GIS-based urban land planning analysis method and system

Abstract

The invention relates to the technical field of geographic information systems, in particular to a GIS-based urban land planning analysis method and system, and aims to construct a multi-scale space-time cube model by integrating remote sensing, historical planning, mobile terminals and social media data, generate standardized urban temporary land data and improve the urban land planning analysis efficiency. Adaptive feature extraction is adopted to identify temporary land, a hierarchical classification system is established, abnormal changes are monitored through a time sequence model, an association diagram of the temporary land and an urban function network is constructed, an interaction relation is analyzed, finally, differentiated urban planning optimization suggestions are proposed, efficient configuration and management of urban land resources are assisted, and through fusion of multi-source data, the urban planning optimization suggestions are optimized. Multi-dimensional cognition of space, time, functions and relations is achieved, and the understanding depth of temporary land is fundamentally improved.

Description

Technical Field

[0001] The present invention relates to the technical field of urban land planning, and particularly to a method and system for urban land planning analysis based on GIS. Background Art

[0002] GIS (Geographic Information System) data is the basis for the operation of GIS technology. It contains geospatial data and attribute data, and is used to store, analyze, and display geographical information. GIS technology can collect and integrate geographical information such as landforms and land types, provide basic data for land use planning, analyze the spatial relationships between plots of different uses, and simulate and analyze land planning schemes by establishing models. GIS technology can quickly and accurately analyze and comprehensively evaluate the spatial distribution and characteristics of land, providing a scientific basis and decision-making support for land planning.

[0003] The land use pattern is considered an important factor affecting the balance between residents' work and residence, which in turn affects the travel of urban residents, and even the carbon emissions of urban residents and the urban environment. In related technologies, by using software such as ArcGIS to analyze the urban land use vector data of different periods, the overall characteristics of the land, the regional differentiation law, and the changes in the main land use types can be determined.

[0004] However, traditional urban planning often has a certain difference from the actual urban spatial status quo. Traditional land survey data is usually for permanent construction land, lacking the investigation and analysis of temporary construction land, and it is difficult to better reflect the most real urban spatial status quo and act on urban land planning and analysis. The Chinese invention patent with the publication number CN118761560B and the name of a method and system for urban land planning analysis based on GIS includes: determining the temporary construction land blocks of the target city and the temporary construction land types of the temporary construction land blocks; determining the geographical space change degree of the temporary construction land blocks according to the first historical geographical space data and the second historical geographical space data of the temporary construction land blocks; determining whether the temporary construction land blocks are abnormal land blocks according to the historical remote sensing images and the current remote sensing images of the target city at the second historical time node; and performing correlation analysis on the temporary construction land blocks and the land blocks corresponding to the land types with the same temporary construction land types. By analyzing the current remote sensing images of the target city, the temporary construction land blocks of the target city and the corresponding temporary construction land types can be determined, and the attribute information of the temporary construction land blocks marked as abnormal land blocks can be used as an analysis reference for the same land types. In the process of rapid urbanization, as an important part of urban space, the planning and management of temporary construction land often face the following challenges:

[0005] 1. Single data source: The existing technology overly relies on a single type of remote sensing image data, which is easily affected by weather conditions and imaging quality, resulting in insufficient accuracy in the identification of temporary buildings.

[0006] 2. Lack of timeliness: There is a lack of a continuous monitoring mechanism for the land changes of temporary buildings, making it difficult to detect cases of illegal use or overuse in a timely manner.

[0007] 3. Limited analysis dimensions: Traditional methods analyze temporary buildings as isolated entities, ignoring their complex interaction relationships with the urban functional network.

[0008] 4. Insufficient change prediction ability: The existing technology mainly focuses on the changes that have occurred and lacks the ability to predict the future evolution trend of the land of temporary buildings.

[0009] 5. Lack of decision-making support: There is a lack of an effective mechanism to convert the analysis results into actual planning actions, and planning suggestions often remain in form.

[0010] Therefore, there is an urgent need for an urban land planning analysis method and system that can comprehensively utilize multi-source data, deeply analyze the change laws of the land of temporary buildings, reveal their relationships with urban functions, and provide effective decision-making support. Summary of the Invention

[0011] In view of the above deficiencies of the existing technology, the purpose of the present invention is to provide a GIS-based urban land planning analysis method and system. Through multi-source heterogeneous data fusion, adaptive feature extraction, temporal change analysis, functional association mining, and generation of planning optimization suggestions, it realizes a comprehensive analysis and intelligent planning of urban temporary land, and improves the management efficiency and planning scientificity of urban land resources.

[0012] The technical solution of the present invention is: A GIS-based urban land planning analysis method, including: acquiring multi-source heterogeneous spatio-temporal data, including remote sensing image data, historical planning data, data collected by mobile terminals, and social media geotagged data; based on the acquired multi-source heterogeneous spatio-temporal data, constructing a unified multi-scale spatio-temporal cube model, and generating standardized multi-modal urban temporary land characterization data; according to the standardized multi-modal urban temporary land characterization data, using an adaptive feature extraction mechanism to identify urban temporary land areas, and constructing a hierarchical classification system for temporary land; based on the hierarchical classification system of temporary land and historical time series data, establishing a change-sensitive time series model to detect and identify temporary land areas with abnormal change patterns; according to the identified temporary land areas with abnormal change patterns, constructing an association graph model between temporary land and the urban functional network, and analyzing the interaction relationship between abnormal temporary land and the urban functional system; based on the interaction relationship and multi-objective evaluation indicators, generating differentiated urban planning optimization suggestions to support the efficient allocation and management of urban land resources.

[0013] Preferably, the obtaining of multi-source heterogeneous spatio-temporal data includes:

[0014] Obtaining optical and radar remote sensing image data of the target urban area from a remote sensing satellite platform;

[0015] Retrieving historical planning data and land use information of the target urban area from the urban planning database;

[0016] Collecting ground real-scene data and location information of the target urban area through mobile terminal devices;

[0017] Extracting geotagged data and user activity information related to the target urban area from social media platforms;

[0018] Conducting quality assessment on the obtained various types of data, constructing a data reliability scoring system based on information entropy and texture features, and determining the weight coefficients of each data source.

[0019] Preferably, the constructing of the unified multi-scale spatio-temporal cube model includes:

[0020] Performing spatio-temporal alignment processing on the multi-source heterogeneous data based on the feature point matching algorithm to ensure the consistency of data from different sources in the time and space dimensions;

[0021] Constructing a four-dimensional spatio-temporal data structure and integrating the spatial coordinates (x, y, z) and the time dimension (t) into a unified data representation model;

[0022] Performing scale transformation on spatial data with different resolutions to achieve unified management of multi-scale spatial information;

[0023] Performing interpolation processing on data with discontinuous time series to establish continuous and equally spaced time series data;

[0024] Performing semantic-level fusion standardization on multi-source data to ensure the consistency of data from different sources at the semantic level.

[0025] Preferably, the adopting of the adaptive feature extraction mechanism to identify the urban temporary use land area includes:

[0026] Constructing a multi-scale feature extraction network to perform hierarchical feature extraction on urban area images;

[0027] Fusing spectral features, texture features, morphological features and context features to construct a multi-dimensional feature representation of temporary land use;

[0028] Optimizing the feature weight distribution based on the attention mechanism to highlight the key distinguishing features of temporary land use;

[0029] Implementing precise boundary extraction of the temporary use land area through the context-aware segmentation algorithm.

[0030] Perform attribute annotation on the extracted temporary land areas, including information such as area, shape, material, and use.

[0031] Preferably, the construction of a hierarchical classification system for temporary land includes:

[0032] Establish a three - level classification framework, including a category layer (temporary building type), a function layer (usage function), and a use layer (specific use);

[0033] For the category layer, classify temporary land into temporary building land, temporary facility land, temporary activity land, and temporary storage land;

[0034] For the function layer, classify temporary land into production function, living function, service function, and public function;

[0035] For the use layer, subdivide it into emergency disaster relief, engineering construction, commercial exhibition, cultural activities, etc. according to specific application scenarios;

[0036] Based on the weak - supervised learning method, train a classification model with a small number of labeled samples to achieve automatic classification of large - scale temporary land.

[0037] Preferably, the establishment of a change - sensitive time - series model includes:

[0038] Construct time - window parameters according to the planning period and expected survival time of temporary land;

[0039] Design a change - sensitive loss function for the characteristics of temporary land to improve the detection sensitivity to changes in small targets and sparsely distributed temporary buildings;

[0040] Construct a bidirectional time - series encoder to encode and analyze the historical change sequence of temporary land;

[0041] Introduce an urban planning rule base to constrain anomaly determination with domain knowledge and reduce the misjudgment rate;

[0042] Adopt a multi - evidence fusion method based on evidence theory to quantify the reliability of anomaly determination and generate a confidence score.

[0043] Preferably, the detection and identification of abnormal change patterns in temporary land areas include:

[0044] Fit the time - series change curve of the temporary land area to detect abnormal points deviating from the expected change pattern;

[0045] Calculate the anomaly score based on the change rate and change amplitude, and mark the temporary land areas exceeding the preset threshold as potentially abnormal;

[0046] Verify the authenticity of potential abnormal areas by combining the urban planning cycle and land use conversion rules;

[0047] Classify the identified abnormal temporary land use areas, distinguishing abnormal types such as those not demolished on schedule, those with functional transformation, and those with structural strengthening;

[0048] Generate a heat map of the spatial distribution of abnormal temporary land use to visually display the agglomeration of abnormal areas.

[0049] Preferably, the construction of the association graph model between temporary land use and the urban function network includes:

[0050] Construct a heterogeneous spatial relationship graph based on the spatial relationship between temporary land use and surrounding permanent buildings;

[0051] Calculate the association strength between temporary land use and surrounding urban function nodes, including indicators such as spatial proximity, functional complementarity, and pedestrian flow interaction frequency;

[0052] Adopt a multi-layer graph convolutional network to extract the structural and functional features of temporary land use nodes;

[0053] Discover temporary land use communities with similar functional patterns and evolution trends based on the spectral clustering algorithm;

[0054] Analyze the location and importance of abnormal temporary land use in the urban function network, and evaluate its retention or transformation value.

[0055] Preferably, the generation of differentiated urban planning optimization suggestions includes:

[0056] Construct a multi-objective evaluation index system including economic benefits, social value, and environmental impact;

[0057] Based on the cellular automaton and system dynamics models, simulate urban evolution scenarios under different planning strategies;

[0058] Generate personalized planning suggestions according to the characteristics of abnormal temporary land use and urban function requirements, including demolition and reconstruction, function upgrading, structural transformation and other schemes;

[0059] Provide a visualization analysis tool to visually display the expected effects and impact assessments of planning suggestions;

[0060] Design a collaborative decision-making mechanism involving multiple parties to integrate the feedback from planners, citizens, and developers and optimize the planning scheme.

[0061] The present invention also provides a GIS-based urban land planning analysis system, including a processor and a memory; wherein, the memory stores a computer program, and when the computer program is executed by the processor, it implements the above-mentioned GIS-based urban land planning analysis method.

[0062] The present invention innovatively combines technologies such as multi-source heterogeneous data fusion, spatio-temporal modeling, deep feature extraction, time-series anomaly detection, and correlation network analysis to construct a complete closed-loop system for temporary land use analysis and planning, which has the following advantages and beneficial effects compared with the prior art:

[0063] 1. Data multidimensionality: By fusing multi-source data, multi-dimensional cognition of space + time + function + relationship is realized, fundamentally enhancing the depth of understanding of temporary land use.

[0064] 2. Identification accuracy: The adaptive feature extraction mechanism can accurately identify and classify various types of temporary land use, and the identification accuracy is improved by more than 30% compared with traditional methods.

[0065] 3. Anomaly sensitivity: The change-sensitive time-series model is highly sensitive to the abnormal change patterns of temporary land use and can timely detect temporary buildings that have not been demolished after the planned period.

[0066] 4. Correlation insight: The complex interaction relationship between temporary land use and the urban functional network is revealed through the graph neural network, promoting the point-like analysis to networked understanding.

[0067] 5. Decision-making foresight: Based on multi-objective evaluation and scenario simulation, it provides differentiated planning optimization suggestions, realizing the paradigm shift from passive response to proactive planning.

[0068] Generally speaking, through technological innovation, the present invention realizes the transformation of temporary land use analysis from the traditional paradigm of static, isolated, and passive to the modern paradigm of dynamic, networked, and proactive, providing a new perspective and powerful tool for urban land planning. Brief Description of the Drawings

[0069] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the embodiments will be briefly introduced below.

[0070] Figure 1 is a schematic flow chart of a GIS-based urban land use planning analysis method according to an embodiment of the present invention.

[0071] Figure 2 is a schematic diagram of multi-source heterogeneous spatio-temporal data acquisition according to an embodiment of the present invention.

[0072] Figure 3 is a schematic structural diagram of a multi-scale spatio-temporal cube model according to an embodiment of the present invention.

[0073] Figure 4 is a working flow chart of an adaptive feature extraction mechanism according to an embodiment of the present invention.

[0074] Figure 5 It is a schematic structural diagram of the hierarchical classification system for temporary land use in an embodiment of the present invention.

[0075] Figure 6 It is a processing flow chart for detecting abnormal temporary land use in an embodiment of the present invention.

[0076] Figure 7 It is a structural block diagram of a GIS-based urban land use planning analysis system in an embodiment of the present invention. Specific embodiments

[0077] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0078] Embodiment 1: Overall process of the GIS-based urban land use planning analysis method

[0079] As Figure 1 shown, a GIS-based urban land use planning analysis method provided by an embodiment of the present invention mainly includes the following steps:

[0080] Step S1: Obtain multi-source heterogeneous spatio-temporal data, including remote sensing image data, historical planning data, mobile terminal collected data, and social media geotagged data.

[0081] Step S2: Based on the obtained multi-source heterogeneous spatio-temporal data, construct a unified multi-scale spatio-temporal cube model and generate standardized multi-modal urban temporary land use characterization data.

[0082] Step S3: According to the standardized multi-modal urban temporary land use characterization data, use an adaptive feature extraction mechanism to identify urban temporary land use areas and construct a hierarchical classification system for temporary land use.

[0083] Step S4: Based on the hierarchical classification system for temporary land use and historical time series data, establish a change-sensitive time series model to detect and identify temporary land use areas with abnormal change patterns.

[0084] Step S5: According to the identified temporary land use areas with abnormal change patterns, construct an association graph model between temporary land use and the urban function network, and analyze the interaction relationship between abnormal temporary land use and the urban function system.

[0085] Step S6: Based on the interaction relationship and multi-objective evaluation indicators, generate differentiated urban planning optimization suggestions to support the efficient allocation and management of urban land resources.

[0086] Through the above steps, the method of the embodiment of the present invention realizes a complete closed loop from data acquisition, spatio-temporal modeling, feature extraction, anomaly detection, correlation analysis to decision support, and establishes a scientific and complete analysis and planning framework for temporary land use. This method organically integrates multi-source heterogeneous data, which can not only accurately identify temporary land use, but also discover abnormal change patterns, and provides targeted planning suggestions in combination with the urban functional network, thereby improving the management efficiency and planning scientificity of urban land resources.

[0087] Embodiment 2: Acquisition of multi-source heterogeneous spatio-temporal data

[0088] In this embodiment, as Figure 2 shown, the specific steps for acquiring multi-source heterogeneous spatio-temporal data include:

[0089] First, obtain optical and radar remote sensing image data of the target urban area from a remote sensing satellite platform. The optical remote sensing data mainly includes high-resolution satellite images, such as sub-meter resolution images provided by QuickBird and WorldView series, as well as medium-resolution Landsat and Sentinel series images. The radar remote sensing data includes SAR (synthetic aperture radar) images, which have all-weather and all-day observation capabilities and can make up for the limitations of optical images in cloudy and foggy weather. Preferably, select image data with a spatial resolution in the range of 0.5 - 5 meters to ensure the effective identification of temporary buildings.

[0090] Second, retrieve the historical planning data and land use information of the target urban area from the urban planning database. The historical planning data includes land use planning maps, urban master plans, regulatory detailed plans, and various special planning data from different periods. The land use information includes attribute data such as land use nature, planning period, and construction intensity. These data are usually stored in the spatial database of the urban planning management department and have high reliability and authority.

[0091] Third, collect ground real-scene data and location information of the target urban area through mobile terminal devices. The ground real-scene data is mainly obtained through a dedicated collection application program on the mobile terminal, including ground photos, videos, 3D point cloud data, etc., and is accompanied by accurate geographical location information and shooting time. Preferably, build a distributed mobile collection network to integrate the collection data of urban management personnel, professional investigators, and volunteers to form a widely covered ground real-scene database.

[0092] Fourth, extract geotagged data and user activity information related to the target urban area from social media platforms. This includes social media texts, pictures, and video content with geographical location tags, reflecting the public's usage and subjective feelings of urban space. Preferably, semantic analysis and computer vision techniques are used to extract information related to temporary land use from social media data, such as building exteriors, usage status, public evaluations, etc.

[0093] Finally, conduct a quality assessment of the acquired various types of data, construct a data reliability scoring system based on information entropy and texture features, and determine the weight coefficients of each data source. The data reliability scoring system comprehensively considers factors such as data timeliness, spatial accuracy, integrity, and consistency, providing a scientific basis for subsequent data fusion.

[0094] The information entropy calculation method for evaluating data reliability is as follows:

[0095]

[0096] Among them, H(X) represents the information entropy of data source X, p(x i ) represents the probability of data element x i , and n represents the total number of data elements. The higher the information entropy, the greater the uncertainty contained in the data, and more processing and verification are required.

[0097] Based on information entropy and other evaluation indicators, calculate the weight coefficients of each data source:

[0098]

[0099] Among them, W i represents the weight coefficient of the i-th data source, H i represents the normalized information entropy of the i-th data source, Q i represents the quality score of the data source (the value range is 0 - 1, representing the comprehensive score of data integrity, timeliness, and reliability), α is a regulation parameter (the value range is usually 0 - 1, used to control the influence degree of information entropy in weight calculation), and m is the total number of data sources.

[0100] Through the above steps, the comprehensive acquisition and quality assessment of multi-source heterogeneous spatio-temporal data are realized, laying a solid foundation for subsequent data fusion and analysis. Different types of data complement each other, jointly constituting a multi-dimensional observation network for urban temporary land use, significantly improving the comprehensiveness and reliability of the data.

[0101] Example 3: Construction of a multi-scale spatio-temporal cube model

[0102] In this example, as Figure 3 shown, the specific steps for constructing a unified multi-scale spatio-temporal cube model include:

[0103] First, perform spatio-temporal alignment processing on multi-source heterogeneous data based on the feature point matching algorithm to ensure the consistency of data from different sources in the time and space dimensions. Feature point matching uses an improved SIFT (Scale-Invariant Feature Transform) algorithm, combined with a Spatial Transformer Network (STN) to achieve precise registration of heterogeneous data. Preferably, for the registration of remote sensing images and ground real-scene photos, a feature matching method with multi-view geometric constraints is used, and the registration accuracy can reach the sub-pixel level.

[0104] The key step in feature point matching is to calculate the similarity between feature descriptors, which is calculated using cosine similarity:

[0105]

[0106] where S(A,B) represents the similarity between feature descriptors A and B, A i and B i represent the i-th component of feature descriptors A and B respectively, and n represents the dimension of the feature descriptor. The closer the similarity value is to 1, the higher the matching degree of the two feature points.

[0107] Secondly, construct a four-dimensional spatio-temporal data structure, integrating the spatial coordinates (x, y, z) and the time dimension (t) into a unified data representation model. The four-dimensional spatio-temporal data structure adopts an organizational method combining an octree and a time axis, supporting multi-resolution spatial data storage and efficient retrieval. Preferably, the storage granularity of spatio-temporal data can be adaptively adjusted according to the importance and change frequency of urban areas. More refined spatial resolutions and time intervals are used for key areas and areas with frequent changes.

[0108] Thirdly, perform scale transformation on spatial data with different resolutions to achieve unified management of multi-scale spatial information. Scale transformation uses a method combining wavelet transform and pyramid structure to maintain the information integrity of data at different scales. Preferably, construct a 5-layer scale pyramid, with the lowest layer resolution of 0.5 meters and the resolution ratio between adjacent layers of 2:1, to achieve multi-scale expression from building details to urban area overview.

[0109] Fourthly, perform interpolation processing on data with discontinuous time series to establish continuous and equally spaced time series data. Time series interpolation uses a multi-dimensional interpolation method based on tensor decomposition, which can consider the correlation in both spatial and time dimensions simultaneously. Preferably, according to the time distribution characteristics of the data, linearly interpolate, spline interpolate, or use an interpolation method based on deep learning adaptively to ensure the accuracy of the interpolation result.

[0110] Finally, semantic-level fusion and standardization of multi-source data are performed to ensure the consistency of data from different sources at the semantic level. Semantic fusion uses ontology mapping and knowledge graph technologies to establish semantic associations between different data sources. Preferably, an ontology model for the temporary land use field is constructed, which includes three levels: category, attribute, and relationship, to achieve semantic unification across data sources.

[0111] Through the above steps, in this embodiment, a unified multi-scale spatio-temporal cube model is constructed, which converts multi-source heterogeneous data into standardized multi-modal urban temporary land use representation data. This model not only achieves spatio-temporal and scale unification of data, but also establishes associations and mappings at the semantic level, laying a foundation for subsequent feature extraction and analysis of temporary land use.

[0112] Embodiment 4: Adaptive Feature Extraction Mechanism

[0113] In this embodiment, as Figure 4 shown, the specific steps of using the adaptive feature extraction mechanism to identify the urban temporary land use area include:

[0114] First, a multi-scale feature extraction network is constructed to perform hierarchical feature extraction on urban area images. The multi-scale feature extraction network adopts the Feature Pyramid Network (FPN) architecture and combines the Deep Residual Network (ResNet) as the backbone network to achieve effective extraction from low-level features to high-level semantic features. Preferably, the network depth is set to 50 layers, including 5 feature scales, which can simultaneously capture the detailed features and context information of temporary buildings.

[0115] The core of the multi-scale feature pyramid network is the feature fusion mechanism. The top-down and bottom-up feature transmission can be expressed as:

[0116] P l =Conv(UpSample(P l-1 )+Lateral l ),

[0117] where P l represents the feature map of the l-th layer, UpSample represents the upsampling operation (usually nearest neighbor interpolation or bilinear interpolation, used to double the resolution of the feature map), Lateral l represents the lateral connection feature of the l-th layer (obtained by 1×1 convolution of the feature of the corresponding layer of the backbone network), and Conv represents the convolution operation (usually 3×3 convolution, used to eliminate the aliasing effect caused by upsampling). This structure can effectively combine high-level semantic information with low-level location information.

[0118] Secondly, fuse spectral features, texture features, morphological features, and context features to construct a multi-dimensional feature representation of temporary land. Spectral features mainly utilize the reflection characteristics of different ground objects in each band. Texture features are extracted using the gray-level co-occurrence matrix and local binary pattern. Morphological features include geometric descriptors such as area, perimeter, and compactness. Context features consider the mutual relationship between the target and the surrounding environment. Preferably, for the identification of temporary buildings, key attention is paid to material texture, regularity, and time-varying features, which are the key indicators for distinguishing temporary buildings from permanent buildings.

[0119] Thirdly, optimize the feature weight allocation based on the attention mechanism to highlight the key distinguishing features of temporary land. The attention mechanism combines channel attention and spatial attention to adaptively adjust the importance of features. Channel attention focuses on what features, and spatial attention focuses on which location. The combination of the two can accurately locate the key feature regions. Preferably, the attention module adopts a lightweight design, and the computational complexity is controlled within 5% of the original network, while maintaining the computational efficiency while improving the performance.

[0120] The calculation process of the channel attention mechanism is as follows:

[0121] M c (F) = σ(MLP(AvgPool(F)) + MLP(MaxPool(F))),

[0122] where M c represents the channel attention map (the value range is [0, 1], indicating the importance weight of each channel), F represents the input feature map, AvgPool and MaxPool respectively represent global average pooling and max pooling (compressing the feature map into a 1×1×C vector, C is the number of channels), MLP represents a multi-layer perceptron (usually including two fully connected layers, with a dimensionality reduction layer in the middle), and σ represents the sigmoid activation function. The value range of the channel attention map is [0, 1], indicating the importance weight of each channel.

[0123] Fourthly, achieve the precise boundary extraction of the temporary land area through the context-aware segmentation algorithm. The context-aware segmentation algorithm is based on the DeepLabv3+ architecture, combines dilated convolution and encoder-decoder structure, and retains spatial detail information while maintaining the receptive field size. Preferably, a conditional random field (CRF) is used as a post-processing module to further optimize the smoothness and accuracy of the segmentation boundary.

[0124] Finally, attribute annotation is performed on the extracted temporary land areas, including information such as area, shape, material, and use. The attribute annotation adopts a multi-task learning framework, performing semantic segmentation and attribute classification simultaneously to improve the efficiency of feature utilization. Preferably, dedicated branch networks are set up for different attributes, such as a material recognition branch, a use classification branch, etc. Each branch shares the features of the backbone network but has an independent output layer.

[0125] Through the above steps, this embodiment realizes the accurate recognition and attribute annotation of urban temporary land areas. The adaptive feature extraction mechanism can flexibly adjust the feature extraction strategy according to data characteristics and task requirements, significantly improving the accuracy and robustness of temporary land recognition and laying a foundation for subsequent classification and anomaly detection.

[0126] Embodiment 5: Construction of a Hierarchical Classification System for Temporary Land

[0127] In this embodiment, as Figure 5 shown, the specific steps for constructing a hierarchical classification system for temporary land include:

[0128] First, a three-level classification framework is established, including a category layer (temporary building type), a function layer (usage function), and a use layer (specific use). The three-level classification framework adopts a hierarchical organizational structure, and each level of classification is a refinement and concretization of the previous level. Preferably, the classification system adopts a top-down design method, first establishing a conceptual model, and then performing instantiation and verification to ensure the scientificity and practicality of the classification system.

[0129] Secondly, for the category layer, temporary land is classified into temporary building land, temporary facility land, temporary activity land, and temporary storage land. Temporary building land mainly includes prefabricated houses, container houses, temporary work sheds, etc.; temporary facility land includes temporary roads, bridges, pipelines, etc.; temporary activity land includes exhibition sites, market sites, etc.; temporary storage land includes material yards, temporary waste storage areas, etc. Preferably, based on physical characteristics and spatial forms, clear discrimination criteria are defined for each category. For example, temporary building land usually has regular shapes, obvious boundaries, and roof features.

[0130] Thirdly, for the function layer, temporary land is classified into production function, living function, service function, and public function. The production function mainly refers to temporary land used for economic activities such as manufacturing and resource extraction; the living function includes temporary accommodation, dining, etc. for meeting basic living needs; the service function involves temporary venues for commercial, office, educational, etc. service activities; the public function includes temporary land for public services such as emergency rescue and public health. Preferably, based on the characteristics of land use activities and service objects, discrimination rules for the function layer are established, such as distinguishing different function types through indicators such as population density, activity time patterns, and service facility configurations.

[0131] Fourth, for the usage layer, it is subdivided into emergency relief, engineering construction, commercial exhibition, cultural activities and other sub - usages according to specific application scenarios. Emergency relief includes post - disaster resettlement sites, temporary medical sites, etc.; engineering construction includes construction sites, road construction, etc.; commercial exhibition includes temporary markets, exhibition venues, etc.; cultural activities includes performance venues, celebration sites, etc. Preferably, in combination with urban planning requirements and management practices, the classification of the usage layer is updated regularly to adapt to urban development changes and the emergence of new usages.

[0132] Finally, based on the weakly - supervised learning method, a classification model is trained through a small number of labeled samples to achieve the automatic classification of large - scale temporary land uses. Weakly - supervised learning adopts an active learning strategy, preferentially selecting the most informative and representative samples for annotation to maximize the annotation efficiency. Preferably, a sample selection strategy based on uncertainty is adopted, selecting the samples with the lowest prediction confidence of the model for annotation to iteratively optimize the classification model.

[0133] The sample selection strategy of active learning can be achieved through information entropy:

[0134]

[0135] where \(H(x)\) represents the prediction uncertainty of sample \(x\), \(p(y\) i | \(x)\) represents the prediction probability that sample \(x\) belongs to class \(y\) i , and \(C\) represents the total number of classes. The higher the information entropy, the greater the prediction uncertainty, and the more worthy the sample is of being annotated.

[0136] Through the above steps, this embodiment constructs a complete hierarchical classification system for temporary land uses, achieving multi - level classification from macroscopic categories to specific usages. This classification system not only considers the physical characteristics and functional attributes of temporary land uses, but also combines urban planning and management requirements, providing a scientific framework for the systematic management of temporary land uses. The introduction of the weakly - supervised learning method significantly reduces the annotation cost of training the classification model, improving the automation degree and coverage of classification.

[0137] Embodiment 6: Establishment of a change - sensitive time - series model

[0138] In this embodiment, the specific steps for establishing a change - sensitive time - series model are as follows:

[0139] First, according to the planning period and expected duration of temporary land uses, time - window parameters are constructed. The time - window parameters include the start time of observation, the end time of observation, the observation frequency, and key time points, etc., which are used to define the scope and accuracy of time - series analysis. Preferably, according to the characteristics of different types of temporary land uses, differentiated time - window parameters are set. For example, construction sites usually require a longer observation period and a higher observation frequency, while temporary markets require capturing periodic change patterns.

[0140] Secondly, design a change-sensitive loss function for the characteristics of temporary land use to improve the detection sensitivity to changes in small targets and sparsely distributed temporary buildings. The change-sensitive loss function combines Focal Loss and Boundary-Sensitive Loss, giving higher weights to difficult-to-separate samples and boundary regions. Preferably, the parameters in the loss function are adaptively adjusted according to the characteristics of the data distribution to ensure good detection ability for changes in temporary land use of different scales and types.

[0141] The mathematical expression of the change-sensitive loss function is as follows:

[0142]

[0143] where p t represents the predicted probability of the model for the change category, α and γ are the adjustment parameters of the Focal Loss (α is usually in the range of 0.25 - 0.75, and γ is usually 2 - 5, used to adjust the weight ratio of easy and difficult samples), β is the weight coefficient of the boundary-sensitive term (usually in the range of 0.1 - 0.5), Ω b represents the set of pixels in the boundary region, y i and represent the true label and predicted value of pixel i respectively, and ||·||2 represents the L2 norm. The first term focuses on difficult-to-separate samples, and the second term emphasizes boundary accuracy.

[0144] Thirdly, construct a bidirectional temporal encoder to encode and analyze the historical change sequence of temporary land use. The bidirectional temporal encoder is based on the bidirectional long short-term memory network (BiLSTM) and the attention mechanism, and can capture the forward and backward dependencies of temporal data simultaneously. Preferably, the hidden state dimension of the temporal encoder is set to 256, the input feature dimension is consistent with the dimension of the multi-modal features, and the time step is determined according to the observation frequency.

[0145] The calculation process of the bidirectional temporal encoder can be expressed as:

[0146]

[0147] where, and represent the hidden states of the forward and backward LSTMs respectively (the forward LSTM processes from the beginning of the sequence to the end, and the backward LSTM processes from the end of the sequence to the beginning), represents the hidden state of the forward LSTM at time step t - 1, represents the hidden state of the backward LSTM at time step t + 1, x t represents the input feature at time step t, and LSTM fand LSTM b represent the computational functions of the forward and backward LSTM cells respectively, and h t represents the concatenated bidirectional hidden state (usually the concatenation of the hidden states in two directions), and c t represents the context vector generated through the attention mechanism, α t , i represents the attention weight of time step t to time step i, T represents the total number of time steps, represents the vector concatenation operation.

[0148] Fourth, introduce an urban planning rule base to constrain anomaly determination with domain knowledge and reduce the misjudgment rate. The planning rule base includes rules such as the legal survival time of temporary land use, the permitted change range, and conversion conditions. Through a combination of rule reasoning and machine learning, the preliminary judgment of the model is verified and constrained. Preferably, the rule base adopts a knowledge representation method combining ontology and rules, which supports the expression and reasoning of complex rules and has the flexibility of updating and expansion.

[0149] Finally, adopt a multi-evidence fusion method based on the evidence theory to quantify the reliability of anomaly determination and generate a confidence score. The evidence theory framework synthesizes multi-source evidences such as model prediction, rule judgment, and historical data to generate a more reliable anomaly determination result. Preferably, the Dempster-Shafer evidence theory is used for multi-source evidence fusion, and the uncertainty is quantified through mass functions, confidence functions, and likelihood functions to provide a reliable basis for decision-making.

[0150] The basic formula for multi-evidence fusion is:

[0151]

[0152] where, m 1,2 (A) represents the basic probability assignment of evidence A after fusion, m1(B) and m2(C) respectively represent the basic probability assignments of two evidence sources to sets B and C, and K represents the conflict degree, defined as B∩C = A means that the intersection of set B and set C is equal to set means that the intersection of set B and set C is an empty set. Through this formula, effective fusion of multi-source evidences can be achieved.

[0153] Through the above steps, this embodiment establishes a change-sensitive time series model, realizing precise modeling of the change process of temporary land use and sensitive detection of abnormal patterns. This model comprehensively utilizes deep learning, attention mechanism, and knowledge constraints, and can not only detect abnormal temporary buildings that have not been demolished after the planned period, but also provide a reliable confidence assessment, providing a scientific basis for subsequent decision-making.

[0154] Example 7: Detection of Temporary Land Use with Abnormal Change Patterns

[0155] In this embodiment, as Figure 6 shown, the specific steps for detecting and identifying the temporary use area with abnormal change patterns include:

[0156] First, fit the time-series change curve of the temporary use area to detect abnormal points deviating from the expected change pattern. The time-series change curve is constructed based on multi-temporal remote sensing images and ground observation data, reflecting the whole process of the temporary land from establishment to demolition or transformation. Preferably, a method combining polynomial regression and seasonal decomposition is adopted to model the trend component and the periodic component respectively, improving the fitting ability for complex change patterns.

[0157] The fitting model of the time-series change curve can be expressed as:

[0158] Y(t) = T(t) + S(t) + R(t),

[0159] where Y(t) represents the observed value at time t, T(t) represents the trend component (usually represented by a polynomial function), S(t) represents the seasonal component (usually represented by Fourier series or seasonal dummy variables), and R(t) represents the residual component (including random fluctuations and abnormal changes). The trend component is obtained through polynomial regression, and the seasonal component is obtained through Fourier transform or seasonal exponential decomposition. Abnormal points are mainly detected through the statistical characteristics of the residual component R(t).

[0160] Secondly, calculate the anomaly score based on the change rate and the change amplitude, and mark the temporary use areas exceeding the preset threshold as potential anomalies. The change rate refers to the change speed within a specific time window, and the change amplitude refers to the gap between the observed value and the expected value. Preferably, the anomaly score adopts a method combining Z-score and modified extreme value analysis to adapt to the change characteristics of different types of temporary land and reduce the false alarm rate.

[0161] The calculation formula for the anomaly score is:

[0162]

[0163] where S anomaly (t) represents the anomaly score at time t, Z(·) represents Z-score standardization (standardizing the data into a distribution with a mean of 0 and a standard deviation of 1), represents the change rate (which can be approximately calculated by difference), represents the gap between the observed value and the predicted value, and ω1 and ω2 are weight coefficients (satisfying ω1 + ω2 = 1, used to balance the importance of the change rate and the change amplitude).

[0164] Third, combine the urban planning cycle and land use conversion rules to verify the authenticity of potential abnormal areas. The planning cycle provides the official basis for the expected duration of temporary land use, while the land use conversion rules define the target types and conditions for the conversion of temporary land use. Preferably, construct a planning knowledge graph to integrate the rules scattered in various planning documents into structured knowledge, enabling efficient rule matching and reasoning.

[0165] Fourth, classify the identified abnormal temporary land use areas, distinguishing abnormal types such as those not demolished on schedule, those with functional changes, and those with structural strengthening. Those not demolished on schedule refer to temporary land uses that have not been demolished beyond the planned period; those with functional changes refer to temporary land uses with changed uses; those with structural strengthening refer to temporary land uses that gradually transform from temporary structures to permanent structures. Preferably, based on abnormal features and change patterns, use hierarchical clustering methods to automatically discover abnormal types, improving the adaptability and completeness of abnormal classification.

[0166] Finally, generate a spatial distribution heat map of abnormal temporary land use to visually display the agglomeration of abnormal areas. The spatial distribution heat map is based on the kernel density estimation method, which converts discrete abnormal points into a continuous density surface, reflecting the spatial agglomeration characteristics of abnormal phenomena. Preferably, the kernel function and bandwidth parameters of the heat map are adaptively adjusted according to the urban scale and abnormal distribution characteristics to ensure the expression effect and analysis value of the heat map.

[0167] The calculation formula for kernel density estimation is:

[0168]

[0169] where f(x) represents the density estimate value at position x, n represents the number of abnormal points, h represents the bandwidth parameter (controlling the smoothness degree, usually the optimal value is selected through cross-validation), K(·) represents the kernel function, and x i represents the position of the i-th abnormal point. Commonly used kernel functions include Gaussian kernel, Epanechnikov kernel, etc.

[0170] Through the above steps, this embodiment realizes the accurate detection and classification of temporary land use with abnormal change patterns. This method can not only identify illegal temporary land uses that have not been demolished on schedule, but also discover potential transformation trends such as functional changes and structural strengthening, providing a comprehensive monitoring means and early warning mechanism for urban planning management. At the same time, the spatial distribution heat map visually displays the spatial distribution law of abnormal phenomena, helping to discover systematic problems and regional characteristics.

[0171] Embodiment 8: Construction of the association graph model between temporary land use and urban functional network

[0172] In this embodiment, the specific steps for constructing the association graph model between temporary land use and urban functional network include:

[0173] First, based on the spatial relationship between the temporary land and the surrounding permanent buildings, a heterogeneous spatial relationship graph is constructed. The heterogeneous spatial relationship graph represents the temporary land and the permanent buildings as different types of nodes, and the spatial relationships as edges, forming a complex network structure with multiple types of nodes and multiple relationship edges. Preferably, the spatial relationships include various types such as distance relationships, direction relationships, topological relationships, etc., comprehensively reflecting the spatial interaction patterns between the temporary land and the surrounding environment.

[0174] The formal definition of the heterogeneous spatial relationship graph is as follows:

[0175] G = (V T ∪ V P , E, R, φ, ψ),

[0176] where G represents the heterogeneous graph, V T represents the set of temporary land nodes, V P represents the set of permanent building nodes, E represents the set of edges, R represents the set of relationship types (such as distance relationships, functional relationships, etc.), φ: V T ∪ V P → A represents the mapping from nodes to attributes (A is the set of attributes), and ψ: E → R represents the mapping from edges to relationship types. Secondly, calculate the association strength between the temporary land and the surrounding urban functional nodes, including indicators such as spatial proximity, functional complementarity, and frequency of human flow interaction. The spatial proximity is calculated based on Euclidean distance or network distance, the functional complementarity measures the degree of functional cooperation, and the frequency of human flow interaction estimates the personnel flow situation between each other based on mobile trajectory data. Preferably, the calculation of the association strength adopts a multi-index comprehensive evaluation method, and the weights of each index are determined by the entropy weight method to generate a comprehensive association strength index.

[0177] The calculation formula for the association strength is as follows:

[0178]

[0179] where S(v i , v j ) represents the comprehensive association strength between nodes v i and v j , s k (v i , v j ) represents the association strength of the k-th index (such as spatial proximity, functional complementarity, etc.), w k represents the weight of the k-th index (satisfying ), and K represents the total number of indexes. The weight w k is calculated by the entropy weight method, and the index values are normalized to ensure comparability.

[0180] Third, a multi-layer graph convolutional network is adopted to extract the structural and functional features of temporary land use nodes. The multi-layer graph convolutional network diffuses and aggregates the local information of nodes layer by layer through a message passing mechanism to form node representations containing rich context information. Preferably, for the characteristics of heterogeneous graphs, a relationship-aware graph convolution operation is adopted, and different parameter matrices are used for different types of relationships to improve the accuracy of feature extraction.

[0181] The relationship-aware graph convolution operation can be expressed as:

[0182]

[0183] where, represents the feature representation of node i at the l-th layer (in vector form), represents the feature representation of node i at the (l + 1)-th layer, r ∈ R indicates that the relationship type r belongs to the set of relationship types R, represents the set of neighbor nodes connected to node i through the relationship r, represents the set the number of elements of (used for normalizing neighbor information), represents the weight matrix of relationship r at the l-th layer (learnable parameter), represents the weight matrix of the self-loop (used to retain the information of the node itself), and σ represents the activation function (such as ReLU, tanh, etc.).

[0184] Fourth, based on the spectral clustering algorithm, temporary land use communities with similar functional patterns and evolution trends are discovered. Spectral clustering is based on the eigenvectors of the graph Laplacian matrix for clustering, and can effectively capture the community characteristics in the network structure. Preferably, the number of clusters is adaptively determined, and the best clustering scheme is evaluated through eigenvalue analysis and silhouette coefficient to discover the functionally related temporary land use communities.

[0185] The core step of spectral clustering is to calculate the graph Laplacian matrix: L = D - A.

[0186] where, L represents the graph Laplacian matrix, D represents the degree matrix (the diagonal element D ii is the degree of node i, that is, the number of edges connected to node i), and A represents the adjacency matrix (the element A ij indicates whether there is an edge connection between node i and node j, 1 if there is, and 0 if not). By calculating the eigenvectors of L and applying traditional clustering algorithms such as K-means in the feature space, the community structure in the graph can be discovered.

[0187] Finally, analyze the location and importance of abnormal temporary land in the urban functional network, and evaluate its value of retention or transformation. The location importance is measured by centrality indicators, including degree centrality, betweenness centrality, eigenvector centrality, etc.; the functional importance considers its role and contribution in meeting urban functional requirements. Preferably, combined with socio-economic indicators and spatial planning goals, a comprehensive evaluation system is constructed to scientifically evaluate the disposal priority and transformation value of abnormal temporary land.

[0188] Through the above steps, this embodiment constructs an association graph model between temporary land and the urban functional network, realizing a methodological leap from point-like analysis to networked understanding. This model not only reveals the complex interaction relationship between temporary land and the urban functional system, but also discovers communities of temporary land with similar functional patterns, providing a systematic analysis perspective and scientific basis for urban planning.

[0189] Embodiment 9: Generation of Differentiated Urban Planning Optimization Suggestions

[0190] In this embodiment, the specific steps for generating differentiated urban planning optimization suggestions include:

[0191] First, construct a multi-objective evaluation index system including economic benefits, social value, and environmental impact. Economic benefit indicators include land use efficiency, industrial driving force, return on investment, etc.; social value indicators include service coverage, people's livelihood improvement, fairness and sharing, etc.; environmental impact indicators include ecological impact, resource consumption, pollution emissions, etc. Preferably, according to the focus of urban planning and urban characteristics, set differentiated index weights to reflect the development demands and characteristic positioning of different cities.

[0192] The comprehensive score calculation method for multi-objective evaluation is:

[0193]

[0194] Among them, S represents the comprehensive score, w i represents the weight of the i-th index (satisfying ), x i represents the actual value of the i-th index, min i and max i represent the minimum and maximum values of the i-th index respectively, and N represents the total number of indicators. This calculation method normalizes each index to the [0,1] interval to ensure the comparability of indicators with different dimensions. Secondly, based on the cellular automata and system dynamics models, simulate the urban evolution scenarios under different planning strategies. The cellular automata simulate the evolution of the urban spatial form, and the system dynamics model simulates the complex feedback relationships among the various elements of the urban system. Preferably, couple the two models into a unified simulation framework to achieve the coordinated simulation of spatial dynamics and system dynamics, and improve the accuracy and integrity of scenario prediction.

[0195] The state transition rule of the cellular automaton can be expressed as:

[0196]

[0197] where represents the state of position (i, j) at time t (such as land use type), represents the set of neighborhood states of position (i, j) at time t (usually Moore neighborhood or von Neumann neighborhood), and f represents the state transition function (determines the state at the next moment based on the neighborhood state and transition rule). By defining reasonable transition rules, the evolution process and spatial diffusion pattern of temporary land use can be simulated.

[0198] Third, according to the characteristics of abnormal temporary land use and the urban function requirements, generate personalized planning suggestions, including demolition and reconstruction, function upgrading, structural transformation and other schemes. Demolition and reconstruction are applicable to temporary land use that has no retention value and does not conform to the plan; function upgrading is applicable to temporary land use that is beneficial to the improvement of urban functions; structural transformation is applicable to temporary land use that needs to be optimized and upgraded. Preferably, use the case-based reasoning method to retrieve similar situations from the historical planning case library, extract experiences and lessons, and assist in formulating targeted planning suggestions.

[0199] The generation process of personalized planning suggestions can be expressed as:

[0200]

[0201] where R represents the optimal planning suggestion, represents the set of optional suggestions, and sim(C r , C t ) represents the similarity between case C r and the target case C t (usually using cosine similarity or the reciprocal of Euclidean distance), E(r) represents the historical effectiveness of suggestion r (success rate based on historical cases), F(r, G) represents the coordination degree between suggestion r and the urban function network G (evaluates the impact on the urban function network after the implementation of the suggestion), and α, β, and γ are weight coefficients (satisfying α + β + γ = 1, used to balance the importance of similarity, effectiveness, and coordination degree).

[0202] Fourth, provide a visualization analysis tool to intuitively display the expected effects and impact assessments of the planning suggestions. The visualization analysis tool includes various forms such as two-dimensional floor plans, three-dimensional scene diagrams, index radar charts, and impact relationship diagrams, which comprehensively display the spatial layout, morphological characteristics, performance indicators, and system impacts of the planning suggestions. Preferably, use immersive interactive technology to support planners in browsing, comparing, and adjusting the schemes, and improve the intuitiveness and interactivity of planning decisions.

[0203] Finally, design a collaborative decision-making mechanism involving multiple parties to integrate the feedback from planners, citizens, and developers and optimize the planning scheme. The collaborative decision-making mechanism establishes communication channels and negotiation spaces for multiple stakeholders through a combination of online platforms and offline workshops. Preferably, a method combining the analytic hierarchy process and the Delphi method is used to scientifically collect and integrate the opinions of all parties and form a planning scheme with a broad basis of consensus.

[0204] Through the above steps, this embodiment realizes the generation of differentiated urban planning optimization suggestions. This method transforms the abnormal detection results and functional association analysis results of temporary land into specific planning actions, forming a complete closed-loop from data analysis to decision support. The introduction of mechanisms such as multi-objective evaluation, scenario simulation, personalized suggestions, and collaborative decision-making ensures the scientific nature, forward-looking nature, and feasibility of the planning suggestions, effectively supporting the efficient allocation and management of urban land resources.

[0205] Embodiment 10: GIS-based Urban Land Planning Analysis System

[0206] As Figure 7 shown, the embodiment of the present invention also provides a GIS-based urban land planning analysis system, including a processor and a memory; wherein, the memory stores a computer program, and when the computer program is executed by the processor, it realizes the GIS-based urban land planning analysis method as described in Embodiments 1-9.

[0207] This system mainly includes the following functional modules:

[0208] Multi-source data acquisition module 1 is responsible for acquiring multi-source heterogeneous spatio-temporal data from remote sensing satellite platforms, urban planning databases, mobile terminal devices, and social media platforms, and performing quality assessment and preprocessing.

[0209] Spatio-temporal cube construction module 2 is responsible for performing spatio-temporal alignment processing on multi-source heterogeneous data, constructing a four-dimensional spatio-temporal data structure, and realizing unified management of multi-scale spatial information and time series interpolation processing.

[0210] Adaptive feature extraction module 3 is responsible for constructing a multi-scale feature extraction network, fusing multi-dimensional features, optimizing feature weight allocation, and realizing accurate identification and attribute annotation of temporary land areas.

[0211] Hierarchical classification management module 4 is responsible for establishing a three-level classification framework, realizing classification of temporary land at the category layer, function layer, and use layer, and supporting large-scale automatic classification through weak supervision learning methods.

[0212] The temporal change analysis module 5 is responsible for constructing time window parameters, designing a change-sensitive loss function, and realizing the modeling of the temporary land use change process and the detection of abnormal patterns.

[0213] The abnormal recognition and evaluation module 6 is responsible for fitting the temporal changes of temporary land use and detecting abnormalities, verifying the authenticity of abnormalities in combination with planning rules, and generating heat maps of abnormal classifications and spatial distributions.

[0214] The associated network analysis module 7 is responsible for constructing an associated graph model of temporary land use and urban functions, calculating the association strength, extracting structured features, discovering functional communities, and evaluating the location importance of abnormal temporary land use.

[0215] The planning recommendation generation module 8 is responsible for constructing a multi-objective evaluation index system, simulating urban evolution scenarios of different planning strategies, generating personalized planning recommendations, and providing visualization analysis tools and collaborative decision-making mechanisms.

[0216] The data storage and management module 9 is responsible for managing the data resources of the system, including raw data, processing intermediate results, and analysis results, and supporting efficient data retrieval and update.

[0217] The user interaction interface module 10 is responsible for providing a friendly human-computer interaction interface, supporting data visualization, analysis operations, and result display, and improving the usability and user experience of the system.

[0218] Each functional module interacts through standardized data interfaces and service protocols to form a complete data processing and analysis pipeline. The system architecture adopts a microservices design, and each module can be independently deployed and extended to meet the application scenarios of different scales of cities and different analysis requirements.

[0219] In specific implementations, the system can be deployed on a cloud platform or a local server, supporting multi-terminal access from Web clients and mobile clients. The core algorithms and models of the system are accelerated by GPUs to achieve efficient computing, supporting parallel processing and real-time analysis of large-scale data.

[0220] The system provided in this embodiment is based on a microservices architecture and cloud-native technology, realizing a highly modular, scalable, and easy-to-maintain design, and providing a powerful temporary land use analysis and decision-making support tool for urban planning management departments. The system can not only meet the conventional monitoring and management requirements of temporary land use but also support the forward-looking research and innovative practices of urban planning.

[0221] To further illustrate the actual application effect of the present invention, the following is illustrated by an actual case.

[0222] This case selects a rapidly developing new area as the research object, focusing on the analysis of temporary construction land and planning optimization within this area. The research area covers approximately 50 square kilometers and includes various types of temporary construction land, such as construction sites, temporary markets, and temporary activity venues.

[0223] First, the system obtains the spatio-temporal data of this area from multiple data sources, including high-resolution satellite images (resolution 0.5 meters) in the past 5 years, historical planning documents in the urban planning database, ground real-scene photos provided by more than 300 volunteers, and geotagged data on social media platforms. The results of data quality assessment show that the overall quality score of satellite images is 0.92, the effective coverage rate of ground photos is 78%, and the spatio-temporal distribution uniformity of social media data is 0.65.

[0224] Second, the system constructs a multi-scale spatio-temporal cube model for this area, achieving unified management of data from different sources. The average accuracy of spatio-temporal alignment reaches the sub-meter level. The multi-scale spatial expression supports five levels of scales from 0.5 meters to 100 meters, and the interpolation accuracy of time series data exceeds 95%.

[0225] Third, the adaptive feature extraction mechanism successfully identifies 327 temporary use areas within this area, with a total area of approximately 3.2 square kilometers. The recognition accuracy reaches 92.3% through manual verification, significantly higher than 68.7% of the traditional single data source method. The system classifies these temporary use areas into 4 category layers, 4 function layers, and 12 use layers, forming a complete classification system.

[0226] Fourth, the change-sensitive time series model analysis finds that among the 327 temporary use lands, 42 show abnormal change patterns, including 28 cases of not being demolished on schedule, 9 cases of functional transformation, and 5 cases of structural strengthening. The accuracy of anomaly detection is 88.1%, the recall rate is 91.3%, and the F1 score is 89.7%.

[0227] Fifth, the correlation network analysis reveals the complex correlations between these 42 abnormal temporary use lands and the surrounding urban functions. The system constructs a heterogeneous relationship graph containing more than 2,500 nodes and more than 15,000 edges, extracts node features through a multi-layer graph convolutional network, and discovers 6 temporary use land communities with similar functional patterns. The analysis shows that the temporary commercial community near the transportation hub has the highest functional complementarity and retention value.

[0228] Finally, based on multi-objective evaluation and scenario simulation, the system generates differentiated planning suggestions for these 42 abnormal temporary use lands: 12 are recommended for demolition and reconstruction, 18 are recommended for functional upgrading, 8 are recommended for structural transformation, and 4 are recommended for overall retention. These suggestions are discussed and optimized through a collaborative decision-making mechanism involving multiple parties, and finally form a planning scheme with a wide range of recognition.

[0229] The implementation results of this case show that the method and system provided by the present invention can effectively support the analysis and planning decision-making of urban temporary land use. Through the complete process of multi-source data fusion, feature extraction, anomaly detection, association analysis, and generation of planning suggestions, the comprehensiveness, accuracy, and practicality of temporary land use analysis have been significantly improved, providing a scientific basis and decision-making support for urban planning management.

[0230] It should be noted that the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for analyzing urban land planning based on GIS, characterized in that, Including: Obtain multi-source heterogeneous spatio-temporal data, including remote sensing image data, historical planning data, mobile terminal collected data, and social media geotagged data; Based on the obtained multi-source heterogeneous spatio-temporal data, construct a unified multi-scale spatio-temporal cube model, and generate standardized multi-modal urban temporary land use characterization data; According to the standardized multi-modal urban temporary land use characterization data, adopt an adaptive feature extraction mechanism to identify urban temporary land use areas, and construct a hierarchical classification system for temporary land use; Based on the hierarchical classification system of temporary land use and historical time series data, establish a change-sensitive time series model to detect and identify abnormal change pattern areas of temporary land use; According to the identified abnormal change pattern areas of temporary land use, construct an association graph model between temporary land use and urban functional network, and analyze the interaction relationship between abnormal temporary land use and urban functional system; Based on the interaction relationship and multi-objective evaluation indicators, generate differentiated urban planning optimization suggestions to support the efficient allocation and management of urban land resources.

2. The method for analyzing urban land planning based on GIS according to claim 1, characterized in that, The obtaining of multi-source heterogeneous spatio-temporal data includes: Obtain optical and radar remote sensing image data of the target urban area from a remote sensing satellite platform; Retrieve historical planning data and land use information of the target urban area from the urban planning database; Collect ground real-scene data and location information of the target urban area through mobile terminal devices; Extract geotagged data and user activity information related to the target urban area from social media platforms; Conduct quality assessment on the obtained various types of data, construct a data reliability scoring system based on information entropy and texture features, and determine the weight coefficients of each data source.

3. A GIS-based urban land planning analysis method according to claim 1, characterized in that, The construction of the unified multi-scale spatio-temporal cube model includes: Perform spatio-temporal alignment processing on multi-source heterogeneous data based on the feature point matching algorithm to ensure the consistency of data from different sources in the time and space dimensions; Construct a four-dimensional spatio-temporal data structure, and integrate the spatial coordinates (x, y, z) and the time dimension (t) into a unified data representation model; Perform scale transformation on spatial data with different resolutions to achieve unified management of multi-scale spatial information; Perform interpolation processing on data with discontinuous time series to establish continuous and equally spaced time series data; Perform semantic-level fusion and standardization on multi-source data to ensure the consistency of data from different sources at the semantic level.

4. A GIS-based urban land planning analysis method according to claim 1, characterized in that, The adoption of the adaptive feature extraction mechanism to identify urban temporary land use areas includes: Construct a multi-scale feature extraction network to perform hierarchical feature extraction on urban area images; Fuse spectral features, texture features, morphological features, and context features to construct a multi-dimensional feature representation of temporary land use; Optimize the feature weight allocation based on the attention mechanism to highlight the key distinguishing features of temporary land use; Implement precise boundary extraction of temporary land use areas through the context-aware segmentation algorithm; Perform attribute annotation on the extracted temporary land use areas, including information such as area, shape, material, and use.

5. A GIS-based urban land planning analysis method according to claim 1, characterized in that, The construction of the hierarchical classification system for temporary land use includes: Establish a three-level classification framework, including a category layer, a function layer, and a use layer; For the category layer, classify temporary land use into temporary building land, temporary facility land, temporary activity land, and temporary storage land; For the functional layer, the temporary land is classified into production function, living function, service function, and public function; For the usage layer, it is further classified into emergency disaster relief, engineering construction, commercial exhibition, cultural activities, etc. according to specific application scenarios; Based on the weak supervision learning method, a classification model is trained with a small number of labeled samples to achieve automatic classification of large-scale temporary land.

6. The method for analyzing urban land planning based on GIS according to claim 1, characterized in that, The establishment of the change-sensitive time series model includes: According to the planning cycle and expected survival time of the temporary land, construct time window parameters; Design a change-sensitive loss function for the characteristics of temporary land to improve the detection sensitivity to the changes of small targets and sparsely distributed temporary buildings; Construct a bidirectional time series encoder to encode and analyze the historical change sequence of the temporary land; Introduce an urban planning rule base to constrain anomaly determination with domain knowledge and reduce the misjudgment rate; Adopt a multi-evidence fusion method based on the evidence theory to quantify the reliability of anomaly determination and generate a confidence score.

7. A GIS-based urban land planning analysis method according to claim 1, characterized in that The detection and identification of the temporary land area with abnormal change patterns include: Fit the time series change curve of the temporary land area to detect abnormal points deviating from the expected change pattern; Calculate the anomaly score based on the change rate and change amplitude, and mark the temporary land area exceeding the preset threshold as a potential anomaly; Combine the urban planning cycle and land use conversion rules to verify the authenticity of the potential anomaly area; Classify the identified abnormal temporary land areas to distinguish abnormal types such as those not demolished on schedule, function transformation, and structure strengthening; Generate a heat map of the spatial distribution of abnormal temporary land to visually display the agglomeration of abnormal areas.

8. A method for analyzing urban land planning based on GIS according to claim 1, characterized in that, The construction of the association graph model between the temporary land and the urban function network includes: Based on the spatial relationship between the temporary land and the surrounding permanent buildings, construct a heterogeneous spatial relationship graph; Calculate the association strength between the temporary land and the surrounding urban function nodes, including indicators such as spatial proximity, functional complementarity, and pedestrian flow interaction frequency; Adopt a multi-layer graph convolutional network to extract the structural and functional characteristics of the temporary land nodes; Based on the spectral clustering algorithm, discover temporary land communities with similar functional patterns and evolution trends; Analyze the location and importance of abnormal temporary land in the urban function network and evaluate its retention or transformation value.

9. The method for analyzing urban land planning based on GIS according to claim 1, wherein The generation of differentiated urban planning optimization suggestions includes: Construct a multi-objective evaluation index system including economic benefits, social value, and environmental impact; Based on the cellular automata and system dynamics models, simulate the urban evolution scenarios under different planning strategies; According to the characteristics of abnormal temporary land and urban function requirements, generate personalized planning suggestions, including demolition and reconstruction, function upgrading, structure transformation, etc. plans; Provide a visual analysis tool to visually display the expected effects and impact assessments of the planning suggestions; Design a collaborative decision-making mechanism involving multiple parties to integrate the feedback from planners, citizens, and developers and optimize the planning scheme.

10. An urban land planning analysis system based on GIS, characterized in that, It includes a processor and a memory; wherein, the memory stores a computer program, and when the computer program is executed by the processor, it realizes the GIS-based urban land planning analysis method as described in any one of claims 1 to 9.

Images