Method and device for analyzing factors affecting urban surface temperature, storage medium, and computer equipment

Abstract

The present application discloses a method and device for analyzing factors affecting urban surface temperature, a storage medium, and a computer device. The method comprises: performing surface temperature inversion calculation on Landsat‑8 remote sensing image data to determine surface temperature information; performing regional segmentation on a target research city according to airborne laser radar point cloud data and aerial image data, identifying the land use type of each area, and obtaining a land use classification distribution map; extracting a building height model for land use classified as a building and a canopy height model for land use classified as a forest, and assigning the building height model and the canopy height model to the land use classification distribution map; performing block division on the target research city according to aerial image data and OSM data, extracting POI features, two-dimensional morphological features, and three-dimensional morphological features of each block, and identifying the functional area type of each block; and for each functional area type, identifying the factors affecting surface temperature in the two-dimensional morphological features and the three-dimensional morphological features.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and device for analyzing factors affecting urban surface temperature, a storage medium, and a computer device. Background Art

[0002] With the development of global economy and society, urbanization has become an unstoppable development trend in the world and one of the most significant features of global change. The accelerated pace of urbanization has made human social activities such as production, life and practice increasingly frequent. A large number of artificial structures in cities, such as buildings, roads, and parking lots, have changed the thermal properties of the underlying surface. At the same time, the growth of urban population has led to the intensification of production, transportation, consumption and other activities, resulting in a large amount of heat and greenhouse gas emissions to the outside world, triggering a series of urban thermal environment problems. The most significant impact is the urban heat island effect. The urban heat island effect is a natural phenomenon in which the temperature in the city center is higher than that in the suburbs due to the loss of ecological environment regulation. The urban heat island effect can cause extreme weather such as high temperature and heavy rain, accelerate energy consumption, environmental pollution, affect people's health and reduce the happiness of life, and has attracted widespread attention from all walks of life. Land surface temperature (LST) is the most direct manifestation of the urban heat island effect. It regulates the temperature of the near-surface space and has a great impact on the living comfort of urban residents. Urban form describes the spatial composition and distribution characteristics of urban elements, and is the result of the combined effects of natural, social and economic factors on urban space. Existing research results show that urban morphology has a significant impact on the urban thermal environment. Therefore, identifying the spatial distribution of urban LST and optimizing the spatial layout of the city based on understanding the impact of urban morphology on LST are effective ways to alleviate thermal environment problems such as the urban heat island effect, which is of great significance to urban planning and management.

[0003] In traditional thermal environment research, temperature data is mainly obtained from meteorological station measurements. The limited and scattered meteorological station network can only provide partial non-uniform urban / suburban temperature changes with poor resolution, making it difficult to identify and obtain the overall temperature distribution of the city. In current research, there is still a lack of research on the impact of surface temperature. Summary of the invention

[0004] In view of this, the embodiments of the present application provide a method and device for analyzing factors affecting urban surface temperature, a storage medium, and a computer device, which help to more accurately analyze the impact of different dimensional morphological characteristics on the surface temperature.

[0005] According to one aspect of the present application, a method for analyzing factors affecting urban surface temperature is provided, the method comprising:

[0006] Obtain airborne LiDAR point cloud data, aerial image data, Landsat-8 remote sensing image data, POI (Point of Interest) data, and OSM (Open Street Map) data corresponding to the target research city;

[0007] Using the radiation transfer equation method, the surface temperature inversion calculation is performed on the Landsat-8 remote sensing image data to determine the surface temperature information of the target research city;

[0008] Performing regional segmentation on the target research city according to the airborne laser radar point cloud data and the aerial image data, and identifying the land use type of each area to obtain a land use classification distribution map of the target research city;

[0009] Based on the land use classification distribution map, extracting a building height model whose land use is classified as a building and a canopy height model whose land use is classified as a forest in the elevation model corresponding to the airborne laser radar point cloud data, and assigning the building height model and the canopy height model to the land use classification distribution map;

[0010] Divide the target research city into blocks according to the aerial image data and the OSM data, extract POI features of each block based on the POI data, extract at least one two-dimensional morphological feature and at least one three-dimensional morphological feature of each block based on the land use classification distribution map, and identify the functional area type corresponding to each block based on the POI features, the two-dimensional morphological features and the three-dimensional morphological features;

[0011] For each block corresponding to each functional area type, the surface temperature influencing factors in the two-dimensional morphological features and the three-dimensional morphological features are identified according to the surface temperature information, the two-dimensional morphological features and the three-dimensional morphological features.

[0012] According to another aspect of the present application, a device for analyzing factors affecting urban surface temperature is provided, the device comprising:

[0013] The data acquisition module is used to obtain airborne lidar point cloud data, aerial image data, Landsat-8 remote sensing image data, POI (Point of Interest) data, and OSM (OpenStreet Map) data corresponding to the target research city;

[0014] A surface temperature inversion module is used to perform surface temperature inversion calculation on the Landsat-8 remote sensing image data using a radiation transfer equation method to determine the surface temperature information of the target research city;

[0015] A type classification module is used to segment the target research city according to the airborne laser radar point cloud data and the aerial image data, and identify the land use type of each area to obtain a land use classification distribution map of the target research city;

[0016] A feature extraction module is used to extract, based on the land use classification distribution map, a building height model of a land use classified as a building and a canopy height model of a land use classified as a forest in the elevation model corresponding to the airborne laser radar point cloud data, and assign the building height model and the canopy height model to the land use classification distribution map;

[0017] A functional area classification module is used to divide the target research city into blocks according to the aerial image data and the OSM data, extract POI features of each block based on the POI data, extract at least one two-dimensional morphological feature and at least one three-dimensional morphological feature of each block based on the land use classification distribution map, and identify the functional area type corresponding to each block based on the POI features, the two-dimensional morphological features and the three-dimensional morphological features;

[0018] The influencing factor analysis module is used to identify the surface temperature influencing factors in the two-dimensional morphological features and the three-dimensional morphological features for the blocks corresponding to each functional area type according to the surface temperature information, the two-dimensional morphological features and the three-dimensional morphological features.

[0019] According to another aspect of the present application, a storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned method for analyzing factors affecting urban surface temperature is implemented.

[0020] According to another aspect of the present application, a computer device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor implements the above-mentioned method for analyzing factors affecting urban surface temperature when executing the program.

[0021] By means of the above technical scheme, a method and device for analyzing factors affecting urban surface temperature, a storage medium, and a computer device provided in the embodiment of the present application integrate airborne laser radar point cloud data, aerial image data, Landsat-8 remote sensing image data, POI data, and OSM data. The combination of multiple data sources can provide rich urban spatial information and provide comprehensive data support for the analysis of factors affecting urban surface temperature. The surface temperature inversion calculation of Landsat-8 remote sensing image data using the radiation transfer equation method can accurately determine the surface temperature information of the target research city, providing a reliable basis for subsequent analysis. Through airborne laser radar point cloud data and aerial image data, the regional segmentation and land use type identification of the target research city are realized, and a land use classification distribution map is formed. Further, based on POI data and OSM data, the blocks are divided, and the functional area type of each block is identified. This refined identification helps to more accurately analyze the impact of different functional areas on the surface temperature. In addition, in the elevation model corresponding to the airborne laser radar point cloud data, the building height model and the canopy height model are extracted and combined with the land use classification distribution map. In addition, the POI features, two-dimensional morphological features, and three-dimensional morphological features of each block were extracted. The complete extraction of these features provides a comprehensive analysis dimension for identifying the factors affecting the surface temperature. At the same time, for the blocks corresponding to different functional area types, the key factors affecting the surface temperature were identified based on the surface temperature information, two-dimensional morphological features, and three-dimensional morphological features. This targeted analysis helps to provide a scientific basis for urban planning, heat island effect mitigation, etc. Through this scheme, the main influencing factors of surface temperature in different functional areas can be identified, providing targeted decision-making support for urban planners and managers. For example, when planning a new urban area, the building layout and greening configuration can be optimized according to the analysis results to improve the urban thermal environment. In summary, the influencing factor analysis scheme of urban surface temperature provides strong support for urban thermal environment management and planning through multi-source data integration, accurate surface temperature inversion, refined regional and functional identification, complete feature extraction, and targeted identification of influencing factors.

[0022] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0024] Figure 1 A schematic diagram of a flow chart of a method for analyzing factors affecting urban surface temperature provided in an embodiment of the present application is shown;

[0025] Figure 2 A schematic diagram of a process for analyzing factors affecting urban surface temperature provided in another embodiment of the present application is shown;

[0026] Figure 3 A schematic diagram of the structure of a device for analyzing factors affecting urban surface temperature provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0027] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0028] In this embodiment, a method for analyzing factors affecting urban surface temperature is provided. Figure 1 As shown, the method includes:

[0029] Step 101, obtain airborne lidar point cloud data, aerial image data, Landsat-8 remote sensing image data, POI (Point of Interest) data, and OSM (Open Street Map) data corresponding to the target research city.

[0030] In the embodiment of the present application, for the target research city where the influencing factors of the surface temperature need to be studied, the multi-source data corresponding to the city is first obtained for subsequent analysis, and the multi-source data specifically includes airborne laser radar point cloud data, aerial image data, Landsat-8 remote sensing image data, POI data, and OSM data. These data sets cover information from multiple dimensions such as topography, surface coverage, and human activities, providing a comprehensive basis for subsequent analysis. Optionally, the airborne laser radar point cloud data respectively obtains the airborne laser radar point cloud data, aerial image data, Landsat-8 remote sensing image data in summer and winter, POI data, and OSM data corresponding to the target research city, so as to analyze the influencing factors of the urban surface temperature in summer and winter respectively for the target research city. After obtaining the data of summer and winter, the influencing factors of the surface temperature in summer and winter can be studied separately, and the Landsat-8 remote sensing image data in summer and winter should be calculated according to the surface inversion method proposed in the embodiment of the present application to realize the separate analysis of the influencing factors of the surface temperature in summer and winter. The specific data acquisition method may be:

[0031] (1) Airborne LiDAR point cloud data and aerial image data

[0032] The airborne laser radar point cloud and aerial image data were acquired synchronously by the airborne laser scanning system, and the data were acquired on September 9, 2017. The airborne laser scanning system is integrated with hardware such as RIEGL VUX-1LR laser radar scanner, PHASEONE IXU1000-R high-resolution digital camera and POS system. In the embodiment of the present application, the average point cloud density of the entire study area is 20pts / m2, the horizontal resolution is 0.17m, the vertical resolution is 0.2m, the minimum elevation of the point cloud is 338m, and the maximum elevation is 519.19m. In this embodiment, point cloud data is used for land use classification and extraction of two- and three-dimensional morphological features of the city. Each original aerial image is 11608pixel×8708pixel wide and has a resolution of 3.33mm. We use Pix4Dmapper to perform pre-processing such as orthorectification and stitching on the image, and set the resolution of the aerial orthophoto image finally used for research to 1m. Aerial images include information on three spectral bands of visible light: red, green, and blue, which are used in the embodiments of the present application for land use classification, extraction of two- and three-dimensional urban morphological features, and block division.

[0033] (2) Landsat-8 remote sensing image data

[0034] Landsat 8 remote sensing images are used for the inversion of LST. This paper obtains Landsat-8 remote sensing image data from the USGS official website (URL: https: / / earthexplorer.usgs.gov / ), including 9 OLI (Operational LandImager, OLI) bands and 2 TIRS (Thermal Infrared Sensor, TIRS) bands. For example, Landsat 8 data from the past three years corresponding to the acquisition time of airborne lidar point clouds and aerial images can be selected, that is, Landsat 8 remote sensing images from January 2017 to December 2019. Three scenes of the study area with no cloud coverage that meet the conditions are selected in summer (June to August) and winter (January, February, and December). Two images are included in winter, and only one image data that meets the requirements is selected in summer due to more cloudy weather.

[0035] (3) POI data

[0036] POI is a representative of the rapid development of Internet technology and the increasingly updated network electronic maps. It connects the daily life trajectory of human beings with the world of geographic information, has strong computing and expression capabilities, and has important reference value for the mining of geospatial data and the analysis of people in the region based on location. POI is a series of point type data in the Internet electronic map, mainly including attribute information such as name, address, coordinates (latitude and longitude), category, etc. In the embodiment of the present application, POI data is obtained from the Amap open platform (URL: https: / / lbs.amap.com / ), and the data interface provided by Amap is used to crawl the POI data of the study area based on Python. After data screening, a total of 36,216 vector point data of the study area were obtained. In the embodiment of the present application, POI data is used as one of the important data sources for the classification of urban functional areas.

[0037] (4) OSM data

[0038] OSM is a free, open source, editable map database created and maintained by volunteers around the world. OSM has global coverage and provides rich geographic data, including streets, buildings, rivers, lakes, mountains, public facilities, etc. OSM is widely used in applications such as urban planning, 3D modeling, and block boundary demarcation. For example, this application can be downloaded from the OSM official website

[0039] (URL: https: / / www.openstreetmap.org / ) to obtain the road network data of the study area, and select the main road network data of urban trunk roads, urban secondary roads, urban branch roads, urban elevated roads and expressways, railways, etc. as the basis for block division. In the embodiment of this application, OSM data is mainly used to obtain the road network distribution information of the study area and assist in the division of blocks in the study area.

[0040] Step 102, using the radiation transfer equation method, performs surface temperature inversion calculation on the Landsat-8 remote sensing image data to determine the surface temperature information of the target research city.

[0041] In the present application embodiment, Figure 2 As shown in the figure, the radiation transfer equation method is used to invert LST. The basic principle is: first, estimate the influence of the atmosphere on the surface thermal radiation; second, calculate the surface thermal radiation intensity by subtracting the influence of this part of the atmosphere from the total amount of thermal radiation observed by the satellite sensor; finally, convert the thermal radiation intensity into the corresponding LST. Specifically, the surface temperature calculation formula based on the radiation transfer equation method is:

[0042]

[0043] where λ is the wavelength of the emitted radiation (e.g. 10.9 μm for TIRS band 10), and ρ = 1.438 × 10 -2 mK, K 1 , K 2 are calibration constants (e.g., in TIRS band 10, K 1 =774.89 / m 2 *μm*sr), K 2 =1321.08K), L λ is the thermal infrared radiation brightness value received by the satellite sensor, ε is the surface emissivity, B(T s ) is the black body thermal radiation brightness; T s is the actual surface temperature (in thermodynamic temperature K); (T s ) is the black body thermal radiation brightness; τ is the transmittance of the atmosphere in the thermal infrared band; L ↑ L is the upward radiant brightness of the atmosphere. The actual radiant brightness of the ground is the energy that reaches the satellite sensor after passing through the atmosphere. ↓ It is the energy reflected by the atmosphere after it radiates downward and reaches the ground. ↑ , L ↓ You can query and obtain the information on the relevant website provided by NASA (URL: http: / / atmcorr.gsfc.nasa.gov / ) based on the imaging time and central longitude and latitude of the Landsat8 image.

[0044] Step 103, segmenting the target research city into regions according to the airborne laser radar point cloud data and the aerial image data, and identifying the land use type of each region to obtain a land use classification distribution map of the target research city.

[0045] In the embodiment of the present application, the laser radar and aerial image data are combined to perform regional segmentation, and then the land use type of each area (such as residential area, commercial area, park, etc.) is identified to form a land use classification map. This helps to understand the possible impact of different use areas on the surface temperature.

[0046] Optionally, in step 103, identifying the land use type of each area to obtain a land use classification distribution map of the target research city includes:

[0047] Extracting elevation model information, intensity model information, roughness model information, and texture feature information based on the elevation model based on the airborne laser radar point cloud data, and extracting RGB band information and visible light vegetation index information based on the aerial image data;

[0048] Selecting a plurality of sample areas in the region, marking land type labels of the sample areas, and constructing a first training sample and a first verification sample based on the sample areas and the corresponding land type labels;

[0049] Respectively taking each of the elevation model information, the intensity model information, the roughness model information, the texture feature information, the RGB band information, and the visible light vegetation index information as a regional classification feature, and constructing multiple first feature fusion schemes, wherein each first feature fusion scheme includes at least one regional classification feature;

[0050] For each first feature fusion scheme, a land use classification model is constructed according to the regional classification features included in the first feature fusion scheme, and the land use classification model is trained based on the first training samples, and the trained land use classification model is verified based on the first verification samples;

[0051] The target first feature fusion scheme is determined according to the verification results corresponding to each first feature fusion scheme, and the land use classification model corresponding to the target first feature fusion scheme is used to identify the land use type of each area to obtain the land use classification distribution map of the target research city.

[0052] In this embodiment, first, the elevation, intensity, roughness and texture features are extracted from the airborne laser radar point cloud data, which reflect the three-dimensional structure and physical properties of the surface. RGB color information and visible light vegetation index (such as NDVI) are extracted from the aerial image data, which help to distinguish vegetation coverage and ground material. These data together constitute a rich regional feature set. Secondly, representative sample areas are selected in the study area, and their land use types (such as building land, bare soil, farmland, grassland, road, woodland, water area) are manually marked to create training and verification sample sets. This step ensures the accuracy of subsequent model training. Again, in order to find the most effective feature combination to distinguish different land use types, the above features can be tried to be fused in various combinations (for example, only using elevation and vegetation index, or combining all features, etc.) to form different feature fusion schemes. Then, for each feature fusion scheme, a machine learning model (such as KNN, RF and XGBoost, etc.) is constructed and trained using the training sample set established in the first step. Then, the model performance is evaluated using the verification sample set, such as by indicators such as accuracy and recall. Finally, the best performing feature fusion scheme is selected based on the verification results, which means that the model under this scheme can most accurately identify the land use type. Further, this optimal model is used to classify the entire study area to obtain a land use classification distribution map. The embodiment of the present application improves the accuracy and precision of land use classification through automatic identification by machine learning, combined with multi-source data and feature fusion strategies, and can more accurately reflect the urban spatial structure than traditional methods. The application of automated processing and machine learning algorithms makes it possible to process large-area urban data, greatly improving work efficiency and processing capabilities.

[0053] Step 104, based on the land use classification distribution map, extract the building height model of the land use classified as buildings and the canopy height model of the land use classified as woodland in the elevation model corresponding to the airborne lidar point cloud data, and assign the building height model and the canopy height model to the land use classification distribution map.

[0054] In the embodiment of the present application, after determining the land use classification distribution map, the building height model of the building and the canopy height model of the forest land can be extracted from the elevation model according to the land use type of each area in the map, and combined with the land use classification distribution map, the three-dimensional characteristics of the urban structure are further refined. This helps to analyze the specific contribution of building density, green coverage rate, etc. to the urban heat island effect. Specifically, in this embodiment, buildings and trees are used as the main objects for three-dimensional morphological feature extraction, and the three-dimensional elevation models of buildings and trees are extracted in combination with the two-dimensional land object distribution map of land use classification and the elevation information based on airborne lidar point cloud data. First, according to the results of land use classification, the two-dimensional distribution of ground objects is determined, where buildings correspond to building land in the ground object category, and trees correspond to forest land in the ground object category. The two-dimensional distribution ranges of buildings and trees can be extracted according to the class attribute information of building land and forest land. Secondly, the buildings and trees are post-processed to obtain a two-dimensional distribution map with a relatively regular distribution. The more fragmented objects in the two-dimensional distribution map are removed by morphological opening operation, and the morphological closing operation is used to fill smaller holes. The output object can be made smoother by eight-neighborhood majority filtering operation. Next, the distribution map of buildings and trees obtained in the previous step is subjected to a pickling operation with nDSM to extract the building height model and canopy height model corresponding to the range of buildings and trees, respectively.

[0055] Step 105, dividing the target research city into blocks according to the aerial image data and the OSM data, extracting POI features of each block based on the POI data, extracting at least one two-dimensional morphological feature and at least one three-dimensional morphological feature of each block based on the land use classification distribution map, and identifying the functional area type corresponding to each block based on the POI features, the two-dimensional morphological features and the three-dimensional morphological features.

[0056] In the embodiment of the present application, aerial images and OSM data are combined to divide the city into blocks, and the POI features of each block are extracted based on the POI data, so as to identify the functional area types (such as commercial areas, residential areas, etc.) of different blocks based on human activity characteristics (POI) and urban morphological characteristics (two-dimensional and three-dimensional). This helps to understand how social and economic activities affect the distribution of surface temperature.

[0057] Optionally, in step 105, POI features of each block are extracted based on the POI data, including: reclassifying different types of POI data according to the definition of functional area type, so that the POI data is divided into blocks of corresponding functional area types; based on the kernel density analysis method, the divided POI data is rasterized, and the kernel density mean, kernel density standard deviation and kernel density sum within each block are calculated based on the rasterized POI data, wherein the POI features include the kernel density mean, the kernel density standard deviation and the kernel density sum.

[0058] And, optionally, in step 105, based on the POI feature, the two-dimensional morphological feature and the three-dimensional morphological feature, identifying the functional area type corresponding to each block includes:

[0059] Selecting a plurality of sample blocks from the blocks, marking the functional area type labels of the sample blocks, and constructing a second training sample and a second verification sample based on the sample blocks and the corresponding functional area type labels;

[0060] Each two-dimensional morphological feature, each three-dimensional morphological feature and each POI feature is respectively used as a functional area classification feature to construct multiple second feature fusion schemes, wherein each second feature fusion scheme includes at least one functional area classification feature;

[0061] For each second feature fusion scheme, a functional area classification model is constructed according to the functional area classification features included in the second feature fusion scheme, and the functional area classification model is trained based on the second training samples, and the trained functional area classification model is verified based on the second verification samples;

[0062] A target second feature fusion scheme is determined according to the verification results corresponding to each second feature fusion scheme, and the functional area classification model corresponding to the target second feature fusion scheme is used to identify the functional area type of each block.

[0063] In the above embodiment, the urban functional area divides the city into spatially separated areas with different attributes according to relevant standards or functional types. The areas are both independent and interconnected. It represents a complex and diverse land use object with similar spatial landscape structure and social and economic activities. Different functional areas often have different spatial landscape structures and social and economic activities. For example, the study area mainly includes residential houses, factories, hotels, shopping centers, urban villages and rural settlements. Combined with airborne aerial image data and high-precision Google maps, the study area is divided into built-up areas and non-built-up areas according to the social and economic functional characteristics of each block. The built-up area includes residential areas, commercial areas, industrial areas, public services and open spaces; the non-built-up area includes agricultural areas, green spaces, waters and unused areas. The reasonable division of blocks is the basic premise for functional area classification. The blocks are further divided in combination with OSM road network data and airborne aerial images. Finally, the study area is divided into multiple blocks so that block-scale feature extraction and functional area classification research can be carried out based on the scope of each block. In the embodiment of the present application, the two-dimensional and three-dimensional morphological features of each functional area are extracted to describe its landscape structure, and the POI data is used to describe the basic activities of human beings, and the functional area classification is carried out by integrating the two-dimensional and three-dimensional morphological features of the city and the POI features. The classification scheme mainly includes the following four steps: 1) Feature extraction, extracting the two-dimensional and three-dimensional morphological features and POI features of the city at the block scale. The POI data is reclassified according to the definition of the functional area category, and the discrete POI point data is converted into raster data that can be spatially analyzed and expressed using the kernel density analysis method, and finally the mean, standard deviation and sum of the kernel density of residential areas, commercial areas, industrial areas, public services and open spaces within each block are extracted. 2) Data fusion, according to the characteristics of different input data sources (i.e., two-dimensional and three-dimensional morphological features, and POI features), a total of six data fusion classification schemes (including two-dimensional morphological features, three-dimensional morphological features, POI features, two-dimensional morphological features + POI features, three-dimensional morphological features + POI features, two-dimensional morphological features + three-dimensional morphological features + POI features) are designed to obtain the optimal classification scheme. 3) Sample data selection: Combine aerial images and geographic data to complete the selection of sample data sets for each functional area, including the determination of training samples and test sample data sets based on the block scale. The samples include sample blocks and corresponding functional area type labels. 4) Use three classifier algorithms, KNN, RF and XGBoost, to conduct classification experiments with different schemes, and perform feature optimization on all input features to obtain the optimized classification scheme. Finally, the classification scheme with the highest classification accuracy is selected for functional area mapping output.

[0064] In addition, in an embodiment of the present application, optionally, after assigning the building height model and the canopy height model to the land use classification distribution map, the method further includes: dividing the land use classification distribution map into unit grids; for each grid, identifying the surface temperature influencing factors in the two-dimensional morphological features and the three-dimensional morphological features according to the surface temperature information, the two-dimensional morphological features and the three-dimensional morphological features.

[0065] Among them, according to the inversion results of LST, the LST distribution data of a certain resolution can be obtained. For example, the Landsat 8 remote sensing image corresponds to a 30m resolution. Therefore, the study area can be further divided into a grid of 30m×30m, and the two-dimensional and three-dimensional morphological features based on the grid-level (grid scale) are extracted according to the classification results of the ground object (land use classification), that is, a pixel in the LST corresponds to a grid in the high-resolution image, so as to associate the two-dimensional and three-dimensional morphological features of the grid-level city with the LST and study their corresponding relationship; secondly, according to the functional area classification results, the two-dimensional and three-dimensional morphological features of the objects in each functional area are extracted, and the average value of LST in each functional area is counted as the LST data of the functional area, and the influence of the two-dimensional and three-dimensional morphological features of the block-level city on LST is further explored; by comparing the influence of two-dimensional and three-dimensional morphological features on LST at different scales in summer and winter, the mutual influence between them and the seasonal and hierarchical differences are studied. The specific two-dimensional and three-dimensional morphological feature selection and extraction methods are as follows:

[0066] (1) Two-dimensional morphological characteristics

[0067] Based on the high-precision land use classification results, the distribution of seven types of land objects was obtained, and on this basis, the two-dimensional morphological feature extraction of the seven types of land objects was carried out. From the perspective of landscape pattern, the structural composition and spatial configuration of land objects in the landscape are extracted. The landscape pattern index is an effective method for quantitatively studying the composition, spatial configuration and dynamic changes of the landscape by establishing the connection between the pattern and the landscape process. Fragstats4.2 is a commonly used landscape index calculation software. From the patch, type and landscape level, hundreds of landscape indexes can be calculated, including area edge index, shape index, core area index, contrast index, aggregation / dispersion index and diversity index. In addition, according to the research results of predecessors, it is found that the landscape index of the patch type level and the landscape level can better reflect the composition, spatial configuration and degree of fragmentation of the landscape pattern. Therefore, the embodiment of the present application is based on the patch type and landscape level, and chooses to use Fragstats4.2 to quantitatively describe the distribution and mutual relationship of the land use types in the study area on a two-dimensional plane. According to previous studies, ten commonly used landscape indices were selected for the extraction of two-dimensional morphological features, including area indicators: type coverage (Percentage of Landscape, PLAND) and edge density (Edge Density, ED); shape indicators: area-weighted Mean Shape Index (SHAPE_AM) and area-weighted Mean Fractal Dimension Index (FRAC_AM); aggregation / dispersion indicators: patch density (PD), landscape shape index (LSI), mean proximity index (PROX_MN), Euclidean Nearest-Neighbor Mean Distance (ENN_MN), patch cohesion index (COHESION) and Shannon's Diversity Index (SHDI).

[0068] (2) Three-dimensional morphological characteristics

[0069] In this embodiment, the three-dimensional morphological characteristics of the city mainly refer to the spatial distribution, changes and mutual relationships of the three-dimensional information of buildings and trees. Based on the two-dimensional distribution of three-dimensional objects and the elevation model characteristics, a total of 16 three-dimensional building morphological characteristics and 12 three-dimensional tree morphological characteristics are obtained, including three-dimensional space elevation indicators, area indicators, shape indicators and spatial distribution indicators.

[0070] Based on the two-dimensional distribution of buildings in the study area and the elevation information in the building elevation model, 16 different three-dimensional building morphological features are mainly extracted, including building mean height (BMH), building maximum height (BMaH), building height variation (BHV), normalized building height variance (NBHV), building height range (BHR), building surface area (BSA), building volume (BV), street height-to-width ratio H / W (Ratio of Street Height and Building Width, BHW), street height-to-length ratio H / L (Ratio of Street Height and Length, BHL), building surface area coverage (PBSA), building volume ratio (PBV), building edge density (BED), building shape coefficient (BSC), building landscape shape index in 3D Space, BLSI), Frontal Area Index (FAI), and Building Sky View Factor (BSVF).

[0071] The extraction of three-dimensional tree morphological parameters is similar to that of buildings. Here, based on the two-dimensional distribution of trees and the elevation information provided by the canopy elevation model, we mainly extract 12 three-dimensional tree morphological characteristics, including tree mean height (TMH), tree canopy maximum height (TMaH), tree height variation (THV), normalized tree height variation (Normalized TreeHeight Variance, NTHV), tree height range (THR), tree surface area (TSA), tree volume (TV), tree surface area coverage (PTSA), tree volume percentage (PTV), tree edge density (TED), tree shape coefficient (TSC), and tree landscape shape index (TLSI) in 3D space.

[0072] Table 1 details the description of various types of 2D and 3D morphological features.

[0073] Table 1

[0074]

[0075]

[0076]

[0077] Step 106, for each block corresponding to each functional area type, based on the surface temperature information, the two-dimensional morphological features and the three-dimensional morphological features, identify the surface temperature influencing factors in the two-dimensional morphological features and the three-dimensional morphological features.

[0078] In the embodiment of the present application, after determining the functional area type to which each block belongs, the surface temperature influencing factors can be analyzed for each block of different functional area types, and the two-dimensional morphological features and three-dimensional morphological features that have an impact on the surface temperature of the blocks of different functional area types can be identified using the block as the scale. In addition, it can also include: dividing the land use classification distribution map into unit grids; for each grid, based on the surface temperature information, the two-dimensional morphological features and the three-dimensional morphological features, identifying the surface temperature influencing factors in the two-dimensional morphological features and the three-dimensional morphological features. That is, using the grid as the scale, the two-dimensional morphological features and the three-dimensional morphological features that have an impact on the surface temperature are identified.

[0079] In the embodiment of the present application, taking the analysis at the block scale as an example, optionally, step 106 includes: for the blocks corresponding to each functional area type, according to the surface temperature information, performing a single factor analysis on the two-dimensional morphological features and the three-dimensional morphological features, and determining the factors whose correlation between the surface temperature of the blocks corresponding to the functional area type meets the preset correlation conditions as analysis factors for regression analysis; according to the analysis factors, constructing multiple factor analysis schemes, wherein the factor analysis schemes include a two-dimensional morphological feature analysis scheme, a three-dimensional building morphological analysis scheme, a three-dimensional tree morphological analysis scheme, a two-dimensional morphological feature and a three-dimensional building morphological feature analysis scheme, a two-dimensional morphological feature and a three-dimensional tree morphological feature analysis scheme, and a two-dimensional morphological feature and a three-dimensional building and tree morphological feature analysis scheme; constructing a multivariate linear regression model based on each factor analysis scheme respectively, and performing a regression analysis between each feature in the factor analysis scheme and the surface temperature for the blocks corresponding to each functional area type through the multivariate linear regression model, determining the surface temperature influencing factors corresponding to each factor analysis scheme, and determining the surface temperature influencing factors in all analysis factors corresponding to each functional area type according to the surface temperature influencing factors in each factor analysis scheme corresponding to each functional area type.

[0080] In this embodiment, for the blocks corresponding to each functional area type, the surface temperature information is first used to perform a single factor analysis on the two-dimensional morphological features and the three-dimensional morphological features. Determine which factors (such as the aspect ratio of the block, the average height of the building, the green coverage rate, etc.) are significantly correlated with the surface temperature of the corresponding block. Statistical methods (such as correlation coefficient, t-test, etc.) are used to quantify this correlation, and factors that meet the preset correlation conditions are screened out as the basis for subsequent regression analysis. Through single factor analysis, the key factors closely related to the surface temperature can be quickly located, providing direction for subsequent multi-factor analysis.

[0081] Furthermore, based on the selected analysis factors, a variety of factor analysis schemes with different combinations are constructed. These schemes include only considering two-dimensional features, only considering three-dimensional building features, only considering three-dimensional tree features, and combinations of these features. Through different combinations of schemes, the comprehensive impact of various factors on the surface temperature is comprehensively examined. The construction of multiple schemes helps to comprehensively analyze the influencing factors of the surface temperature and ensure the accuracy and comprehensiveness of the analysis results. Specifically, based on each factor analysis scheme, a multivariate linear regression model is constructed, and the model is used for regression analysis. Through regression analysis, the quantitative relationship between each feature in each factor analysis scheme and the surface temperature is determined, and the main influencing factors of the surface temperature are identified. Using the multivariate linear regression model, the coefficient corresponding to each factor can be calculated to quantify their influence on the surface temperature. Multivariate linear regression analysis can provide accurate quantitative results, which helps decision makers understand the specific influence of different factors on the surface temperature and provide a scientific basis for urban planning, environmental protection, etc. Among them, the distribution of urban features is complex, and the features affect each other. The analysis of a single feature and LST cannot represent the relationship between the features of the entire study area and LST. Therefore, the influence of urban two-dimensional and three-dimensional morphological characteristics on LST is a complex process. In order to further compare and analyze the influence and mutual relationship of urban two-dimensional and three-dimensional morphological characteristics on LST and reduce the mutual influence between characteristics, the method of multivariate linear stepwise regression analysis can be used. This method introduces variables into the regression model one by one, and performs F test each time an explanatory variable is introduced. At the same time, t test is performed on the selected explanatory variables one by one. When the previously introduced explanatory variable becomes no longer significant due to the introduction of the new explanatory variable, the previous variable will be deleted, so as to ensure that only significant variables are included in the regression equation each time a new variable is introduced. This is a more efficient and stable method. By constructing multiple multivariate linear stepwise regression models, the explanatory ability of regression models constructed with different morphological characteristics on LST is comprehensively analyzed and compared. In addition, considering the impact of human activities on LST, the regression model of urban two-dimensional and three-dimensional morphological characteristics and LST in summer and winter is constructed at the Block-level to further study the impact of human activities on LST and the stability of the analysis model. According to the characteristics of the five functional areas of the built-up area, including residential area, commercial area, industrial area, public service and open space, the influence of two-dimensional and three-dimensional morphological characteristics on LST in different functional areas is analyzed and compared. Finally, the factors affecting the surface temperature in all the analysis factors corresponding to each functional area type can be comprehensively determined based on the factors affecting the surface temperature in the various factor analysis schemes corresponding to each functional area type. The analysis results of different schemes are integrated to form a comprehensive understanding of the factors affecting the surface temperature of each functional area type. The comprehensive analysis results can provide targeted suggestions for the planning and management of different functional areas, which will help to achieve the optimization and sustainable development of the urban thermal environment.

[0082] By applying the technical solution of this embodiment, a study on the influence of fine two-dimensional and three-dimensional morphological features of cities of different scales on LST is carried out based on high-resolution remote sensing data. First, based on airborne laser radar point cloud, aerial image data and open geographic information data sources (POI data and OSM road network data), high-precision ground object distribution and functional area distribution are obtained. Secondly, fine two-dimensional and three-dimensional morphological features and their corresponding LST are extracted from the Grid-level and Block-level scales. Finally, the relationship between the two-dimensional and three-dimensional morphological features of the city and the LST in summer and winter is studied at the Grid-level and Block-level scales using statistical analysis methods. The embodiment of this application integrates airborne laser radar point cloud data, aerial image data, Landsat-8 remote sensing image data, POI data and OSM data. The combination of multiple data sources can provide rich urban spatial information and provide comprehensive data support for the analysis of factors affecting urban surface temperature. The surface temperature inversion calculation of Landsat-8 remote sensing image data using the radiation transfer equation method can more accurately determine the surface temperature information of the target research city, providing a reliable basis for subsequent analysis. Through airborne LiDAR point cloud data and aerial image data, the regional segmentation and land use type identification of the target research city were realized, and a land use classification distribution map was formed. Furthermore, based on POI data and OSM data, the blocks were divided and the functional area type of each block was identified. This refined identification helps to more accurately analyze the impact of different functional areas on the surface temperature. In addition, the building height model and canopy height model were extracted from the elevation model corresponding to the airborne LiDAR point cloud data and combined with the land use classification distribution map. In addition, the POI features, two-dimensional morphological features and three-dimensional morphological features of each block were extracted. The complete extraction of these features provides a comprehensive analysis dimension for identifying the factors affecting the surface temperature. At the same time, for the blocks corresponding to different functional area types, the key factors affecting the surface temperature were identified based on the surface temperature information, two-dimensional morphological features and three-dimensional morphological features. This targeted analysis helps to provide a scientific basis for urban planning, heat island effect mitigation, etc. Through this solution, the main influencing factors of surface temperature in different functional areas can be identified, providing targeted decision support for urban planners and managers. For example, when planning a new urban area, the building layout and greening configuration can be optimized according to the analysis results to improve the urban thermal environment. In summary, the influencing factor analysis scheme of urban surface temperature provides strong support for urban thermal environment management and planning through multi-source data integration, accurate surface temperature inversion, refined regional and functional identification, complete feature extraction, and targeted influencing factor identification.

[0083] Further, as Figure 1The specific implementation of the method, the embodiment of the present application provides a device for analyzing factors affecting urban surface temperature, such as Figure 3 As shown, the device comprises:

[0084] The data acquisition module is used to obtain airborne lidar point cloud data, aerial image data, Landsat-8 remote sensing image data, POI (Point of Interest) data, and OSM (OpenStreet Map) data corresponding to the target research city;

[0085] A surface temperature inversion module is used to perform surface temperature inversion calculation on the Landsat-8 remote sensing image data using a radiation transfer equation method to determine the surface temperature information of the target research city;

[0086] A type classification module is used to segment the target research city according to the airborne laser radar point cloud data and the aerial image data, and identify the land use type of each area to obtain a land use classification distribution map of the target research city;

[0087] A feature extraction module is used to extract, based on the land use classification distribution map, a building height model of a land use classified as a building and a canopy height model of a land use classified as a forest in the elevation model corresponding to the airborne laser radar point cloud data, and assign the building height model and the canopy height model to the land use classification distribution map;

[0088] A functional area classification module is used to divide the target research city into blocks according to the aerial image data and the OSM data, extract POI features of each block based on the POI data, extract at least one two-dimensional morphological feature and at least one three-dimensional morphological feature of each block based on the land use classification distribution map, and identify the functional area type corresponding to each block based on the POI features, the two-dimensional morphological features and the three-dimensional morphological features;

[0089] The influencing factor analysis module is used to identify the surface temperature influencing factors in the two-dimensional morphological features and the three-dimensional morphological features for the blocks corresponding to each functional area type according to the surface temperature information, the two-dimensional morphological features and the three-dimensional morphological features.

[0090] Optionally, the type classification module is also used to:

[0091] Extracting elevation model information, intensity model information, roughness model information, and texture feature information based on the elevation model based on the airborne laser radar point cloud data, and extracting RGB band information and visible light vegetation index information based on the aerial image data;

[0092] Selecting a plurality of sample areas in the region, marking land type labels of the sample areas, and constructing a first training sample and a first verification sample based on the sample areas and the corresponding land type labels;

[0093] Respectively taking each of the elevation model information, the intensity model information, the roughness model information, the texture feature information, the RGB band information, and the visible light vegetation index information as a regional classification feature, and constructing multiple first feature fusion schemes, wherein each first feature fusion scheme includes at least one regional classification feature;

[0094] For each first feature fusion scheme, a land use classification model is constructed according to the regional classification features included in the first feature fusion scheme, and the land use classification model is trained based on the first training samples, and the trained land use classification model is verified based on the first verification samples;

[0095] The target first feature fusion scheme is determined according to the verification results corresponding to each first feature fusion scheme, and the land use classification model corresponding to the target first feature fusion scheme is used to identify the land use type of each area to obtain the land use classification distribution map of the target research city.

[0096] Optionally, the functional area classification module is further used to:

[0097] Reclassifying different types of POI data according to the definition of functional area types, so that the POI data are divided into blocks of corresponding functional area types;

[0098] Based on the kernel density analysis method, the divided POI data are rasterized, and the kernel density mean, kernel density standard deviation and kernel density sum are calculated within each block based on the rasterized POI data, wherein the POI features include the kernel density mean, the kernel density standard deviation and the kernel density sum.

[0099] Optionally, the functional area classification module is further used to:

[0100] Selecting a plurality of sample blocks from the blocks, marking the functional area type labels of the sample blocks, and constructing a second training sample and a second verification sample based on the sample blocks and the corresponding functional area type labels;

[0101] Each two-dimensional morphological feature, each three-dimensional morphological feature and each POI feature is respectively used as a functional area classification feature to construct multiple second feature fusion schemes, wherein each second feature fusion scheme includes at least one functional area classification feature;

[0102] For each second feature fusion scheme, a functional area classification model is constructed according to the functional area classification features included in the second feature fusion scheme, and the functional area classification model is trained based on the second training samples, and the trained functional area classification model is verified based on the second verification samples;

[0103] A target second feature fusion scheme is determined according to the verification results corresponding to each second feature fusion scheme, and the functional area classification model corresponding to the target second feature fusion scheme is used to identify the functional area type of each block.

[0104] Optionally, the two-dimensional morphological features include type coverage, edge density, area-weighted average shape index, area-weighted average fractal dimension index, patch density, landscape shape index, proximity index mean, nearest distance mean, patch cohesion index and Shannon diversity index; the three-dimensional morphological features include three-dimensional building morphological features and three-dimensional tree morphological features, the three-dimensional building morphological features include average building height, maximum building height, building height variation, standardized building height variation, building height range, building surface area, building volume, street aspect ratio H / W, street height-to-length ratio H / L, building surface area coverage, building volume proportion, building edge density, building shape coefficient, building landscape shape index, windward area index, sky view coefficient, the three-dimensional tree morphological features include average tree height, maximum tree height, tree height variation, standardized tree height variation, tree height range, tree surface area, tree volume, tree surface area coverage, tree volume proportion, tree edge density, tree shape coefficient, tree landscape shape index;

[0105] The influencing factor analysis module is also used for:

[0106] For each block corresponding to each functional area type, a single factor analysis is performed on the two-dimensional morphological features and the three-dimensional morphological features according to the surface temperature information, and factors whose correlations with the surface temperatures of the blocks corresponding to the functional area type meet preset correlation conditions are determined as analysis factors for regression analysis;

[0107] According to the analysis factors, a plurality of factor analysis schemes are constructed, wherein the factor analysis schemes include a two-dimensional morphological feature analysis scheme, a three-dimensional building morphological feature analysis scheme, a three-dimensional tree morphological feature analysis scheme, a two-dimensional morphological feature and three-dimensional building morphological feature analysis scheme, a two-dimensional morphological feature and three-dimensional tree morphological feature analysis scheme, and a two-dimensional morphological feature and three-dimensional building and tree morphological feature analysis scheme;

[0108] Based on each factor analysis scheme, a multivariate linear regression model is constructed respectively. For the blocks corresponding to each functional area type, regression analysis is performed between each feature in the factor analysis scheme and the surface temperature through the multivariate linear regression model to determine the surface temperature influencing factors corresponding to each factor analysis scheme. The surface temperature influencing factors in all analysis factors corresponding to each functional area type are determined based on the surface temperature influencing factors in each factor analysis scheme corresponding to each functional area type.

[0109] Optionally, the influencing factor analysis module is further used to:

[0110] Dividing the land use classification distribution map into unit grids;

[0111] For each grid, the ground surface temperature influencing factors in the two-dimensional morphological features and the three-dimensional morphological features are identified according to the ground surface temperature information, the two-dimensional morphological features and the three-dimensional morphological features.

[0112] Optionally, the data acquisition module is further used to:

[0113] The airborne lidar point cloud data, aerial image data, Landsat-8 remote sensing image data in summer and winter, POI data, and OSM data corresponding to the target research city are obtained respectively to analyze the influencing factors of urban surface temperature in summer and winter of the target research city.

[0114] It should be noted that for other corresponding descriptions of the functional units involved in the device for analyzing factors affecting urban surface temperature provided in the embodiment of the present application, reference can be made to Figure 1 to Figure 2 The corresponding description in the method will not be repeated here.

[0115] The embodiment of the present application also provides a computer device, which can be a personal computer, a server, a network device, etc. The computer device includes a bus, a processor, a memory and a communication interface, and can also include an input and output interface and a display device. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store location information. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the steps in each method embodiment are implemented.

[0116] Those skilled in the art will appreciate that the structure of the above-mentioned computer device is only a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components, or combine certain components, or have a different arrangement of components.

[0117] In one embodiment, a computer-readable storage medium is provided. The computer-readable storage medium may be non-volatile or volatile, and stores a computer program thereon. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0118] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0119] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0120] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0121] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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 this specification.

[0122] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for analyzing factors affecting urban surface temperature, characterized in that: The method comprises: Obtain airborne LiDAR point cloud data, aerial image data, Landsat-8 remote sensing image data, POI (Point of Interest) data, and OSM (Open Street Map) data corresponding to the target research city; Using the radiation transfer equation method, the surface temperature inversion calculation is performed on the Landsat-8 remote sensing image data to determine the surface temperature information of the target research city; Performing regional segmentation on the target research city according to the airborne laser radar point cloud data and the aerial image data, and identifying the land use type of each area to obtain a land use classification distribution map of the target research city; Based on the land use classification distribution map, extracting a building height model whose land use is classified as a building and a canopy height model whose land use is classified as a forest in the elevation model corresponding to the airborne laser radar point cloud data, and assigning the building height model and the canopy height model to the land use classification distribution map; Divide the target research city into blocks according to the aerial image data and the OSM data, extract POI features of each block based on the POI data, extract at least one two-dimensional morphological feature and at least one three-dimensional morphological feature of each block based on the land use classification distribution map, and identify the functional area type corresponding to each block based on the POI features, the two-dimensional morphological features and the three-dimensional morphological features; For each block corresponding to each functional area type, according to the surface temperature information, the two-dimensional morphological characteristics and the three-dimensional morphological characteristics, the surface temperature influencing factors in the two-dimensional morphological characteristics and the three-dimensional morphological characteristics are identified, including: for each block corresponding to the functional area type, according to the surface temperature information, the two-dimensional morphological characteristics and the three-dimensional morphological characteristics are single-factor analysis is performed to determine the factors whose correlation between the surface temperature of the blocks corresponding to the functional area type meets the preset correlation conditions as the analysis factors of the regression analysis; according to the analysis factors, a plurality of factor analysis schemes are constructed, wherein the factor analysis schemes include a two-dimensional morphological feature analysis scheme and a three-dimensional building morphology analysis scheme. , three-dimensional tree morphology analysis scheme, two-dimensional morphology characteristics and three-dimensional building morphology characteristics analysis scheme, two-dimensional morphology characteristics and three-dimensional tree morphology characteristics analysis scheme, two-dimensional morphology characteristics and three-dimensional building and tree morphology characteristics analysis scheme; based on each factor analysis scheme, a multivariate linear regression model is constructed respectively, and through the multivariate linear regression model, a regression analysis is performed between each feature in the factor analysis scheme and the surface temperature for the blocks corresponding to each functional area type, and the surface temperature influencing factors corresponding to each factor analysis scheme are determined; and according to the surface temperature influencing factors in the factor analysis schemes corresponding to each functional area type, the surface temperature influencing factors in all analysis factors corresponding to each functional area type are determined.

2. The method according to claim 1, characterized in that Identify the land use type of each area and obtain the land use classification distribution map of the target research city, including: Extracting elevation model information, intensity model information, roughness model information, and texture feature information based on the elevation model based on the airborne laser radar point cloud data, and extracting RGB band information and visible light vegetation index information based on the aerial image data; Selecting a plurality of sample areas in the region, marking land type labels of the sample areas, and constructing a first training sample and a first verification sample based on the sample areas and the corresponding land type labels; Respectively taking each of the elevation model information, the intensity model information, the roughness model information, the texture feature information, the RGB band information, and the visible light vegetation index information as a regional classification feature, and constructing multiple first feature fusion schemes, wherein each first feature fusion scheme includes at least one regional classification feature; For each first feature fusion scheme, a land use classification model is constructed according to the regional classification features included in the first feature fusion scheme, and the land use classification model is trained based on the first training samples, and the trained land use classification model is verified based on the first verification samples; The target first feature fusion scheme is determined according to the verification results corresponding to each first feature fusion scheme, and the land use classification model corresponding to the target first feature fusion scheme is used to identify the land use type of each area to obtain the land use classification distribution map of the target research city.

3. The method according to claim 1, characterized in that Extracting POI features of each block based on the POI data includes: Reclassifying different types of POI data according to the definition of functional area types, so that the POI data are divided into blocks of corresponding functional area types; Based on the kernel density analysis method, the divided POI data are rasterized, and the kernel density mean, kernel density standard deviation and kernel density sum are calculated within each block based on the rasterized POI data, wherein the POI features include the kernel density mean, the kernel density standard deviation and the kernel density sum.

4. The method according to claim 3, characterized in that Based on the POI feature, the two-dimensional morphological feature and the three-dimensional morphological feature, identifying the functional area type corresponding to each block, including: Selecting a plurality of sample blocks from the blocks, marking the functional area type labels of the sample blocks, and constructing a second training sample and a second verification sample based on the sample blocks and the corresponding functional area type labels; Each two-dimensional morphological feature, each three-dimensional morphological feature and each POI feature is respectively used as a functional area classification feature to construct multiple second feature fusion schemes, wherein each second feature fusion scheme includes at least one functional area classification feature; For each second feature fusion scheme, a functional area classification model is constructed according to the functional area classification features included in the second feature fusion scheme, and the functional area classification model is trained based on the second training samples, and the trained functional area classification model is verified based on the second verification samples; A target second feature fusion scheme is determined according to the verification results corresponding to each second feature fusion scheme, and the functional area classification model corresponding to the target second feature fusion scheme is used to identify the functional area type of each block.

5. The method according to any one of claims 1 to 4, characterized in that The two-dimensional morphological features include type coverage, edge density, area-weighted average shape index, area-weighted average fractal dimension index, patch density, landscape shape index, proximity index mean, nearest distance mean, patch cohesion index and Shannon diversity index; the three-dimensional morphological features include three-dimensional building morphological features and three-dimensional tree morphological features, the three-dimensional building morphological features include average building height, maximum building height, building height variation, standardized building height variation, building height range, building surface area, building volume, street aspect ratio H / W, street height-to-length ratio H / L, building surface area coverage, building volume proportion, building edge density, building shape coefficient, building landscape shape index, windward area index, sky view coefficient, the three-dimensional tree morphological features include average tree height, maximum tree height, tree height variation, standardized tree height variation, tree height range, tree surface area, tree volume, tree surface area coverage, tree volume proportion, tree edge density, tree shape coefficient, tree landscape shape index.

6. The method according to claim 5, characterized in that After assigning the building height model and the canopy height model to the land use classification distribution map, the method further includes: Dividing the land use classification distribution map into unit grids; For each grid, the ground surface temperature influencing factors in the two-dimensional morphological features and the three-dimensional morphological features are identified according to the ground surface temperature information, the two-dimensional morphological features and the three-dimensional morphological features.

7. The method according to claim 5, characterized in that Obtain airborne lidar point cloud data, aerial image data, Landsat-8 remote sensing image data, POI data, and OSM data corresponding to the target research city, including: The airborne lidar point cloud data, aerial image data, Landsat-8 remote sensing image data in summer and winter, POI data, and OSM data corresponding to the target research city are obtained respectively to analyze the influencing factors of urban surface temperature in summer and winter of the target research city.

8. A device for analyzing factors affecting urban surface temperature, characterized in that: The device comprises: Data acquisition module, used to obtain airborne lidar point cloud data, aerial image data, Landsat-8 remote sensing image data, POI (Point of Interest) data, and OSM (Open Street Map) data corresponding to the target research city; A surface temperature inversion module is used to perform surface temperature inversion calculation on the Landsat-8 remote sensing image data using a radiation transfer equation method to determine the surface temperature information of the target research city; A type classification module is used to segment the target research city according to the airborne laser radar point cloud data and the aerial image data, and identify the land use type of each area to obtain a land use classification distribution map of the target research city; A feature extraction module is used to extract, based on the land use classification distribution map, a building height model of a land use classified as a building and a canopy height model of a land use classified as a forest in the elevation model corresponding to the airborne laser radar point cloud data, and assign the building height model and the canopy height model to the land use classification distribution map; A functional area classification module is used to divide the target research city into blocks according to the aerial image data and the OSM data, extract POI features of each block based on the POI data, extract at least one two-dimensional morphological feature and at least one three-dimensional morphological feature of each block based on the land use classification distribution map, and identify the functional area type corresponding to each block based on the POI features, the two-dimensional morphological features and the three-dimensional morphological features; The influencing factor analysis module is used to identify the surface temperature influencing factors in the two-dimensional morphological features and the three-dimensional morphological features for the blocks corresponding to each functional area type according to the surface temperature information, the two-dimensional morphological features and the three-dimensional morphological features; wherein, for the blocks corresponding to each functional area type, the two-dimensional morphological features and the three-dimensional morphological features are subjected to single factor analysis according to the surface temperature information, and the factors whose correlation with the surface temperature of the blocks corresponding to the functional area type meets the preset correlation conditions are determined as the analysis factors of the regression analysis; and according to the analysis factors, a plurality of factor analysis schemes are constructed, wherein the factor analysis schemes include a two-dimensional morphological feature analysis scheme, a three-dimensional building ... A multivariate linear regression model is constructed based on each factor analysis scheme respectively, and a regression analysis is performed between each feature in the factor analysis scheme and the ground surface temperature for the blocks corresponding to each functional area type through the multivariate linear regression model to determine the ground surface temperature influencing factors corresponding to each factor analysis scheme, and the ground surface temperature influencing factors in all analysis factors corresponding to each functional area type are determined according to the ground surface temperature influencing factors in the factor analysis schemes corresponding to each functional area type.

9. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

10. A computer device comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

Images