Data analysis method and system based on land planning surveying and mapping, and electronic equipment

Abstract

The embodiment of the invention relates to a data analysis method and system based on land planning surveying and mapping, and electronic equipment, and belongs to the technical field of electronic data processing and measurement planning, and the method comprises the steps: measuring a target region to obtain a first surveying and mapping image and spatial data, generating a spatial coordinate data set, extracting a multi-dimensional feature value of the first surveying and mapping image, comparing the multi-dimensional feature value with a preset feature threshold value, adaptively selecting a target method from a plurality of quality evaluation methods according to a comparison result, calculating a quantitative quality evaluation value, and screening out a second surveying and mapping image and associated spatial data thereof, and respectively analyzing the space geometric structure information and the analysis image sequence association information, and generating a fusion data result representing the land planning surveying and mapping elements. According to the embodiment of the invention, the automation degree and the result precision of data processing are improved, and efficient and intelligent conversion of the land planning surveying and mapping data from original acquisition to structured achievement is realized.

Description

Technical Field

[0001] This application relates to a data analysis method, system, and electronic device based on land planning and surveying, belonging to the technical fields of electronic data processing and surveying planning. Background Technology

[0002] Urban and rural planning and land space management are important foundations for modern social development. Land planning and surveying rely on high-precision mapping of spatial elements such as land, buildings, and real estate. Current surveying operations mostly rely on single types of surveying instruments (such as total stations and GNSS receivers) for data collection, and generate surveying reports through manual recording and post-processing. Although this method can obtain basic spatial coordinates, it suffers from problems such as separation of data collection and recording, lengthy processing procedures, and poor real-time performance, making it difficult to meet the needs of large-scale, high-time-efficiency planning monitoring.

[0003] With the development of remote sensing and information technology, images and spatial data of surveyed areas can be acquired simultaneously through drones, satellites, and other means. Current technologies typically involve simply overlaying the acquired images and coordinate data, lacking an effective mechanism for evaluating and filtering image quality. This results in databases storing a large number of redundant or low-resolution images, occupying significant storage space and increasing the computational burden of subsequent data processing and analysis. Furthermore, the fusion of image and spatial data largely relies on manual judgment, and a standardized methodology capable of automatically selecting high-quality images and achieving intelligent fusion of multi-source data has not yet been established.

[0004] In conclusion, existing technologies can no longer meet people's needs and urgently need to be improved. Summary of the Invention

[0005] The main objective of this application is to provide a data analysis method, system, and electronic device based on land planning and surveying, in order to solve the technical problems in the prior art, such as difficulty in synchronizing surveying images and spatial data, poor real-time transmission, ineffective intelligent quality assessment and data fusion, and low degree of automation in data processing.

[0006] The embodiments of this application are implemented using the following technical solutions: According to one aspect of the embodiments of this application, a data analysis method based on land planning and mapping is provided, comprising: measuring a target area of ​​land planning and mapping to obtain a first survey image and spatial data; associating the first survey image and the spatial data and mapping them to spatial coordinates to form a spatial coordinate dataset; and uploading the spatial coordinate dataset to a remote data processing terminal in real time; the remote data processing terminal receiving the spatial coordinate dataset, performing spatiotemporal reference unification processing on the spatial coordinate dataset, performing multi-dimensional feature analysis and extraction on the first survey image in the spatial coordinate dataset to generate multi-dimensional feature values; and comparing and calculating the multi-dimensional feature values ​​with a preset feature threshold to generate a comparison. Based on the comparison results, a target evaluation method is adaptively selected from a set of preset computational methods for evaluating image quality to obtain a quantitative quality evaluation value for the first survey image. A second survey image is then selected from the spatial coordinate dataset based on the quantitative quality evaluation value. This second survey image is input into a fusion calculation model, which includes a spatial feature extraction branch and a sequence feature extraction branch. The spatial feature extraction branch is used to parse the geometric and structural information of the spatial coordinate dataset, and / or the sequence feature extraction branch is used to analyze the dynamic correlation information of the spatial coordinate dataset to generate a fusion data result. This fusion data result is used to characterize specific data elements in land planning and surveying.

[0007] According to at least one specific embodiment of the present application, a target area for land planning and mapping is measured based on a spatial data acquisition method. The spatial data acquisition method includes a rangefinder, a GPS area measuring instrument, or a three-dimensional laser scanning unit. The rangefinder is used to obtain linear distance measurement values ​​of the target area. The GPS area measuring instrument is used to obtain the boundary trajectory of the target area based on GPS and calculate the area of ​​the target area. The three-dimensional laser scanning unit is used to obtain three-dimensional point cloud data of the surface of the target area.

[0008] According to at least one specific embodiment of the present application, the remote data processing terminal receives the spatial coordinate dataset, performs spatiotemporal reference unification processing on the spatial coordinate dataset, performs multi-dimensional feature analysis and extraction on the first survey image in the spatial coordinate dataset, and generates multi-dimensional feature values. The process further includes: the remote data processing terminal receives the spatial coordinate dataset, performs spatiotemporal reference unification processing on the spatial data in the spatial coordinate dataset based on a preset national geodetic coordinate system, and generates a standard dataset with a unified spatiotemporal reference system through coordinate transformation and timestamp alignment; performs multi-dimensional feature analysis and extraction on the first survey image in the standard dataset, calculating color histogram statistics, local binary mode histograms, and gradient direction histograms from the dimensions of color space, texture distribution, and edge gradient, respectively, to generate corresponding multi-dimensional feature values; inputs the multi-dimensional feature values ​​into a preset feature evaluation model, which, based on a built-in multilayer perceptron mechanism, calculates the matching degree between the multi-dimensional feature values ​​and a preset feature template through forward propagation, fuses the matching degree corresponding to the multi-dimensional feature values ​​through a normalized weighting function, and outputs a comprehensive feature quantization value of the first survey image.

[0009] According to at least one specific embodiment of the present application, the step of comparing and calculating the multi-dimensional feature values ​​with preset feature thresholds and generating comparison results, and adaptively selecting a target evaluation method from a plurality of preset computational methods for evaluating image quality based on the comparison results to obtain a quantitative quality evaluation value of the first surveyed image, further includes: comparing the extracted multi-dimensional feature values ​​with a preset multi-dimensional feature threshold vector dimension by dimension; obtaining a feature saliency index array by calculating the ratio of the value of each feature dimension to the corresponding threshold; using the out-of-standard identifiers of the feature saliency index array as preliminary comparison results; analyzing the combination patterns of out-of-standard identifiers using a preset decision tree model based on the preliminary comparison results, wherein the decision tree model is trained based on historical application scenario data; calculating the target quality evaluation method index of the current combination pattern based on the decision tree model; calling the corresponding evaluation algorithm program based on the target quality evaluation method index; and performing weighted fusion or transformation operations on the multi-dimensional feature values ​​through the evaluation algorithm program to calculate the comprehensive quantitative quality evaluation value of the first surveyed image.

[0010] According to at least one specific embodiment of the present application, the feature saliency index array is an ordered data set composed of the ratios of multi-dimensional feature values ​​to their corresponding preset thresholds, used to quantify the saliency of each feature dimension; the exceeding identifier is used to mark whether the feature saliency index exceeds the set threshold as a Boolean variable, and the combination mode of the exceeding identifier is specifically a specific logical state sequence composed of multiple exceeding identifiers; the target quality assessment method index is an identifier output by the decision tree model, used to point to the storage location or call number in the preset method library, and the storage location or call number is used to store a specific image quality assessment algorithm adapted to the current feature combination mode; the assessment algorithm program is an executable computer program module that receives multi-dimensional feature values ​​as input, performs calculations through preset weighting, fusion, or mathematical transformation rules, and outputs the corresponding comprehensive quantitative quality evaluation value.

[0011] According to at least one specific embodiment of the present application, the step of selecting a second survey image from the spatial coordinate dataset based on the quantitative quality evaluation value, inputting the second survey image into a fusion calculation model, the fusion calculation model including a spatial feature extraction branch and a sequence feature extraction branch, parsing the geometric and structural information of the spatial coordinate dataset through the spatial feature extraction branch, and / or: analyzing the dynamic correlation information of the spatial coordinate dataset through the sequence feature extraction branch to generate fusion data results, the fusion data results being used to characterize specific data elements of land planning surveying, further including: based on the quantitative quality evaluation value of the first survey image, selecting the second survey image with the highest quantitative quality evaluation value and its associated spatial coordinate dataset from the spatial coordinate dataset. The second survey image and its associated spatial data subset are input into a fusion computing model. The fusion computing model is configured with a spatial feature extraction branch and a sequence feature extraction branch. The spatial feature extraction branch analyzes the coordinates and topological relationships in the spatial data subset through a geometric feature extraction network and outputs a spatial feature vector. The sequence feature extraction branch uses a recurrent neural network structure to obtain the sequence dependencies of pixels or features in the second survey image and outputs a sequence feature vector. The spatial feature vector and the sequence feature vector are input into a fully connected fusion layer of the fusion computing model. The fusion layer performs weighted concatenation and nonlinear transformation operations to generate fused data results. The fused data results are used to characterize the specific data elements of the land planning survey of the target area.

[0012] According to at least one specific embodiment of the present application, in the process of inputting the second survey image into the fusion calculation model, the second survey image and its associated spatial coordinate dataset are input into the fusion calculation model; when filtering the second survey image and its associated spatial data subset, a sorting method and a threshold comparison method are used for filtering.

[0013] According to another aspect of the embodiments of this application, a data analysis system based on land planning and surveying is provided to implement the data analysis method based on land planning and surveying, comprising: a spatial coordinate dataset generation module, which measures the target area of ​​land planning and surveying to obtain a first survey image and spatial data, associates the first survey image and the spatial data and maps them to spatial coordinates to form a spatial coordinate dataset, and uploads the spatial coordinate dataset to a remote data processing terminal in real time; a multi-dimensional feature value generation module, in which the remote data processing terminal receives the spatial coordinate dataset, performs spatiotemporal benchmark unification processing on the spatial coordinate dataset, performs multi-dimensional feature analysis and extraction on the first survey image in the spatial coordinate dataset, and generates multi-dimensional feature values; and a quantitative quality evaluation value calculation module, which calculates the... Multi-dimensional feature values ​​are compared and calculated with preset feature thresholds to generate comparison results. Based on the comparison results, a target evaluation method is adaptively selected from multiple preset calculation methods for evaluating image quality to obtain a quantitative quality evaluation value for the first surveying image. The surveying image fusion calculation module selects a second surveying image from the spatial coordinate dataset based on the quantitative quality evaluation value and inputs the second surveying image into the fusion calculation model. The fusion calculation model includes a spatial feature extraction branch and a sequence feature extraction branch. The spatial feature extraction branch parses the geometric and structural information of the spatial coordinate dataset, and / or the sequence feature extraction branch analyzes the dynamic correlation information of the spatial coordinate dataset to generate fusion data results. The fusion data results are used to characterize specific data elements of land planning and surveying.

[0014] According to another aspect of the embodiments of this application, an electronic device is provided, including: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method.

[0015] According to another aspect of the embodiments of this application, a computer-readable storage medium is provided that stores a computer program executable by an electronic device, which, when run on the electronic device, causes the electronic device to perform the steps of the method.

[0016] The beneficial technical effects of the embodiments of this application are: This application's embodiments, based on land planning and surveying data analysis methods, construct an intelligent processing flow from data acquisition, real-time uploading, intelligent quality assessment to multi-source data fusion. By performing multi-dimensional feature analysis and quantitative quality assessment on surveyed images, and selecting high-quality images based on the assessment results, these images are finally processed collaboratively with associated spatial data in a fusion computing model to generate fused data results representing land planning and surveying elements. The technical effects of this application's embodiments are not only reflected in the efficiency improvement of individual steps, but also in the optimization of the entire land planning and surveying data processing process through logical connections and data fusion between steps, providing reliable technical support for land planning, surveying, and governance. Attached Figure Description

[0017] To more clearly illustrate the specific implementation methods of the embodiments of this application or the technical solutions in the prior art, the drawings used in the description of the specific implementation methods or the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a data analysis method based on land planning and surveying.

[0019] Figure 2 This is a flowchart of the optimized technical solution from steps S11 to S13.

[0020] Figure 3 This is a flowchart of the optimized technical solution from steps S21 to S23.

[0021] Figure 4 This is a flowchart of the optimized technical solution from step S31 to step S33.

[0022] Figure 5 This is a flowchart of the optimized technical solution from steps S41 to S43.

[0023] Figure 6 This is an architecture diagram of a data analysis system based on land planning and surveying.

[0024] Figure 7 This is a structural diagram of an electronic device. Detailed Implementation

[0025] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the embodiments of this application, and not all embodiments. Based on the specific implementation methods in the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of the embodiments of this application.

[0026] The technical solutions disclosed in the embodiments of this application can be implemented in multiple application scenarios. It is easy to understand that the implementation methods of the embodiments of this application in specific application scenarios are not unique. The description of the specific application scenario can only be regarded as an explanation of the embodiments of this application, rather than a limitation.

[0027] This application scenario primarily targets the fields of high-precision surveying and dynamic monitoring in land spatial planning and integrated urban-rural management. With the comprehensive implementation of new urbanization and the land spatial planning system, local natural resources authorities, urban and rural planning units, real estate registration agencies, and engineering surveying companies face an urgent need for large-scale, efficient, and real-time surveying and monitoring of spatial elements such as land, buildings, and infrastructure.

[0028] This application scenario is suitable for the following situations: 1. Rural homestead land ownership confirmation and illegal construction monitoring: In the process of carrying out unified registration of rural real estate, it is necessary to quickly survey and map homesteads that are scattered and have complex terrain, and at the same time identify illegal constructions that exceed the approved area in real time. This application scenario can achieve automated surveying and change recognition with millimeter-level accuracy in complex environments through multi-source data collection and intelligent fusion.

[0029] 2. Supervision of the implementation of territorial spatial planning: During the implementation of the plan, it is necessary to regularly monitor changes in land use, progress of construction projects, and compliance with ecological protection red lines within the planning area. This application scenario can achieve dynamic monitoring with an adjustable cycle of 1-72 hours through real-time data collection and intelligent analysis, and promptly detect illegal construction, illegal land use and other behaviors.

[0030] 3. Full-cycle management of major engineering construction projects: In the construction of major projects such as highways, railways, and water conservancy projects, continuous mapping and monitoring of the project area and its surrounding environment are required to ensure that construction meets planning requirements and controls environmental impact. This application scenario, through multi-source data collaborative acquisition and intelligent fusion processing, can generate comprehensive data containing information such as three-dimensional coordinates, surface morphology, and structural integrity, providing accurate spatial data support for the full-cycle management of projects.

[0031] In these application scenarios, the technical solutions of this application embodiment solve the shortcomings of traditional surveying and mapping methods in terms of real-time performance, automation level, and data processing depth by constructing a fully automated system from data acquisition, transmission, processing to fusion analysis. This can meet the urgent needs of modern land spatial planning and management for high-timeliness, high-precision, and intelligent spatial data services.

[0032] like Figure 1 The data analysis methods based on land planning and surveying shown include: Step S1: Measure the target area of ​​land planning and mapping to obtain a first survey image and spatial data. Associate the first survey image and the spatial data and map them to spatial coordinates to form a spatial coordinate dataset. Upload the spatial coordinate dataset to a remote data processing terminal in real time.

[0033] Step S2: The remote data processing terminal receives the spatial coordinate dataset, performs spatiotemporal reference unification processing on the spatial coordinate dataset, performs multi-dimensional feature analysis and extraction on the first survey image in the spatial coordinate dataset, and generates multi-dimensional feature values.

[0034] Step S3: Compare and calculate the multi-dimensional feature values ​​with the preset feature thresholds and generate comparison results. Based on the comparison results, adaptively select the target evaluation method from a plurality of preset calculation methods for evaluating image quality to obtain the quantitative quality evaluation value of the first mapped image.

[0035] Step S4: Based on the quantitative quality evaluation value, a second survey image is selected from the spatial coordinate dataset. The second survey image is input into the fusion calculation model, which includes a spatial feature extraction branch and a sequence feature extraction branch. The spatial feature extraction branch is used to parse the geometric and structural information of the spatial coordinate dataset, and / or the sequence feature extraction branch is used to analyze the dynamic correlation information of the spatial coordinate dataset to generate fusion data results. The fusion data results are used to characterize the specific data elements of land planning and mapping.

[0036] In the technical solutions provided in steps S1 to S4, the surveyed image is a visualized image of the target area acquired through a surveying device. Association mapping is the process of establishing a correspondence between the surveyed image and its corresponding spatial data and projecting them uniformly onto a specific spatial coordinate system. Spatiotemporal reference unification processing is the process of converting spatial data from different times and using different coordinate systems to a unified national geodetic coordinate system and time reference system. Multi-dimensional feature values ​​refer to a set of numerical features obtained by quantitatively analyzing the surveyed image in multiple dimensions such as color, texture, and edge. Preset feature thresholds refer to pre-defined numerical limits based on different feature dimensions, used to determine whether a feature is significant or meets requirements. Adaptive target evaluation method selection refers to the system automatically selecting the most suitable evaluation method from multiple available algorithms based on the comparison results of multi-dimensional feature values ​​and preset feature thresholds, thereby achieving dynamic optimization of the evaluation strategy.

[0037] In the technical solutions provided in steps S1 to S4, step S1 achieves structured encapsulation and real-time synchronization of multi-source data by instantly associating the first survey image with spatial data and mapping it into a unified spatial coordinate dataset, providing complete real-time input data for subsequent remote processing. Step S2, upon receiving the data from step S1, first performs spatiotemporal benchmark unification processing on the spatial coordinate dataset to ensure all data are in the same reference system. Multi-dimensional feature extraction is then performed on the image within this unified spatiotemporal framework to generate multi-dimensional feature values. Step S3 generates a quantitative quality evaluation value based on an adaptive selection evaluation method using multi-dimensional feature values. The second survey image is then selected based on the quantitative quality evaluation value, filtering out low-quality images to ensure only high-quality images enter the fusion process, improving data effectiveness and reducing the processing burden of invalid data. Step S4 facilitates data fusion and model collaboration. Step S4 inputs the optimized second survey image into a fusion calculation model with two branches. Geometric information is analyzed through the spatial feature branch, and correlation information is mined through the sequence feature branch, ultimately generating fused data results. This allows information from high-quality images to be more fully coupled with spatial data in the fusion model, improving the completeness and usability of the output data.

[0038] In summary, the technical solutions provided in steps S1 to S4 have achieved automation and intelligence in the entire process of land surveying and mapping data processing. By performing multi-dimensional feature analysis and quantitative quality assessment on the measured images, and selecting high-quality images based on the assessment results, the data is finally processed collaboratively in the fusion computing model to generate fused data that represents the requirements of land planning and mapping. Through the collaborative cooperation between different steps, the entire process of land planning and mapping data processing has been optimized, providing reliable technical support for land planning and surveying.

[0039] like Figure 2As shown, preferably, in step S1, the target area for land planning and mapping is measured based on spatial data acquisition methods. These spatial data acquisition methods include a rangefinder, a GPS area measuring instrument, or a three-dimensional laser scanning unit. The rangefinder is used to acquire linear distance measurements of the target area. The GPS area measuring instrument is used to acquire the boundary trajectory of the target area based on GPS and calculate the area of ​​the target area. The three-dimensional laser scanning unit is used to acquire three-dimensional point cloud data of the target area's surface. Further, it includes: Step S11: Determine the mapping range and measurement requirements of the target area, perform distance measurement on the boundary points within the target area based on the rangefinder, obtain the linear distance measurement value, and record the coordinate identifier and measurement timestamp of each measurement point.

[0040] Step S12: Start the GPS area measuring instrument, obtain the boundary trajectory coordinate sequence of the target area by receiving GPS satellite signals, and calculate the area value of the target area based on the boundary trajectory coordinate sequence using an integration algorithm.

[0041] Step S13: Deploy a three-dimensional laser scanning unit within the target area, and scan the surface of the area with a preset point cloud density using a laser emitting and receiving device to obtain three-dimensional point cloud data. Synchronize and spatially register the linear distance measurement value, boundary trajectory coordinate sequence, target area area value, and three-dimensional point cloud data to generate a spatial coordinate dataset.

[0042] The optimized technical solution provided in steps S11 to S13 utilizes three measurement methods—range measurement, GPS trajectory acquisition, and 3D laser scanning—to fuse the measurement data into a unified spatial coordinate dataset through spatiotemporal registration. Specifically, the rangefinder acquires the linear values ​​of local boundary points, the GPS area measuring instrument acquires the trajectory coordinates of the regional boundaries, and the laser scan acquires the 3D point cloud of the surface, forming a complete spatial information coverage from points to lines to surfaces. These multiple measurement methods complement each other in terms of spatial scale and data type, achieving collaborative acquisition of multi-scale, multi-dimensional spatial data. Each step records coordinate identifiers and timestamps during data acquisition, uniformly performing time synchronization and spatial registration to ensure that all measurement data are based on the same spatiotemporal reference system. From linear measurement to trajectory calculation to 3D scanning, all data is ultimately registered and generated into a spatial coordinate dataset. Steps S11 to S13 enable the integration of data generated by different measurement methods in a structured form, providing a unified and complete data input foundation for subsequent image association mapping, quality assessment, and data fusion.

[0043] like Figure 3As shown, preferably, in step S2, the remote data processing terminal receives the spatial coordinate dataset, performs spatiotemporal reference unification processing on the spatial coordinate dataset, performs multi-dimensional feature analysis and extraction on the first survey image in the spatial coordinate dataset, and generates multi-dimensional feature values, further including: Step S21: The remote data processing terminal receives the spatial coordinate dataset and, based on the preset national geodetic coordinate system, performs spatiotemporal reference unification processing on the spatial data in the spatial coordinate dataset through coordinate transformation and timestamp alignment to generate a standard dataset with a unified spatiotemporal reference system.

[0044] Step S22: Perform multi-dimensional feature analysis and extraction on the first mapping image in the standard dataset, calculate the color histogram statistics, local binary mode histogram and gradient direction histogram from the dimensions of color space, texture distribution and edge gradient respectively, and generate the corresponding multi-dimensional feature values.

[0045] Step S23: Input the multi-dimensional feature values ​​into a preset feature evaluation model. The feature evaluation model is based on a built-in multi-layer perception mechanism. It calculates the matching degree between the multi-dimensional feature values ​​and the preset feature template through forward propagation. It fuses the matching degree corresponding to the multi-dimensional feature values ​​through a normalized weighting function and outputs the comprehensive feature quantization value of the first mapping image.

[0046] The optimization techniques provided in steps S21 to S23 unify the received spatial data into the national geodetic coordinate system, then perform multi-dimensional feature extraction on the surveyed image, and finally output the comprehensive feature quantization value of the image, wherein: Step S21 transforms the spatial data to a unified national geodetic coordinate system, establishing an authoritative spatial reference framework for subsequent image feature analysis and ensuring that all image feature analysis is performed within the same coordinate system. Step S22 extracts features from the image's color, texture, and edge gradient dimensions, generating corresponding multi-dimensional feature values. Step S23 inputs the multi-dimensional feature values ​​output from Step S22 into a feature evaluation model based on a multilayer perceptron, utilizing a forward propagation mechanism to achieve deep fusion of features across various dimensions. This transforms feature analysis from single-dimensional statistics to multi-dimensional intelligent evaluation. A normalized weighting function is used to fuse the matching degrees of features across different dimensions, ultimately outputting the corresponding comprehensive feature quantification value. This provides a standardized quantitative basis for subsequent image quality assessment and selection decisions.

[0047] Steps S21 to S23 utilize various technical means such as benchmark unification, feature multidimensionalization, and intelligent evaluation to construct a complete spatial data processing and feature quantification system. Through logical connections and functional coordination between each step, the transformation from raw spatial data to structured feature values ​​is realized, laying the foundation for the intelligent analysis and application of land surveying data.

[0048] like Figure 4 As shown, preferably, in step S3, the step of comparing and calculating the multi-dimensional feature values ​​with preset feature thresholds and generating comparison results, and adaptively selecting a target evaluation method from a preset set of multiple computational methods for evaluating image quality based on the comparison results to obtain a quantitative quality evaluation value for the first mapped image, further includes: Step S31: The extracted multi-dimensional feature values ​​are compared with the pre-set multi-dimensional feature threshold vector dimension by dimension. By calculating the ratio of the value of each feature dimension to the corresponding threshold, a feature significance index array is obtained. The over-standard identifier of the feature significance index array is used as the preliminary comparison result.

[0049] Step S32: Based on the preliminary comparison results, analyze the combination patterns of out-of-standard identifiers using a preset decision tree model. The decision tree model is trained based on historical application scenario data. Calculate the target quality assessment method index for the current combination pattern based on the decision tree model.

[0050] Step S33: Based on the target quality assessment method index, call the corresponding assessment algorithm program, and perform weighted fusion or transformation operations on the multi-dimensional feature values ​​through the assessment algorithm program to calculate the comprehensive quantitative quality evaluation value of the first survey image.

[0051] For example, in the optimization technical solution provided in steps S31 to S33, the feature saliency index array is an ordered data set composed of the ratios of multi-dimensional feature values ​​to their corresponding preset thresholds, used to quantify the saliency of each feature dimension.

[0052] As an alternative technical solution, this application also provides a technical solution that further optimizes step S3: Step S34: Based on the multi-dimensional feature values ​​and the corresponding preset thresholds, obtain the feature significance index array by calculating the ratio of each dimension, generate the over-standard identifiers corresponding to each dimension, and form the current feature combination pattern based on the over-standard identifier sequence.

[0053] Step S35: Input the feature combination pattern into the pre-trained decision tree model, analyze the logical structure of the pattern through the model, and map out the target quality assessment method index that matches it. The index points to the calling path of a specific assessment algorithm in the preset method library.

[0054] Step S36: Load the corresponding evaluation algorithm program from the method library based on the index, input the original multi-dimensional feature values ​​into the evaluation algorithm program, and calculate the comprehensive quantitative quality evaluation value of the image through the weighted fusion rules or mathematical transformation functions built into the program.

[0055] The out-of-target identifier is a Boolean variable used to mark whether the feature significance index exceeds a set threshold. The combination pattern of the out-of-target identifiers is specifically a specific logical state sequence composed of multiple out-of-target identifiers. Taking the combination pattern formed by the out-of-target identifiers as input, the corresponding target quality assessment method index is output through a decision tree model, establishing a mapping relationship between the specific logical state sequence and the optimal assessment algorithm. This allows for adaptive selection of the most suitable quality assessment strategy based on the actual performance of the image features.

[0056] The target quality assessment method index is an identifier output by the decision tree model, used to point to a storage location or call number in a preset method library. This storage location or call number stores a specific image quality assessment algorithm adapted to the current feature combination pattern. The target quality assessment method index directly points to a specific algorithm program in the preset method library. Through index-based invocation, the evaluation algorithm is accurately located and executed, achieving a connection between the decision result and the executed program. This ensures rapid response to feature changes and the execution of corresponding weighted, fused, or transformed operations.

[0057] The evaluation algorithm program is an executable computer program module that receives multi-dimensional feature values ​​as input, performs calculations according to preset weighting, fusion, or mathematical transformation rules, and outputs a corresponding comprehensive quantitative quality evaluation value. The invoked evaluation algorithm program receives the original multi-dimensional feature values ​​as input, performs calculations according to preset rules, and finally outputs a comprehensive quantitative quality evaluation value, realizing a closed-loop calculation from feature analysis to quality evaluation, providing a reliable basis for subsequent image screening.

[0058] like Figure 5 As shown, preferably, in step S4, the second survey image is selected from the spatial coordinate dataset based on the quantitative quality evaluation value, and the second survey image is input into the fusion calculation model. The fusion calculation model includes a spatial feature extraction branch and a sequence feature extraction branch. The spatial feature extraction branch is used to parse the geometric and structural information of the spatial coordinate dataset, and / or the sequence feature extraction branch is used to analyze the dynamic correlation information of the spatial coordinate dataset to generate fusion data results. The fusion data results are used to characterize the specific data elements of land planning and mapping, and further include: Step S41: Based on the quantitative quality evaluation value of the first survey image, select the second survey image with the highest quantitative quality evaluation value and its associated spatial data subset from the spatial coordinate dataset.

[0059] Step S42: Input the second survey image and its associated spatial data subset into the fusion computing model. The fusion computing model is deployed with a spatial feature extraction branch and a sequence feature extraction branch. The spatial feature extraction branch parses the coordinates and topological relationships in the spatial data subset through a geometric feature extraction network and outputs a spatial feature vector. The sequence feature extraction branch uses a recurrent neural network structure to obtain the sequence dependency of pixels or features in the second survey image and outputs a sequence feature vector.

[0060] For example, in the process of inputting the second survey image into the fusion computing model, the second survey image and its associated spatial coordinate dataset are input into the fusion computing model; when filtering the second survey image and its associated spatial data subset, a sorting method and a threshold comparison method are used for filtering.

[0061] Step S43: Input the spatial feature vector and the sequence feature vector into the fully connected fusion layer of the fusion computing model. The fusion layer performs weighted concatenation and nonlinear transformation operations to generate fused data results. The fused data results are used to characterize the specific data elements of land planning and mapping in the target area.

[0062] The optimization solutions provided in steps S41 to S43 are based on intelligent fusion processing of multi-source data with optimized image quality. By performing dual-branch feature extraction and fusion on the highest-quality image and its corresponding spatial data, structured result data that can characterize land planning elements is generated, wherein: Step S41 sorts and filters images based on quantitative quality evaluation values, extracting the second highest-quality surveying image and its corresponding spatial data subset from multiple images to ensure that both the images and spatial data entering the fusion stage possess high quality. Step S42 processes geometric topological information through a spatial feature extraction branch and analyzes image sequence features through a sequence feature extraction branch. These two branches specialize in processing the structured characteristics of spatial data and the continuous characteristics of image data, respectively, forming an efficient functional synergy between heterogeneous data and a dedicated model, achieving parallel deep extraction of multi-source information. In step S43, the feature vectors output from the two branches undergo weighted concatenation and nonlinear transformation in a fully connected fusion layer, realizing the structural integration of multi-dimensional features from parallel extraction to unified synthesis. This generates fused data results that comprehensively reflect the surveying elements of the target area, providing a structured and interpretable data foundation for subsequent planning decisions.

[0063] The following example, step S43, illustrates the transformation from raw multi-source data to high-quality planning data: Step S43 constructs and executes a multimodal feature fusion and nonlinear structured transformation mechanism. Through the fully connected fusion layer of the model, it performs vector concatenation based on preset weights on spatial feature vectors (geometric topological information) and sequence feature vectors (image sequence features), achieving a linear combination of heterogeneous features in a unified vector space. Based on the vector concatenation, activation functions (such as ReLU and Sigmoid) are then used to perform nonlinear transformations on the combined features, mapping the multidimensional features into structured data representing land planning elements. During this process, the weight parameters of the fusion layer are adaptively optimized through model training, enabling the feature fusion method to be dynamically adjusted according to the specific surveying and mapping task requirements.

[0064] Compared with existing data fusion methods, step S43 improves simple overlay into intelligent fusion. Through a combination of weighted stitching and nonlinear transformation, feature-level deep fusion is achieved, which can uncover the implicit correlation between spatial data and image data. Furthermore, through a trainable fully connected fusion layer, it can learn the optimal fusion strategy from historical data, significantly improving its adaptability to complex surveying and mapping scenarios. The final fused data result has multi-level, analyzable structured characteristics, and can simultaneously contain structured data such as geometric coordinates, topological relationships, and feature sequences, providing richer information dimensions for subsequent planning and analysis.

[0065] Preferably, the spatial feature extraction branch and the sequence feature extraction branch involved in step S43 can be implemented by the following methods: Spatial feature extraction branch: The core architecture adopts a 3D point cloud feature extraction network based on PointNet++: 3D point cloud data from the spatial coordinate dataset is input into the network, and local point cloud regions are constructed through multi-level sampling and grouping operations.

[0066] The geometric features (such as curvature, normal vector, density distribution, etc.) of each local region are extracted using a multilayer perceptron (MLP). Max pooling is then used to aggregate these local features into a global feature representation. Finally, a fully connected layer maps the global features into a multidimensional spatial feature vector. The spatial feature extraction branch in this step is suitable for extracting structured spatial information such as building outlines, terrain undulations, and boundary topology.

[0067] The sequence feature extraction branch adopts a sequence modeling architecture based on bidirectional LSTM (Bi-LSTM), which converts the second survey image into a grayscale sequence and extracts pixel columns along the horizontal direction of the image with a fixed step size to form sequence data with a width of W (sequence length L = image height). Each pixel column is used as the input for the time step. The spatial dependencies between pixel columns are learned through a bidirectional LSTM layer. The hidden states of the forward and backward LSTMs are then concatenated. The sequence features at different positions are weighted and fused through an attention mechanism. Finally, a 128-dimensional sequence feature vector is output through a fully connected layer.

[0068] The sequence feature extraction branch can effectively capture sequential information such as building facade texture, vegetation distribution continuity, and road linear features in images.

[0069] Implementation of dual-branch collaborative work: The two branches are processed in parallel: the spatial feature extraction branch performs analysis based on geometric and topological features, and the sequence feature extraction branch is based on the analysis of spatial continuity and pattern association. The following collaborative strategy is adopted during implementation: Input data alignment: Ensure that point cloud data and image data are strictly registered in a spatial reference frame.

[0070] Feature dimension unification: Unify the output vectors of the two branches to the same dimension.

[0071] Training synchronization optimization: The two branches are trained simultaneously using a joint loss function (such as spatial reconstruction loss + sequence prediction loss) so that the extracted features retain their individual characteristics while also having fusion compatibility.

[0072] When deployed on edge computing devices (such as the NVIDIA Jetson Nano), the spatial feature extraction branch uses a TensorRT-accelerated PointNet++ model, while the sequence feature extraction branch uses an ONNX-formatted Bi-LSTM model. Both branches are implemented through an asynchronous parallel computing pipeline, with computation time controlled within 300ms and memory usage not exceeding 1GB, meeting the requirements of real-time field operations. The two branches can efficiently process different types of spatial information, while a collaborative mechanism ensures the effectiveness of feature fusion, providing high-quality feature input for subsequent fully connected fusion layers.

[0073] For the method steps disclosed in the above embodiments, the method steps are described as a series of actions for the purpose of simplicity. However, those skilled in the art should understand that the embodiments of this application are not limited to the described order of actions, because according to the embodiments of this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily necessary for the embodiments of this application.

[0074] Any flowchart or other description of a process or method can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process. Furthermore, the scope of preferred embodiments of the invention includes additional implementations in which functions may be performed and implemented not in the order shown or discussed, including substantially simultaneously or in reverse order according to the functions involved, or in accordance with program structures such as loops, branches, etc., as will be readily understood by those skilled in the art when implementing the embodiments of this application.

[0075] like Figure 6 The data analysis system based on land planning and surveying shown is used to implement the data analysis method based on land planning and surveying described in any specific embodiment of the present application, including: The spatial coordinate dataset generation module measures the target area of ​​land planning and mapping to obtain a first survey image and spatial data. It associates the first survey image and the spatial data and maps them to spatial coordinates to form a spatial coordinate dataset. The spatial coordinate dataset is then uploaded to a remote data processing terminal in real time.

[0076] The multi-dimensional feature value generation module receives the spatial coordinate dataset from the remote data processing terminal, performs spatiotemporal benchmark unification processing on the spatial coordinate dataset, performs multi-dimensional feature analysis and extraction on the first survey image in the spatial coordinate dataset, and generates multi-dimensional feature values.

[0077] The quantitative quality evaluation value calculation module compares and calculates the multi-dimensional feature values ​​with preset feature thresholds and generates comparison results. Based on the comparison results, it adaptively selects a target evaluation method from a number of preset calculation methods for evaluating image quality to obtain the quantitative quality evaluation value of the first mapped image.

[0078] The mapping image fusion calculation module selects a second mapping image from the spatial coordinate dataset based on the quantitative quality evaluation value, and inputs the second mapping image into the fusion calculation model. The fusion calculation model includes a spatial feature extraction branch and a sequence feature extraction branch. The spatial feature extraction branch parses the geometric and structural information of the spatial coordinate dataset, and / or the sequence feature extraction branch analyzes the dynamic correlation information of the spatial coordinate dataset to generate fusion data results. The fusion data results are used to characterize the specific data elements of land planning and mapping.

[0079] The implementation methods of the system described above are merely illustrative. For example, the various functional modules, units, or subsystems within the system may or may not be physically separate, or they may or may not be physical units; that is, they may be located in the same place or distributed across multiple different systems and their subsystems or modules. Those skilled in the art can select some or all of the functional modules, units, or subsystems to achieve the objectives of the embodiments of the present invention according to actual needs. Those skilled in the art can understand and implement the above-described situations without any creative effort.

[0080] like Figure 7 As shown, this application embodiment, in addition to providing a data analysis method and system based on land planning and surveying, also provides corresponding electronic devices and storage media: An electronic device includes: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus; the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method.

[0081] A computer-readable storage medium storing a computer program executable by an electronic device, which, when run on the electronic device, causes the electronic device to perform the steps of the method.

[0082] Figure 7 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Figure 7 In this device 800, a processor 801, a memory 802, a communication interface 803, and a bus 804 are included. The processor 801, memory 802, and communication interface 803 communicate via the bus 804, or via other means such as wireless transmission. The memory 802 stores instructions, and the processor 801 executes the instructions stored in the memory 802. The memory 802 stores program code 8021, and the processor 801 can call the program code 8021 stored in the memory 802 to execute the steps of the gas detection data analysis method.

[0083] It should be understood that in the embodiments of this application, processor 801 may be a CPU, or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. General-purpose processors may be microprocessors or any conventional processors, etc.

[0084] The memory 802 may include read-only memory (ROM) and random access memory (RAM), and provides instructions and data to the processor 801. The memory 802 may also include non-volatile random access memory. The memory 802 may be volatile memory or non-volatile memory, or may include both. The non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory may be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM).

[0085] In addition to the data bus, bus 804 may also include a power bus, a control bus, and a status signal bus. However, for clarity, all buses are labeled as bus 804 in the diagram.

[0086] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive (SSD).

[0087] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of the specification of the embodiments of this application.

[0088] In the description of the embodiments of this application, the reference to terms such as "an embodiment," "example," "specific example," etc., means that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0089] All features disclosed in the embodiments of this application, or all steps in the disclosed methods or processes, may be combined in any way, except for mutually exclusive features and / or steps. Any feature disclosed in the specification of the embodiments of this application, unless specifically stated otherwise, may be replaced by other equivalent or similar alternative features. That is, unless specifically stated otherwise, each feature is merely one example of a series of equivalent or similar features. Throughout the specification, the same reference numerals indicate the same elements.

[0090] Those skilled in the art will understand that modules in the device of the embodiments can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components. Except where at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all features disclosed in this specification of embodiments (including the corresponding claims, abstract, and drawings) and all processes or units of any method or device so disclosed. Unless expressly stated otherwise, each feature disclosed in this specification of embodiments (including the corresponding claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0091] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of this application, and are not intended to limit them. Although the embodiments of this application have been described in detail with reference to the foregoing specific embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein, and such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the specific embodiments of this application.

Claims

1. A data analysis method based on land planning and surveying, characterized in that, include: The target area of ​​land planning and mapping is measured to obtain a first survey image and spatial data. The first survey image and the spatial data are associated and mapped to spatial coordinates to form a spatial coordinate dataset. The spatial coordinate dataset is uploaded to a remote data processing terminal in real time. The remote data processing terminal receives the spatial coordinate dataset, performs spatiotemporal reference unification processing on the spatial coordinate dataset, performs multi-dimensional feature analysis and extraction on the first survey image in the spatial coordinate dataset, and generates multi-dimensional feature values. The multi-dimensional feature values ​​are compared and calculated with preset feature thresholds to generate comparison results. Based on the comparison results, a target evaluation method is adaptively selected from a number of preset calculation methods for evaluating image quality to obtain the quantitative quality evaluation value of the first mapped image. Based on the quantitative quality evaluation value, a second survey image is selected from the spatial coordinate dataset. The second survey image is then input into a fusion calculation model, which includes a spatial feature extraction branch and a sequence feature extraction branch. The spatial feature extraction branch is used to analyze the geometric and structural information of the spatial coordinate dataset, and / or the sequence feature extraction branch is used to analyze the dynamic correlation information of the spatial coordinate dataset to generate fusion data results. The fusion data results are used to characterize the specific data elements of land planning and mapping.

2. The data analysis method based on land planning and surveying according to claim 1, characterized in that, The target area for land planning and mapping is measured using spatial data acquisition methods, including a rangefinder, a GPS area measuring instrument, or a three-dimensional laser scanning unit. The rangefinder is used to obtain linear distance measurements of the target area, the GPS area measuring instrument is used to obtain the boundary trajectory of the target area based on GPS and calculate the area of ​​the target area, and the three-dimensional laser scanning unit is used to obtain three-dimensional point cloud data of the surface of the target area.

3. The data analysis method based on land planning and surveying according to claim 1, characterized in that, The remote data processing terminal receives the spatial coordinate dataset, performs spatiotemporal reference unification processing on the spatial coordinate dataset, performs multi-dimensional feature analysis and extraction on the first survey image in the spatial coordinate dataset, and generates multi-dimensional feature values, further including: The remote data processing terminal receives the spatial coordinate dataset and, based on the preset national geodetic coordinate system, performs spatiotemporal benchmark unification processing on the spatial data in the spatial coordinate dataset through coordinate transformation and timestamp alignment to generate a standard dataset with a unified spatiotemporal reference system. Multi-dimensional feature analysis and extraction are performed on the first mapping image in the standard dataset. The color histogram statistics, local binary mode histogram and gradient direction histogram are calculated from the dimensions of color space, texture distribution and edge gradient, respectively, and the corresponding multi-dimensional feature values ​​are generated. The multi-dimensional feature values ​​are input into a preset feature evaluation model. The feature evaluation model is based on a built-in multi-layer perception mechanism. It calculates the matching degree between the multi-dimensional feature values ​​and the preset feature template through forward propagation. It then fuses the matching degree corresponding to the multi-dimensional feature values ​​through a normalized weighting function and outputs the comprehensive feature quantization value of the first mapping image.

4. The data analysis method based on land planning and surveying according to claim 1, characterized in that, The step of comparing and calculating the multi-dimensional feature values ​​with preset feature thresholds and generating comparison results, and adaptively selecting a target evaluation method from a plurality of preset computational methods for evaluating image quality based on the comparison results to obtain a quantitative quality evaluation value of the first mapped image, further includes: The extracted multi-dimensional feature values ​​are compared with the pre-set multi-dimensional feature threshold vector dimension by dimension. By calculating the ratio of the value of each feature dimension to the corresponding threshold, a feature significance index array is obtained. The over-standard identifier of the feature significance index array is used as the preliminary comparison result. Based on the preliminary comparison results, the combination patterns of the out-of-standard identifiers are analyzed using a preset decision tree model. The decision tree model is trained based on historical application scenario data. The target quality assessment method index of the current combination pattern is calculated based on the decision tree model. Based on the target quality assessment method index, the corresponding assessment algorithm program is called, and the multi-dimensional feature values ​​are weighted, fused, or transformed by the assessment algorithm program to calculate the comprehensive quantitative quality evaluation value of the first survey image.

5. The data analysis method based on land planning and surveying according to claim 4, characterized in that, The feature significance index array is an ordered data set composed of the ratios of multi-dimensional feature values ​​to their corresponding preset thresholds, used to quantify the significance of each feature dimension. The over-standard identifier is used to mark whether the feature significance index exceeds a set threshold as a Boolean variable. The combination mode of the over-standard identifier is a specific logical state sequence composed of multiple over-standard identifiers. The target quality assessment method index is an identifier output by the decision tree model, used to point to the storage location or call number in the preset method library. The storage location or call number is used to store a specific image quality assessment algorithm that is adapted to the current feature combination mode. The evaluation algorithm is an executable computer program module that receives multi-dimensional feature values ​​as input, performs calculations through preset weighting, fusion, or mathematical transformation rules, and outputs the corresponding comprehensive quantitative quality evaluation value.

6. The data analysis method based on land planning and surveying according to claim 1, characterized in that, The second survey image is selected from the spatial coordinate dataset based on the quantitative quality evaluation value. This second survey image is then input into a fusion calculation model, which includes a spatial feature extraction branch and a sequence feature extraction branch. The spatial feature extraction branch analyzes the geometric and structural information of the spatial coordinate dataset, and / or the sequence feature extraction branch analyzes the dynamic correlation information of the spatial coordinate dataset to generate a fusion data result. This fusion data result is used to characterize specific data elements of land planning surveying, and further includes: Based on the quantitative quality evaluation value of the first survey image, the second survey image with the highest quantitative quality evaluation value and its associated spatial data subset are selected from the spatial coordinate dataset; The second survey image and its associated spatial data subset are input into the fusion computing model, which is deployed with a spatial feature extraction branch and a sequence feature extraction branch. The spatial feature extraction branch parses the coordinates and topological relationships in the spatial data subset through a geometric feature extraction network and outputs a spatial feature vector. The sequence feature extraction branch uses a recurrent neural network structure to obtain the sequence dependencies of pixels or features in the second survey image and outputs a sequence feature vector. The spatial feature vector and the sequence feature vector are input into the fully connected fusion layer of the fusion computing model. The fusion layer performs weighted concatenation and nonlinear transformation operations to generate fused data results. The fused data results are used to characterize the specific data elements of land planning and mapping in the target area.

7. The data analysis method based on land planning and surveying according to claim 6, characterized in that, During the process of inputting the second survey image into the fusion calculation model, the second survey image and its associated spatial coordinate dataset are input into the fusion calculation model; when filtering the second survey image and its associated spatial data subset, sorting method and threshold comparison method are used for filtering.

8. A data analysis system based on land planning and surveying, used to implement the data analysis method based on land planning and surveying as described in any one of claims 1 to 7, characterized in that, include: The spatial coordinate dataset generation module measures the target area of ​​land planning and mapping to obtain a first survey image and spatial data. It associates the first survey image and the spatial data and maps them to spatial coordinates to form a spatial coordinate dataset. The spatial coordinate dataset is then uploaded to a remote data processing terminal in real time. The multi-dimensional feature value generation module receives the spatial coordinate dataset from the remote data processing terminal, performs spatiotemporal reference unification processing on the spatial coordinate dataset, performs multi-dimensional feature analysis and extraction on the first survey image in the spatial coordinate dataset, and generates multi-dimensional feature values. The quantitative quality evaluation value calculation module compares and calculates the multi-dimensional feature values ​​with preset feature thresholds and generates comparison results. Based on the comparison results, it adaptively selects a target evaluation method from a number of preset calculation methods for evaluating image quality to obtain the quantitative quality evaluation value of the first mapped image. The mapping image fusion calculation module selects a second mapping image from the spatial coordinate dataset based on the quantitative quality evaluation value, and inputs the second mapping image into the fusion calculation model. The fusion calculation model includes a spatial feature extraction branch and a sequence feature extraction branch. The spatial feature extraction branch parses the geometric and structural information of the spatial coordinate dataset, and / or the sequence feature extraction branch analyzes the dynamic correlation information of the spatial coordinate dataset to generate fusion data results. The fusion data results are used to characterize the specific data elements of land planning and mapping.

9. An electronic device, characterized in that, include: The system includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus; the memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores a computer program executable by an electronic device, which, when run on the electronic device, causes the electronic device to perform the steps of the method according to any one of claims 1 to 7.

Images