A data processing method for the spatial and traffic diversification characteristics of mountain towns

Abstract

The present application provides a method for processing data on the diversified characteristics of mountain town space and traffic, which relates to the technical field of electrical digital data processing. Remote sensing data is collected at different sampling heights, and then the building characteristics and traffic characteristics of mountain towns are extracted. Hierarchical clustering is performed on all the building characteristics and traffic characteristics respectively to obtain a building characteristic cluster set and a traffic characteristic cluster set; the reward heterogeneity of mountain towns in the spatial dimension at different sampling heights is determined based on a feature reward mechanism of reinforcement learning; spectral difference analysis is performed on the remote sensing data at different sampling heights to obtain spectral response differences; a reward factor for multi-feature reinforcement is determined through the reward heterogeneity and spectral response differences, and all the remote sensing data is fused according to the reward factor to obtain the fusion characteristics of the mountain town space and traffic distribution. By adopting the solution of the present application, the enhanced fusion of multi-features at different sampling heights can be realized, thereby improving the accuracy of the analysis of mountain town space and traffic characteristics.

Description

Technical Field

[0001] This application relates to the technical field of electronic digital data processing. More specifically, this application relates to a method for processing data on the diverse characteristics of the space and transportation in mountain towns. Background Art

[0002] Due to their complex topography and unique ecological environment, mountain towns face many challenges in urban planning, transportation design, and spatial layout. The spatial and transportation characteristics of mountain towns often have diverse features, including multiple clusters, multiple centers, multiple dimensions, and multiple levels. These characteristics not only affect the overall function of the city but also have a profound impact on aspects such as transportation, economic production, social culture, and ecological environment protection.

[0003] Traditional data processing methods have certain limitations in capturing the characteristics of mountain towns. The early manual mapping method, although able to reflect land adaptability through manual analysis, is difficult to effectively cope with the dynamic changes of complex terrain and large-scale data. With the wide application of GIS (Geographic Information System) technology, the ability of spatial data analysis has been significantly improved. However, conventional GIS methods mostly focus on two-dimensional feature extraction and are difficult to fully integrate high-dimensional and multi-source data. Especially when facing multi-sampling heights and multi-spectral data, the ability of feature fusion and heterogeneous expression is insufficient. By analyzing the spectral responses at different sampling heights and fusing multi-source features, the insufficient expression of complexity in traditional methods can be effectively solved, and the accuracy of spatial and transportation feature analysis can be improved. Therefore, how to achieve the enhanced fusion of diverse features at different sampling heights to improve the accuracy of spatial and transportation feature analysis in mountain towns. Summary of the Invention

[0004] This application provides a method for processing data on the diverse characteristics of the space and transportation in mountain towns, which can achieve the enhanced fusion of diverse features at different sampling heights, thereby improving the accuracy of spatial and transportation feature analysis in mountain towns.

[0005] This application provides a method for processing data on the diverse characteristics of the space and transportation in mountain towns. The data processing method includes the following steps:

[0006] Collect remote sensing data of mountain towns at different sampling heights and map all the remote sensing data into a multi-feature space;

[0007] Extract the building features and transportation features of mountain towns at different sampling heights from the multi-feature space, and then perform hierarchical clustering on all the building features and transportation features respectively to obtain a building feature cluster set and a transportation feature cluster set;

[0008] The reward information between the building feature clusters and the traffic feature clusters is determined based on a reinforcement learning-based feature reward mechanism, and the reward heterogeneity of mountain towns in the spatial dimension at different sampling heights is determined through the reward information.

[0009] Spectral difference analysis is performed on the remotely sensed data collected at different sampling heights according to the distribution difference degree of remote sensing pixels in each remotely sensed data, and the spectral response difference at different sampling heights is obtained.

[0010] The reward factor of multi-feature reinforcement is determined through the reward heterogeneity and the spectral response difference, and all the collected remotely sensed data are fused according to the reward factor to obtain the fusion features of the spatial and traffic distributions of mountain towns.

[0011] In this embodiment, multi-spectral remote sensing technology is used to collect remotely sensed data of mountain towns at different heights.

[0012] In this embodiment, mapping all the remotely sensed data into a multi-feature space specifically includes:

[0013] All the remotely sensed data are aligned to obtain all the aligned remotely sensed data.

[0014] The principal component analysis algorithm is used to map all the aligned remotely sensed data into a low-dimensional feature space to obtain a multi-feature space.

[0015] In this embodiment, extracting the building features and traffic features of mountain towns at different sampling heights from the multi-feature space specifically includes:

[0016] For each sampling height, a convolutional neural network model is used to identify the edge features of buildings in the multi-feature space to obtain the building features of mountain towns at this sampling height.

[0017] A road detection algorithm is used to extract the traffic features of mountain towns at this sampling height from the multi-feature space.

[0018] In this embodiment, hierarchical clustering is performed on all the building features and traffic features respectively to obtain building feature clusters and traffic feature clusters, which specifically includes:

[0019] Based on the minimum distance mechanism, hierarchical clustering is performed on all the building features and traffic features respectively to obtain building feature clusters and traffic feature clusters with different density levels.

[0020] The building feature clusters are used to obtain the building feature cluster set.

[0021] The traffic feature clusters are used to obtain the traffic feature cluster set.

[0022] In this embodiment, determining the reward information between the building feature cluster set and the traffic feature cluster set based on the feature reward mechanism of reinforcement learning specifically includes:

[0023] Determine the correlation matrix between the building feature cluster set and the traffic feature cluster set;

[0024] Based on the feature reward mechanism of reinforcement learning, perform reward adjustment on the matrix elements in the correlation matrix that are greater than the preset threshold, and perform penalty adjustment on the matrix elements in the correlation matrix that are less than the preset threshold;

[0025] Use the adjusted correlation matrix to represent the reward information between the building feature cluster set and the traffic feature cluster set.

[0026] In this embodiment, determining the reward heterogeneity of mountain towns in the spatial dimension at different sampling heights through the reward information specifically includes:

[0027] Obtain the difference amount of the spatial distribution of mountain towns at different sampling heights;

[0028] Determine the reward heterogeneity of mountain towns in the spatial dimension at different sampling heights according to the reward information and the difference amount of the spatial distribution.

[0029] In this embodiment, performing spectral difference analysis on the remotely sensed data collected at different sampling heights according to the distribution difference degree of the remote sensing pixels in each remotely sensed data, and obtaining the spectral response difference at different sampling heights specifically includes:

[0030] Use the spectral characteristics of the remote sensing pixels to determine the spectral response value of the remote sensing image at each sampling height;

[0031] Determine the spectral response difference at different sampling heights through all the spectral response values and the distribution difference degree of the remote sensing pixels in each remotely sensed data.

[0032] In this embodiment, fusing all the collected remotely sensed data into the fusion features of the spatial and traffic distributions of mountain towns according to the reward factor specifically includes:

[0033] Weightedly fuse the remotely sensed data at different sampling heights according to the reward factor to generate a fusion image with high spatial resolution;

[0034] Extract the fusion features of the spatial and traffic distributions of mountain towns from the fusion image based on a pre-trained feature extraction model.

[0035] In this embodiment, before mapping all the remotely sensed data into a multi-feature space, noise reduction processing is also included for all the remotely sensed data.

[0036] The technical solutions provided by the disclosed embodiments of the present application have the following beneficial effects:

[0037] By collecting remote sensing data of mountain towns at different sampling heights, all the remote sensing data are mapped into a multi - feature space; building features and traffic features of mountain towns at different sampling heights are extracted from the multi - feature space, and then hierarchical clustering is performed on all the building features and traffic features respectively to obtain a building feature cluster set and a traffic feature cluster set; a reward information between the building feature cluster set and the traffic feature cluster set is determined based on a feature reward mechanism of reinforcement learning, and the reward heterogeneity of mountain towns in the spatial dimension at different sampling heights is determined through the reward information; spectral difference analysis is performed on the remote sensing data collected at different sampling heights according to the distribution difference degree of remote sensing pixels in each remote sensing data to obtain the spectral response difference at different sampling heights; a reward factor for multi - feature reinforcement is determined through the reward heterogeneity and the spectral response difference, and all the collected remote sensing data are fused according to the reward factor to obtain the fusion features of the spatial and traffic distributions of mountain towns.

[0038] It can be seen that by strengthening the fusion of multi - features at different sampling heights, the present application can improve the accuracy of the analysis of the spatial and traffic features of mountain towns; First, by mapping the remote sensing data at different sampling heights to form a multi - feature space, and then extracting building and traffic features and performing hierarchical clustering on them, functional units and traffic networks at different spatial scales can be automatically identified from the data. The use of hierarchical clustering can ensure that the differences between different features can be clearly distinguished and effectively utilized, thereby improving the applicability of the model; Then, the feature reward mechanism based on reinforcement learning provides intelligent adjustment for the relationship between building features and traffic features, ensuring that the matching relevance between different feature clusters is fully explored. This mechanism can not only enhance the expression ability of spatial features, but also improve the processing ability of various data differences in complex environments through reward heterogeneity, further optimizing the effect of feature fusion. In addition, spectral response difference analysis provides a scientific basis for feature fusion by identifying the distribution differences of remote sensing pixels at different sampling heights, ensuring the accurate capture of spectral features during the fusion process; Finally, by combining reward heterogeneity and spectral response difference to determine the reward factor of multi - features, weighted fusion of multi - source remote sensing data can be achieved, and fusion features with high spatial resolution can be generated. The fusion features can not only accurately reflect the spatial layout and traffic distribution of mountain towns, but also adapt to dynamic environments and provide more accurate and detailed analysis results. In summary, the solution of the present application can achieve the enhanced fusion of multi - features at different sampling heights, thereby improving the accuracy of the analysis of the spatial and traffic features of mountain towns. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only the embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0040] Figure 1 It is an exemplary flowchart of a method for processing data on the characteristics of the space and transportation in mountainous towns provided by the present application;

[0041] Figure 2 It is an exemplary flowchart of determining a multi - feature space provided by the present application;

[0042] Figure 3 It is an exemplary flowchart of determining the spectral response difference provided by the present application. Detailed implementation manners

[0043] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0044] The embodiments of the present application provide a method for processing data on the characteristics of the space and transportation in mountainous towns. The core is to collect remote - sensing data of mountainous towns at different sampling heights, map all the remote - sensing data into a multi - feature space; extract the building characteristics and transportation characteristics of mountainous towns at different sampling heights from the multi - feature space, and then perform hierarchical clustering on all the building characteristics and transportation characteristics respectively to obtain a building - characteristic cluster set and a transportation - characteristic cluster set; determine the reward information between the building - characteristic cluster set and the transportation - characteristic cluster set based on a feature - reward mechanism of reinforcement learning, and determine the reward heterogeneity of mountainous towns in the spatial dimension at different sampling heights through the reward information; perform spectral - difference analysis on the remote - sensing data collected at different sampling heights according to the distribution difference degree of remote - sensing pixels in each remote - sensing data to obtain the spectral response difference at different sampling heights; determine the reward factor for multi - feature reinforcement through the reward heterogeneity and the spectral response difference, and fuse all the collected remote - sensing data according to the reward factor to obtain the fusion characteristics of the space and transportation distribution in mountainous towns. By adopting the above - mentioned scheme, the enhanced fusion of multi - features at different sampling heights can be realized, thereby improving the accuracy of the analysis of the space and transportation characteristics of mountainous towns.

[0045] To better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings of the specification and specific embodiments. Refer to Figure 1 As shown, this figure is an exemplary flowchart of a method for processing data on the spatial and traffic diversification characteristics of mountain towns according to an embodiment of the present application. The data processing method includes the following steps:

[0046] In step S1, remote sensing data of mountain towns is collected at different sampling heights, and all the remote sensing data is mapped into a multi-feature space.

[0047] In specific implementation, multi-spectral remote sensing technology can be used to collect remote sensing data of mountain towns at different heights. It should be noted that the remote sensing data at different heights in this application provides different spatial resolutions and perspectives, allowing the observation of mountain towns from both macroscopic and microscopic angles. Among them, the remote sensing data includes spectral data and spatial data in different bands. It should also be noted that before mapping all the remote sensing data into a multi-feature space in this application, noise reduction processing is also included for all the remote sensing data. The mean filtering technology can be used to perform noise reduction processing on all the remote sensing data, which will not be elaborated here.

[0048] Preferably, in this embodiment, refer to Figure 2 As shown, this figure is an exemplary flowchart for determining a multi-feature space in an embodiment of the present application. In this embodiment, in this embodiment, mapping all the remote sensing data into a multi-feature space can be specifically implemented in the following manner, that is:

[0049] First, in step S11, all the remote sensing data is aligned to obtain all the aligned remote sensing data;

[0050] Then, in step S12, the principal component analysis algorithm is used to map all the aligned remote sensing data into a low-dimensional feature space to obtain a multi-feature space.

[0051] T In specific implementation, first, ground control points can be used for image registration, and all remote sensing data is projected into a unified geographic coordinate system to obtain spatially unified corresponding remote sensing data; then, the principal component analysis technology is used to flatten all the aligned remote sensing data (transformed from a multi-dimensional image matrix into a two-dimensional matrix form, where each row represents a pixel and each column represents a feature, such as spectral bands, texture features, etc.), and then the covariance matrix of all features is calculated to quantify the correlation between features. Then, eigenvalue decomposition is performed on the covariance matrix, and the main feature components are selected through the correlation between features to obtain multi-feature data. Finally, a low-dimensional multi-feature space is constructed with the dimensions of multi-feature data at different sampling heights.

[0052] In step S2, building features and traffic features of mountain towns at different sampling heights are extracted from the multi - feature space, and then hierarchical clustering is performed on all the building features and traffic features respectively to obtain a building feature cluster set and a traffic feature cluster set.

[0053] In this embodiment, the extraction of building features and traffic features of mountain towns at different sampling heights from the multi - feature space can be specifically implemented in the following way, that is:

[0054] For each sampling height, a convolutional neural network model is used to identify the edge features of buildings in the multi - feature space to obtain the building features of mountain towns at this sampling height;

[0055] A road detection algorithm is used to extract the traffic features of mountain towns at this sampling height from the multi - feature space.

[0056] It should be noted that the convolutional neural network model in this application is used to extract the edge features of buildings. Specifically, for each sampling height, first, the convolutional layer of the convolutional neural network model can automatically extract spatial features, such as the geometric shape and edge features of buildings, and then the edge features of the output building features of the convolutional neural network model are obtained, and the extracted edge features are used as the building features of mountain towns. The building features include building geometric information, building density information, and building function information; then, the Canny edge detection algorithm in the road detection algorithm can be used to extract traffic features from the multi - feature space. The traffic features include road geometric information, road network topology information, and traffic flow information.

[0057] In this embodiment, the hierarchical clustering of all the building features and traffic features respectively to obtain a building feature cluster set and a traffic feature cluster set can be specifically implemented in the following way, that is:

[0058] Based on the minimum - distance mechanism, hierarchical clustering is performed on all the building features and traffic features respectively to obtain building feature clusters and traffic feature clusters at different density levels;

[0059] A building feature cluster set is obtained through all the building feature clusters;

[0060] A traffic feature cluster set is obtained through all the traffic feature clusters.

[0061] In specific implementation, each building feature can be regarded as a separate cluster, and the building density in the building features is used as the clustering index, that is, the distance of the building density between the building features is calculated. Based on the principle of the minimum distance, the two closest clusters are gradually merged until a preset condition is reached, and multiple building feature clusters are obtained. Then, all the building feature clusters are combined into a set as the building feature cluster set. Similarly, the traffic flow in the traffic features can be used as the clustering index, and based on the principle of the minimum distance, the two closest clusters are gradually merged until a preset condition is reached, and multiple traffic feature clusters are obtained. Then, all the traffic feature clusters are combined into a set as the traffic feature cluster set. It should be noted that the preset condition in this application can be set to divide all features into 4 clusters. In other embodiments, the preset condition can also be set to others, which is not limited here.

[0062] In step S3, based on the feature reward mechanism of reinforcement learning, the reward information between the building feature cluster set and the traffic feature cluster set is determined, and the reward heterogeneity of the mountain town in the spatial dimension at different sampling heights is determined through the reward information.

[0063] In this embodiment, the method for determining the reward information between the building feature cluster set and the traffic feature cluster set based on the feature reward mechanism of reinforcement learning can specifically adopt the following method, that is:

[0064] Determine the association matrix between the building feature cluster set and the traffic feature cluster set;

[0065] Based on the feature reward mechanism of reinforcement learning, the matrix elements in the association matrix greater than the preset threshold are adjusted for reward, and the matrix elements in the association matrix less than the preset threshold are adjusted for penalty;

[0066] The adjusted association matrix is used to represent the reward information between the building feature cluster set and the traffic feature cluster set.

[0067] In specific implementation, first, the building feature clusters and the traffic feature clusters can be input into a pre-trained association model, and an association matrix between the building feature clusters and the traffic feature clusters is output through the association model. The association values in the association matrix describe the matching relationship between the building feature clusters in the building feature clusters and the traffic feature clusters in the traffic feature clusters. For example, the matching degree between traffic flow and building density. Then, a reinforcement learning model is defined. The current association matrix is used as the initial state parameter of the reinforcement learning model. The reinforcement learning model rewards the matching relationship between the building and the traffic through a customized reward mechanism, and an association threshold is set to distinguish the matrix elements with high association and low association. Further, the association matrix is rewarded through the reinforcement learning model, that is, the matrix elements greater than the association threshold in the association matrix are adjusted for reward, and the matrix elements less than the association threshold in the association matrix are adjusted for punishment. It should be noted that the adjustment mechanism in this application can be dynamically adjusted according to the matching degree between the building and the traffic. The reward adjustment means increasing on the basis of the current data, and the punishment adjustment means decreasing on the basis of the current data. Finally, the adjusted association matrix is used as the manifestation form of the reward information between the building feature clusters and the traffic feature clusters.

[0068] It should be noted that in this application, by setting a reward mechanism to evaluate the coordination ability of the urban spatial layout and functions, it can provide a good data basis for the optimization of the urban spatial layout.

[0069] In this embodiment, to determine the reward heterogeneity of mountain towns in the spatial dimension at different sampling heights through the reward information, the following method can be specifically adopted, that is:

[0070] Obtain the difference quantity of the spatial distribution of mountain towns at different sampling heights;

[0071] Determine the reward heterogeneity of mountain towns in the spatial dimension at different sampling heights according to the reward information and the difference quantity of the spatial distribution.

[0072] In specific implementation, first, remote sensing data at different sampling heights can be subjected to multi - feature mapping to obtain a feature space corresponding to each sampling height. Then, the average value of the similarity between the feature space at each sampling height and the feature spaces at other sampling heights is used as the difference measure of the spatial distribution of mountain towns at this sampling height. Here, the similarity can be quantified using the Euclidean distance between the feature spaces, and no specific limitation is made here. Then, for each sampling height, the reciprocal of the difference measure of the spatial distribution of mountain towns at this sampling height is used as the reward value at this sampling height, and thus the reward values for all sampling heights are obtained. Then, all the reward values and the correlation matrix in the reward information are used as the input parameters of the reward model. Through the reward model, the differences in different features (i.e., building features and traffic features) in mountain towns in the spatial dimension can be learned. That is, the difference values of the rewards of mountain towns in the spatial dimension at different sampling heights can be output through the reward model, and the reward heterogeneity at different sampling heights of mountain towns in the spatial dimension can be described through this difference value. Through the reward heterogeneity, the characteristic change characteristics of mountain towns at different sampling heights can be intuitively reflected, providing a key basis for optimizing feature fusion. It should also be noted that the reward model in this application is obtained by training a convolutional neural network, and other algorithm structures can also be used in other embodiments, which are not limited here.

[0073] It should be noted that the reward heterogeneity in this application is a characteristic index used to quantify the degree of change in the distribution of building features and traffic features at different sampling heights in the spatial dimension. By calculating the reward heterogeneity and quantifying the differences in mountain towns in the spatial dimension, it can provide guidance for weight assignment in subsequent multi - feature fusion, further improving the accuracy and rationality of remote sensing data processing.

[0074] In step S4, spectral difference analysis is performed on the remote sensing data collected at different sampling heights according to the distribution difference degree of remote sensing pixels in each remote sensing data, and the spectral response differences at different sampling heights are obtained.

[0075] Preferably, in this embodiment, with reference to Figure 3 As shown, this figure is an exemplary flowchart for determining the spectral response difference in the embodiment of this application. In this embodiment, spectral difference analysis is performed on the remote sensing data collected at different sampling heights according to the distribution difference degree of remote sensing pixels in each remote sensing data, and the spectral response differences at different sampling heights can be obtained specifically in the following manner, that is:

[0076] First, in step S41, the spectral response value of the remote sensing image at each sampling height is determined using the spectral characteristics of the remote sensing pixels.

[0077] Then, in step S42, the spectral response differences at different sampling heights are determined through all the spectral response values and the distribution difference degree of remote sensing pixels in each remote sensing data.

[0078] In specific implementation, first, for each sampling height, spectral features of each pixel can be extracted from remote sensing data. The spectral features include multi-band reflectance values (such as visible light, near-infrared, thermal infrared, etc.), and then spectral features of all pixels are obtained. The spectral features of all pixels are statistically analyzed, and the mean spectrum and standard deviation spectrum of the spectral features of all pixels are calculated. Furthermore, the ratio between the mean spectrum and the standard deviation spectrum can be used as the spectral response value of the remote sensing image at the sampling height. Then, the distribution feature values of remote sensing pixels in the remote sensing data at each sampling height can be used as the distribution difference degree of remote sensing pixels in the corresponding remote sensing data. Among them, the distribution feature values can be quantified using the distribution gradient, which will not be elaborated here. Finally, the product of the spectral response value of the remote sensing image at each sampling height and the distribution difference degree of remote sensing pixels in the corresponding remote sensing data is used as the spectral response difference at each sampling height, and the array composed of the spectral response differences at all sampling heights is used as the spectral response difference at different sampling heights.

[0079] It should be noted that the spectral response difference in this application measures the difference in the spectral characteristics of remote sensing images at different sampling heights. It should also be noted that through the comprehensive analysis of the spectral response value and the distribution difference degree, this application accurately characterizes the spectral difference characteristics of remote sensing data at different sampling heights, which can provide data support for the generation of fusion features and help improve the processing accuracy of the spatial and traffic distribution characteristics of mountain towns.

[0080] In step S5, a reward factor for multi-feature reinforcement is determined through the reward heterogeneity and the spectral response difference, and all the collected remote sensing data are fused according to the reward factor to obtain the fusion features of the spatial and traffic distribution of mountain towns.

[0081] In this embodiment, the following method can be specifically adopted to determine the reward factor for multi-feature reinforcement through the reward heterogeneity and the spectral response difference, that is:

[0082] The spectral response difference is used as the initialization parameter for processing the spectral response at different sampling heights in the reward model;

[0083] The reward heterogeneity is used as the initialization parameter for processing the multi-feature reward difference of mountain towns in the spatial dimension at different sampling heights in the reward model;

[0084] The reward factor for multi-feature reinforcement at different sampling heights is output through the reward model.

[0085] In specific implementation, first, an array composed of spectral response differences at all sampling heights and the reward difference values of multivariate features (such as building features and traffic features) of mountain towns in the spatial dimension at different sampling heights can be input into the reward model, and the reward model assigns reward factors to the multivariate features at different sampling heights. It should be noted that in this application, a reward mechanism based on reinforcement learning (such as deep reinforcement learning) can be used for the reward model. In other embodiments, other types of reward models can also be adopted, which are not limited here. It should also be noted that the distribution mechanism of the reward factors in this application specifically includes: for the feature combination with high spectral response difference, the reward factor is higher; for the spatial feature with high reward heterogeneity, the reward factor is higher.

[0086] In this embodiment, to specifically obtain the fusion features of the spatial and traffic distributions of the mountain town by fusing all the collected remote sensing data according to the reward factors, the following method can be adopted, that is:

[0087] Weightedly fuse the remote sensing data at different sampling heights according to the reward factors to generate a fusion image with high spatial resolution;

[0088] Extract the fusion features of the spatial and traffic distributions of the mountain town from the fusion image based on a pre-trained feature extraction model.

[0089] In specific implementation, first, perform weighted processing on the remote sensing data at different sampling heights according to the reward factors, assign weights to each layer of remote sensing data, and fuse them to generate an image with high spatial resolution. This step fully considers the spectral response differences and spatial feature reward heterogeneity of the remote sensing data at different heights, ensuring that the fusion image has both high-detail expression ability and global consistency. Then, use a pre-trained deep learning feature extraction model (such as a model based on a convolutional neural network) to process the fused high-resolution image, extract the spatial features of the mountain town such as buildings and roads and the traffic network distribution information contained therein. Finally, output the fusion features through the deep learning feature extraction model. The fusion features cover the spatial layout and traffic network distribution of the mountain town, can be used as a data basis for further analysis (such as spatial optimization planning, traffic load assessment), provide a scientific basis for urban development and planning, and at the same time improve the comprehensive utilization efficiency of multi-source data.

[0090] It can be seen that this application enhances the integration of diverse features at different sampling heights, improving the accuracy of spatial and traffic feature analysis in mountainous towns. First, by mapping remote sensing data at different sampling heights, a multi-feature space is formed. Subsequently, building and traffic features are extracted and hierarchically clustered. This enables the automatic identification of functional units and traffic networks at different spatial scales from the data. The use of hierarchical clustering ensures that the differences between different features can be clearly distinguished and effectively utilized, thereby enhancing the granularity and applicability of the model. Then, the feature reward mechanism based on reinforcement learning provides intelligent adjustment for the relationship between building features and traffic features, ensuring that the correlations between different feature clusters are fully explored. This mechanism not only enhances the expression ability of spatial features but also improves the ability to handle data differences in complex environments by rewarding heterogeneity, further optimizing the effect of feature fusion. In addition, spectral response difference analysis provides a scientific basis for feature fusion by identifying the distribution differences of remote sensing pixels at different sampling heights, ensuring the accurate capture of spectral features during the fusion process. Finally, by combining reward heterogeneity and spectral response differences to determine the reward factors of multi-feature, weighted fusion of multi-source remote sensing data can be achieved, and fused features with high spatial resolution can be generated. These fused features can not only accurately reflect the spatial layout and traffic distribution of mountainous towns but also adapt to dynamically changing environments, providing more accurate and detailed analysis results. In summary, the solution of this application can achieve the enhanced fusion of multi-feature at different sampling heights, thereby improving the accuracy of spatial and traffic feature analysis in mountainous towns.

[0091] This application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams can be implemented by computer program instructions, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the functions specified in Figure 1 one or more of the processes or multiple processes and / or blocks Figure 1 one or more of the blocks or multiple blocks.

[0092] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing relevant hardware through a program, and this program can be stored in a computer-readable storage medium. The storage medium includes read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc memories, magnetic disc memories, tape memories, or any other medium that can be used to carry or store data and is computer-readable.

[0093] It should also be noted that the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or also includes elements inherent in such a process, method, commodity or device. Without further limitations, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, method, commodity or device including the element.

Claims

1. A data processing method for the spatial and traffic diversification characteristics of mountain towns, characterized in that, The described data processing method includes the following steps: Collect remote sensing data of mountain towns at different sampling heights, and map all the remote sensing data into a multi - feature space; Extract the building features and traffic features of mountain towns at different sampling heights from the multi - feature space, and then perform hierarchical clustering on all the building features and traffic features respectively to obtain a building feature cluster set and a traffic feature cluster set; Determine the reward information between the building feature cluster set and the traffic feature cluster set based on a feature reward mechanism of reinforcement learning, and determine the reward heterogeneity of mountain towns in the spatial dimension at different sampling heights through the reward information; Perform spectral difference analysis on the remote sensing data collected at different sampling heights according to the distribution difference degree of remote sensing pixels in each remote sensing data to obtain the spectral response difference at different sampling heights; Determine a reward factor for multi - feature reinforcement through the reward heterogeneity and the spectral response difference, and fuse all the collected remote sensing data according to the reward factor to obtain the fusion features of the spatial and traffic distributions of mountain towns.

2. The method for processing data on the diversified characteristics of the space and transportation in mountain towns as described in claim 1, wherein, Use multi - spectral remote sensing technology to collect remote sensing data of mountain towns at different heights.

3. A method for processing data on the characteristics of the spatial and transportation diversification in mountain towns as described in claim 1, characterized in that Mapping all the remote sensing data into a multi - feature space specifically includes: Align all the remote sensing data to obtain all the aligned remote sensing data; Use the principal component analysis algorithm to map all the aligned remote sensing data into a low - dimensional feature space to obtain a multi - feature space.

4. The method for processing characteristic data of mountain town space and traffic diversification according to claim 1, wherein, Extracting the building features and traffic features of mountain towns at different sampling heights from the multi - feature space specifically includes: For each sampling height, use a convolutional neural network model to identify the edge features of buildings in the multi - feature space to obtain the building features of mountain towns at this sampling height; Use a road detection algorithm to extract the traffic features of mountain towns at this sampling height from the multi - feature space.

5. A method for processing data on the characteristics of the spatial and transportation diversification of mountain towns as described in claim 1, characterized in that, Performing hierarchical clustering on all the building features and traffic features respectively to obtain a building feature cluster set and a traffic feature cluster set specifically includes: Cluster all the building features and traffic features respectively based on the minimum distance mechanism to obtain building feature clusters and traffic feature clusters at different density levels; Obtain a building feature cluster set through all the building feature clusters; Obtain a traffic feature cluster set through all the traffic feature clusters.

6. The method for processing data on the characteristics of the spatial and traffic diversification of mountain towns according to claim 1, wherein, Determining the reward information between the building feature cluster set and the traffic feature cluster set based on a feature reward mechanism of reinforcement learning specifically includes: Determine the correlation matrix between the building feature cluster set and the traffic feature cluster set; Based on the feature reward mechanism of reinforcement learning, perform reward adjustment on the matrix elements in the correlation matrix that are greater than a preset threshold, and perform penalty adjustment on the matrix elements in the correlation matrix that are less than the preset threshold; Use the adjusted correlation matrix to represent the reward information between the building feature cluster set and the traffic feature cluster set.

7. A method for processing data on the characteristics of the spatial and transportation diversification of mountain towns as described in claim 1, characterized in that, Determining the reward heterogeneity of mountain towns in the spatial dimension at different sampling heights through the reward information specifically includes: Obtain the difference amount of the spatial distribution of mountain towns at different sampling heights; Determine the reward heterogeneity of mountain towns in the spatial dimension at different sampling heights according to the reward information and the difference amount of spatial distribution.

8. The method for processing characteristic data of the spatial and traffic diversification of a mountain town as described in claim 1, wherein Perform spectral difference analysis on the remote sensing data collected at different sampling heights according to the distribution difference degree of remote sensing pixels in each remote sensing data, and the specific spectral response differences at different sampling heights include: Use the spectral characteristics of remote sensing pixels to determine the spectral response values of remote sensing images at each sampling height; Determine the spectral response differences at different sampling heights through all spectral response values and the distribution difference degree of remote sensing pixels in each remote sensing data.

9. A method for processing data on the characteristics of spatial and traffic diversification in mountain towns as described in claim 1, characterized in that, Fuse all the collected remote sensing data according to the reward factor to obtain the fusion characteristics of the mountain town space and traffic distribution, which specifically includes: Weightedly fuse the remote sensing data at different sampling heights according to the reward factor to generate a fusion image with high spatial resolution; Extract the fusion characteristics of the mountain town space and traffic distribution from the fusion image based on a pre-trained feature extraction model.

10. A method for processing data on the characteristics of spatial and traffic diversification in mountain towns as described in claim 1, characterized in that, Before mapping all the remote sensing data into a multi-feature space, noise reduction processing is also included for all the remote sensing data.

Images