic flow and public facility use on real estate areas. This real-time and dynamic analysis method enhances the accuracy of environmental features, and is particularly suitable for ever-changing urban environments, helping to optimize regional assessments in real time and providing support for intelligent decision-making. Through the high degree of integration and precise analysis of these three characteristic indexes, the present invention provides a more scientific, timely and accurate solution for real estate assessment, planning and decision-making, especially in the face of complex and dynamically changing areas, which can significantly improve the depth and breadth of data analysis and promote the intelligent development of the real estate industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flowchart of the specific steps of the multi-dimensional data comprehensive processing method for real estate surveying and mapping of the present invention; Figure 2 It is a flowchart of the multi-dimensional data comprehensive processing system used for real estate surveying and mapping in the present invention. DETAILED DESCRIPTION

[0019] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0020] See also Figure 1 As shown, the present invention is a multi-dimensional data comprehensive processing method for real estate surveying and mapping, comprising the following steps: Use LiDAR to collect multi-dimensional data of real estate, including terrain data, spatial data and attribute data, and pre-process the collected data to reduce noise and optimize data integrity and accuracy; Identify and separate the element areas in real estate through data segmentation algorithms, including: boundary lines, building outlines, terrain features, analyze the element areas, and extract the spatial distribution characteristics, attribute characteristics, and environmental association characteristics of the element areas; Analyze the spatial distribution characteristics, attribute characteristics and environmental association characteristics of each element area respectively, and calculate the spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index in each element area according to the analysis results; Among them, spatial distribution characteristics include: location coordinates and shape, attribute characteristics include: area, volume and use, and environmental association characteristics include: surrounding facilities and road network; Comprehensively analyze the spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index, and generate a comprehensive score for each real estate element area based on the analysis results; Based on the comprehensive score, the real estate element areas are divided into important areas and ordinary areas; Combining real-time surveying and mapping with dynamic adjustment mechanisms, the data collection frequency and processing parameters of different types of real estate element areas are optimized to achieve intelligent data processing for different surveying and mapping targets.

[0021] Among them, LiDAR is used to collect multi-dimensional data of real estate, including terrain data, spatial data and attribute data, and the collected data is pre-processed to reduce noise and optimize the integrity and accuracy of the data, including: First, the real estate area is fully scanned by using a laser radar (LiDAR) device to obtain multi-dimensional data including terrain data, spatial data and attribute data. The specific steps include: using the laser radar to emit a laser beam and measure the time of the reflected laser pulse to calculate the distance and depth information and obtain high-precision three-dimensional point cloud data. By carrying GPS and inertial measurement units (IMU) for synchronous positioning, the accuracy of spatial data is ensured, and spatial information such as terrain features, building outlines, road networks, etc. is recorded in real time through laser scanning; at the same time, sensors are used to obtain attribute data of objects, such as building height, area, purpose, and material.

[0022] The collected multidimensional data will be preprocessed to reduce noise and optimize the integrity and accuracy of the data. The preprocessing steps include: removing outliers and outliers, eliminating noise points in the lidar data through filtering algorithms, and improving the quality of point cloud data. Then, interpolation and data fusion techniques are used to complete the spatial data and attribute data to ensure the continuity and consistency of the data in different areas. Finally, different processing strategies are used to optimize the data for different surveying and mapping targets (such as terrain, buildings, roads, etc.) to ensure that the data accuracy meets the needs of subsequent analysis and processing.

[0023] The data segmentation algorithm is used to identify and separate the element areas in the real estate, including: boundary lines, building outlines, terrain features, analyze the element areas, and extract the spatial distribution characteristics, attribute characteristics and environmental association characteristics of the element areas, including: Multi-scale segmentation of real estate data collected by LiDAR, including terrain data, spatial data and attribute data, using local area feature coefficients Preprocess and optimize data; Among them, the calculation expression of the local area characteristic coefficient is: ; In the formula, Indicates The attribute value of each data point, Indicates the collected data points. represents the average attribute value of the region, is the total number of data points, is the local area characteristic coefficient, Indicates the coordinates of the area location; The region is segmented using geometric analysis methods, and the normal vector and local curvature of each data point are calculated; Among them, the calculation expression of the normal vector is: ; The calculation expression of local curvature is: ; In the formula, represents the normal vector of the data point, represents the Laplacian of the data points, represents the gradient of a data point, Represents the local curvature of a data point; According to the normal vector and local curvature of each data point, the area score of each area is calculated to identify the feature area. The calculation expression is: ; In the formula, represents the area score, and represents the preset scale factor, and and All are greater than 0; It is determined whether the area score of each area is greater than or equal to a preset threshold. If so, it is identified as a feature area; if not, it is identified as a non-feature area.

[0024] When extracting the spatial distribution features of real estate areas, an adaptive convolutional neural network architecture is used for processing. Unlike the static convolution kernels and pooling factors of the prior art, a dynamic adjustment mechanism is introduced to automatically adjust the convolution kernel size and pooling factor according to the feature complexity of different areas. When faced with buildings and areas with irregular shapes and large size variations, the network can dynamically optimize the convolution operation based on local details to enhance the flexibility of feature extraction. For example, geometric features such as building boundaries, terrain undulations, and building outlines are automatically adjusted through the convolution kernel, thereby accurately extracting spatial features at different scales. In addition, the convolutional neural network not only extracts the overall geometric shape of the building, but can also distinguish subtle structural differences, solving the problem of insufficient extraction accuracy of traditional methods in complex areas.

[0025] In the extraction of attribute features, the spatial data and attribute data of real estate are integrated through a multi-level joint learning network, breaking through the limitations of the traditional method of separate processing of spatial features and attribute features. First, the convolution layer extracts the spatial distribution characteristics of the real estate area, such as the shape and outline of the building; then, the extracted spatial features are deeply mapped with the attribute data of the real estate (such as building area, floor height, functional use, etc.) through the fully connected layer to perform attribute prediction. In order to enhance the accuracy of the prediction, the present invention further adopts a regression analysis method to accurately predict the physical properties of each area by adaptively adjusting the weights during the training process. This method can not only effectively predict the attributes of known areas, but also infer the attributes of unknown areas based on the surrounding environment, with high flexibility and adaptability.

[0026] The extraction of environmental correlation features adopts a method that combines graph convolutional networks (GCN) with convolutional neural networks (CNN) to simultaneously process the complex relationship between spatial data and environmental data. By modeling the real estate area and the surrounding environment (such as roads, public facilities, and traffic flow) as a graph structure, the present invention uses a graph convolutional network to process environmental data, and combines the spatial distribution features extracted by CNN to jointly learn the correlation between real estate and its surrounding environment. The innovation lies in that not only static graph structures are considered, but also time series data, such as dynamic changes in traffic flow and facility usage, are introduced, so that the model can update and accurately evaluate the relationship between real estate and surrounding facilities in real time. Through this dynamic graph convolutional network, the environmental impact of real estate areas can be more comprehensively understood, including the shortest path, accessibility, and the influence of surrounding facilities, providing more accurate environmental features for regional assessment.

[0027] The spatial distribution characteristic attribute characteristics and environmental association characteristics of each element area are analyzed. According to the analysis results, the spatial distribution characteristic index attribute characteristic index and environmental association characteristic index within each element area are calculated, including: The process of obtaining the spatial distribution characteristic index is as follows: Extract the spatial distribution characteristics of the feature area, including location, shape, boundary and contour geometric features, and generate a set of spatial data points , where each point Represents the spatial coordinates of a feature in the property area, Indicates the number of collection points, Indicates The location coordinates of the collection points, Indicates the total number of collection points; For the extracted spatial data point set Apply the density clustering algorithm and preset the neighborhood radius as , and set the density threshold to divide the spatial data points into multiple clustering areas according to their density relationship , where each cluster area Include Feature points ; Calculate the spatial distribution characteristic parameters of each cluster area; Calculate the area of ​​the region, the calculation expression is: ; in, Represents cluster area The area of Convex hull boundary points representing the region; Calculate the regional density, the calculation expression is: ; In the formula, Indicates The regional density of the collection points, Represents cluster area The number of points within Calculate the morphological complexity, the calculation expression is: ; In the formula, Represents the fractal dimension of the region; According to the comprehensive calculation of the regional area, regional density and morphological complexity of each cluster region, the spatial distribution characteristic index of the region is calculated as follows: ; In the formula, Indicates the value of the maximum area among all cluster regions. represents the maximum density value among all cluster areas, Indicates the spatial distribution characteristic index of the cluster area; The process of obtaining the attribute characteristic index is as follows: Extract attribute characteristics of feature areas, including area, volume and usage; Construct attribute features into an attribute vector ; For the attribute vector Perform Haar wavelet transform to decompose the attribute features into low-frequency components and high-frequency components. Indicates The areas corresponding to the collection points include: Perform recursive Haar wavelet transform on the attribute data and transform the data vector Decomposed into multiple levels of low-frequency and high-frequency components, the calculation expression is: ; In the formula, and are two adjacent attribute data, and are the low-frequency and high-frequency components of layer 1, respectively; Repeat the decomposition process to convert the low-frequency components Continue to decompose to obtain higher-level low-frequency and high-frequency components, and the calculation expression is: ; In the formula, Indicates The low-frequency components of the layer, Indicates The high-frequency components of the layer, Indicates the number of layers; According to the low-frequency and high-frequency components obtained after Haar wavelet transform, the attribute characteristic index of the real estate element area is calculated. The calculation expression is: ; In the formula, Represents the attribute characteristic index within the real estate element area, and is the preset weight coefficient, and and The sum of is 1; The process of obtaining the environment-related characteristic index is as follows: Extracting environmental related features of the real estate element area, wherein the environmental related features include: surrounding facilities and road network; Constructing the environment feature vector , Indicates The area corresponding to each collection point; Perform Fourier transform on the environmental feature vector to convert the time domain signal into the frequency domain signal. The calculation expression is: ; in, Indicates the frequency domain signal at The value of the frequency point, Indicates the frequency point, Indicates the number of environmental characteristic data, Indicates the total amount of environmental characteristic data, represents the imaginary unit, Represents the first Environmental characteristic data; Calculate the amplitude of the frequency component after Fourier transformation. The calculation expression is: ; in, For the The amplitude of the frequency component; Use the mean calculation expression to calculate the average frequency of all frequency components, and use the difference between the maximum frequency value and the minimum frequency value to calculate the frequency range value; Calculate the environmental association characteristic index, the calculation expression is: ; In the formula, represents the average frequency of all frequency components, Indicates the frequency range, Represents the environmental association characteristic index.

[0028] Comprehensive analysis is conducted on the spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index. Based on the analysis results, a comprehensive score for each real estate element area is generated, including: The spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index of the real estate element area are obtained, and the spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index are normalized and calculated to obtain the comprehensive score of the real estate element area. The calculation expression is: ; In the formula, Represents the comprehensive score of the real estate element area, Indicates the attribute characteristic index of the real estate element area, Indicates the spatial distribution characteristic index of the real estate element area, Indicates the environmental correlation characteristic index of the real estate element area, , and is the preset scale factor, and , and Both are greater than 0.

[0029] Based on the comprehensive score, the real estate element area is divided into important areas and ordinary areas, including: Determine whether the comprehensive score of each real estate element area is greater than or equal to the preset threshold. If so, it means that the corresponding area is an important area and is marked as an important area. If not, it means that the corresponding area is an ordinary area and is marked as an ordinary area.

[0030] Combined with real-time mapping and dynamic adjustment mechanism, the data collection frequency and processing parameters of different types of real estate elements are optimized to achieve intelligent data processing for different mapping targets, including: According to the spatial distribution characteristics, attribute characteristics and environmental association characteristics of different real estate element areas, a classification model is established to classify the real estate element areas into multiple categories, and the classification basis is not limited to the type, area, form and surrounding environment of the real estate; Dynamically adjust the frequency of data collection based on the classification results of each real estate element area. Specifically, for important areas (such as large commercial areas and high-risk areas), increase the frequency of data collection and improve the accuracy of collection; for ordinary areas (such as residential areas or open areas), reduce the frequency of collection and reduce unnecessary redundant data collection; According to the classification results and real-time surveying and mapping data of each real estate element area, optimize the data processing parameters. For example, for key areas (such as complex building areas or traffic-intensive areas), adjust the fine-grainedness of spatial distribution feature analysis and use more convolutional layers and pooling layers; for the attribute analysis part, adjust the layer depth of Haar wavelet transform according to the physical properties of the area; for environmental correlation feature analysis, adjust the calculation range of Fourier transform according to the complexity of surrounding facilities; Based on the feedback data from real-time mapping, the data collection frequency and processing parameters of different categories of areas are continuously optimized through machine learning algorithms. Specifically, it includes: using adaptive learning rate to optimize the feature extraction strategy of each area; Based on historical data and surveying accuracy feedback, adjust the model parameters to improve the balance between processing accuracy and real-time performance; Through adaptive mechanisms, the characteristic change trends of different regions are evaluated and the computational complexity is dynamically adjusted to ensure the efficiency and accuracy of real-time data processing; Optimize the computing performance of the entire system based on the dynamically adjusted data collection frequency and processing parameters to ensure that there are no data bottlenecks in the real-time data analysis and feedback process in different areas, and ensure that the system can process large-scale real estate data and respond in a timely manner; Generate a real-time feedback mechanism to automatically update the feature classification of real estate areas based on new collected data and processing results, gradually improving surveying and mapping efficiency and accuracy.

[0031] See also Figure 2As shown, the multi-dimensional data comprehensive processing system for real estate surveying and mapping includes: A data acquisition module, which uses a laser radar to collect multi-dimensional data of real estate, including terrain data, spatial data, and attribute data, and pre-processes the collected data to reduce noise and optimize the integrity and accuracy of the data; The element area recognition module uses a data segmentation algorithm to identify and separate element areas in real estate, including: boundary lines, building outlines, and terrain features. A feature extraction module, wherein the feature extraction module analyzes the element region and extracts the spatial distribution characteristics, attribute characteristics and environmental association characteristics of the element region; A characteristic index calculation module, wherein the characteristic index calculation module analyzes the spatial distribution characteristics, attribute characteristics and environmental association characteristics of each element area respectively, and calculates the spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index in each element area according to the analysis results; A comprehensive analysis module, which performs a comprehensive analysis on the spatial distribution characteristic index, the attribute characteristic index and the environmental association characteristic index, and generates a comprehensive score for each real estate element area according to the analysis results; A region division module, wherein the region division module divides the real estate element region into an important region and a common region according to the comprehensive score; A dynamic adjustment module combines real-time surveying and mapping with a dynamic adjustment mechanism to optimize the data collection frequency and processing parameters of different categories of real estate element areas, thereby realizing intelligent data processing for different surveying and mapping targets.

[0032] The working principle of the present invention is to collect the terrain data, spatial data and attribute data of real estate through laser radar technology, and combine innovative data processing and analysis methods to achieve high-precision and comprehensive evaluation of real estate areas. Specifically, firstly, the three-dimensional point cloud data of the real estate area is collected by laser radar equipment, and the noise is removed and the data quality is optimized through preprocessing steps. Then, the data segmentation algorithm is used to identify the area of ​​the data, separate the element area in the real estate, and extract the spatial distribution characteristics, attribute characteristics and environmental association characteristics of each area. In terms of spatial feature extraction, the present invention adopts an adaptive convolutional neural network to dynamically adjust the convolution kernel and pooling factor according to the feature complexity of the region, and accurately extract geometric information such as building outlines and terrain features; in attribute feature analysis, a multi-level joint learning network is used to combine spatial features with attribute data, such as building area, floor height, etc., to perform deep mapping and regression analysis to improve prediction accuracy; in terms of environmental feature extraction, the present invention introduces the combination of graph convolutional network and convolutional neural network to handle the complex relationship between real estate and the surrounding environment, and dynamically adjust the analysis range of environmental data to update the relevance of real estate areas in real time. Finally, by calculating the spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index, the comprehensive score of each real estate area is normalized, and the data collection frequency and processing parameters of different areas are dynamically adjusted in combination with real-time surveying and mapping data to achieve intelligent data processing. This technical solution not only improves the surveying and mapping accuracy, but also can efficiently respond to the surveying and mapping needs of different types of real estate, ensuring the real-time and efficiency of the system.

[0033] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0034] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A multi-dimensional data comprehensive processing method for real estate surveying and mapping, characterized in that: The following steps are involved: Use LiDAR to collect multi-dimensional data of real estate, including terrain data, spatial data and attribute data, and pre-process the collected data to reduce noise and optimize data integrity and accuracy; Identify and separate the element areas in real estate through data segmentation algorithms, including: boundary lines, building outlines, terrain features, analyze the element areas, and extract the spatial distribution characteristics, attribute characteristics, and environmental association characteristics of the element areas; Analyze the spatial distribution characteristics, attribute characteristics and environmental association characteristics of each element area respectively, and calculate the spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index in each element area according to the analysis results; Among them, spatial distribution characteristics include: location coordinates and shape, attribute characteristics include: area, volume and use, and environmental association characteristics include: surrounding facilities and road network; Comprehensively analyze the spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index, and generate a comprehensive score for each real estate element area based on the analysis results; Based on the comprehensive score, the real estate element areas are divided into important areas and ordinary areas; Combining real-time surveying and mapping with dynamic adjustment mechanisms, the data collection frequency and processing parameters of different types of real estate element areas are optimized to achieve intelligent data processing for different surveying and mapping targets.

2. The multi-dimensional data comprehensive processing method for real estate surveying and mapping according to claim 1 is characterized in that: The method of identifying and separating the element areas in the real estate by using the data segmentation algorithm specifically includes: Multi-scale segmentation of real estate data collected by LiDAR, including terrain data, spatial data and attribute data, using local area feature coefficients Preprocess and optimize data; Among them, the calculation expression of the local area characteristic coefficient is: ; In the formula, Indicates The attribute value of each data point, Indicates the collected data points. represents the average attribute value of the region, is the total number of data points, is the local area characteristic coefficient, Indicates the coordinates of the area location; The region is segmented using geometric analysis methods, and the normal vector and local curvature of each data point are calculated; The calculation expression of the normal vector is: ; The calculation expression of local curvature is: ; In the formula, represents the normal vector of the data point, represents the Laplacian of the data points, represents the gradient of a data point, Represents the local curvature of a data point; According to the normal vector and local curvature of each data point, the area score of each area is calculated to identify the feature area. The calculation expression is: ; In the formula, represents the area score, and represents the preset scale factor, and and All are greater than 0; It is determined whether the area score of each area is greater than or equal to a preset threshold. If so, it is identified as a feature area; if not, it is identified as a non-feature area.

3. The multi-dimensional data comprehensive processing method for real estate surveying and mapping according to claim 1 is characterized in that: The analysis of the element region to extract the spatial distribution characteristics, attribute characteristics and environmental association characteristics of the element region specifically includes: The spatial distribution features of real estate areas are extracted through multiple convolutional layers and pooling layers of the convolutional neural network, including the geometric features of location, shape, boundary and outline, including: Use convolution kernels to perform convolution operations on real estate area data to extract building boundaries, terrain features, and building outlines; Through adaptive pooling operations, the pooling factor and convolution kernel size are dynamically adjusted, so that the convolutional neural network can effectively extract spatial features at different scales; Through the joint learning network, the spatial data and attribute data of real estate are integrated to extract the attribute characteristics of the real estate area, including: While using the convolutional layer to extract spatial features, the fully connected layer is used to map the spatial features with the attribute data to predict the attributes of the building; Regression analysis is used to predict the physical attributes of real estate areas; Through graph theory and graph convolutional networks combined with convolutional neural networks, the correlation features between real estate areas and their surrounding environments are extracted, including: The real estate area and its surrounding environment are modeled as graph data, and the graph convolutional network is used to extract the correlation features between the real estate and the surrounding environment, and the shortest path and accessibility environmental factors between the real estate and surrounding facilities are calculated; The environmental impact characteristics of real estate areas are extracted by jointly learning environmental data and the spatial features output by convolutional neural networks.

4. The multi-dimensional data comprehensive processing method for real estate surveying and mapping according to claim 1 is characterized in that: The process of obtaining the spatial distribution characteristic index is as follows: Extract the spatial distribution characteristics of the feature area, including location, shape, boundary and contour geometric features, and generate a set of spatial data points , where each point Represents the spatial coordinates of a feature in the property area, Indicates the number of collection points, Indicates The location coordinates of the collection points, Indicates the total number of collection points; For the extracted spatial data point set Apply the density clustering algorithm and preset the neighborhood radius as , and set the density threshold to divide the spatial data points into multiple clustering areas according to their density relationship , where each cluster area Include Feature points ; Calculate the spatial distribution characteristic parameters of each cluster area; Calculate the area of ​​the region, the calculation expression is: ; in, Represents cluster area The area of Convex hull boundary points representing the region; Calculate the regional density, the calculation expression is: ; In the formula, Indicates The regional density of the collection points, Represents cluster area The number of points within Calculate the morphological complexity, the calculation expression is: ; In the formula, Represents the fractal dimension of the region; According to the comprehensive calculation of the regional area, regional density and morphological complexity of each cluster region, the spatial distribution characteristic index of the region is calculated as follows: ; In the formula, Indicates the value of the maximum area among all cluster regions. represents the maximum density value among all cluster areas, An index representing the spatial distribution characteristics of clustering areas.

5. The multi-dimensional data comprehensive processing method for real estate surveying and mapping according to claim 1 is characterized in that: The process of obtaining the attribute characteristic index is as follows: Extract attribute characteristics of feature areas, including area, volume and usage; Construct attribute features into an attribute vector ; For the attribute vector Perform Haar wavelet transform to decompose the attribute features into low-frequency components and high-frequency components. Indicates The areas corresponding to the collection points include: Perform recursive Haar wavelet transform on the attribute data and transform the data vector Decomposed into multiple levels of low-frequency and high-frequency components, the calculation expression is: ; In the formula, and are two adjacent attribute data, and are the low-frequency and high-frequency components of layer 1, respectively; Repeat the decomposition process to convert the low-frequency components Continue to decompose to obtain higher-level low-frequency and high-frequency components, and the calculation expression is: ; In the formula, Indicates The low-frequency components of the layer, Indicates The high-frequency components of the layer, Indicates the number of layers; According to the low-frequency and high-frequency components obtained after Haar wavelet transform, the attribute characteristic index of the real estate element area is calculated. The calculation expression is: ; In the formula, Represents the attribute characteristic index within the real estate element area, and is the preset weight coefficient, and and The sum of is 1.

6. The multi-dimensional data comprehensive processing method for real estate surveying and mapping according to claim 1 is characterized in that: The process of obtaining the environment-related characteristic index is as follows: Extracting environmental related features of the real estate element area, wherein the environmental related features include: surrounding facilities and road network; Constructing the environment feature vector , Indicates The area corresponding to each collection point; Perform Fourier transform on the environmental feature vector to convert the time domain signal into the frequency domain signal. The calculation expression is: ; in, Indicates the frequency domain signal at The value of the frequency point, Indicates the frequency point, Indicates the number of environmental characteristic data, Indicates the total amount of environmental characteristic data, represents the imaginary unit, Represents the first Environmental characteristic data; Calculate the amplitude of the frequency component after Fourier transformation. The calculation expression is: ; in, For the The amplitude of the frequency component; Use the mean calculation expression to calculate the average frequency of all frequency components, and use the difference between the maximum frequency value and the minimum frequency value to calculate the frequency range value; Calculate the environmental association characteristic index, the calculation expression is: ; In the formula, represents the average frequency of all frequency components, Indicates the frequency range, Represents the environmental association characteristic index.

7. The multi-dimensional data comprehensive processing method for real estate surveying and mapping according to claim 1 is characterized in that: The spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index are comprehensively analyzed, and a comprehensive score of each real estate element area is generated according to the analysis results, specifically including: The spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index of the real estate element area are obtained, and the spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index are normalized and calculated to obtain a comprehensive score of the real estate element area.

8. The multi-dimensional data comprehensive processing method for real estate surveying and mapping according to claim 1 is characterized in that: According to the comprehensive score, the real estate element area is divided into important areas and ordinary areas, including: Determine whether the comprehensive score of each real estate element area is greater than or equal to the preset threshold. If so, it means that the corresponding area is an important area and is marked as an important area. If not, it means that the corresponding area is an ordinary area and is marked as an ordinary area.

9. The multi-dimensional data comprehensive processing method for real estate surveying and mapping according to claim 1 is characterized in that: The above-mentioned combination of real-time surveying and dynamic adjustment mechanism optimizes the data collection frequency and processing parameters of different types of real estate element areas, and realizes intelligent data processing for different surveying and mapping targets, specifically including: According to the spatial distribution characteristics, attribute characteristics and environmental association characteristics of different real estate element areas, a classification model is established to classify the real estate element areas into multiple categories, and the classification basis is not limited to the type, area, form and surrounding environment of the real estate; Dynamically adjust the frequency of data collection based on the classification results of each real estate element area; Optimize data processing parameters based on the classification results and real-time surveying and mapping data of each real estate element area; Based on the feedback data from real-time surveying and mapping, the data collection frequency and processing parameters of different categories of areas are continuously optimized through machine learning algorithms; Adjust the model parameters based on historical data and surveying accuracy feedback; Through adaptive mechanisms, the characteristic change trends of different regions are evaluated and the computational complexity is dynamically adjusted; Optimize the computing performance of the entire system based on the dynamically adjusted data acquisition frequency and processing parameters; Generate a real-time feedback mechanism to automatically update the feature classification of real estate areas based on new collected data and processing results.

10. A multi-dimensional data integrated processing system for real estate surveying and mapping, characterized in that: The multi-dimensional data comprehensive processing method for real estate surveying and mapping as claimed in any one of claims 1 to 9 comprises: A data acquisition module, which uses a laser radar to collect multi-dimensional data of real estate, including terrain data, spatial data, and attribute data, and pre-processes the collected data to reduce noise and optimize the integrity and accuracy of the data; The element area recognition module uses a data segmentation algorithm to identify and separate element areas in real estate, including: boundary lines, building outlines, and terrain features. A feature extraction module, wherein the feature extraction module analyzes the element region and extracts the spatial distribution characteristics, attribute characteristics and environmental association characteristics of the element region; A characteristic index calculation module, wherein the characteristic index calculation module analyzes the spatial distribution characteristics, attribute characteristics and environmental association characteristics of each element region respectively, and calculates the spatial distribution characteristic index, attribute characteristic index and environmental association characteristic index in each element region according to the analysis results; A comprehensive analysis module, which performs a comprehensive analysis on the spatial distribution characteristic index, the attribute characteristic index and the environmental association characteristic index, and generates a comprehensive score for each real estate element area according to the analysis results; A region division module, wherein the region division module divides the real estate element region into an important region and a common region according to the comprehensive score; A dynamic adjustment module combines real-time surveying and mapping with a dynamic adjustment mechanism to optimize the data collection frequency and processing parameters of different categories of real estate element areas, thereby realizing intelligent data processing for different surveying and mapping targets.

Images