GeoAI machine learning-based intelligent downscaling seasonal characterization and analysis method for surface temperature

Abstract

The invention develops a surface temperature intelligent downscaling seasonal characterization and analysis method based on GeoAI machine learning. According to the method, interpolation and downscaling can be carried out on a surface temperature missing area influenced by cloud layer factors through urban landscape characteristic index products and incomplete satellite surface temperature inversion data. Through a specific stratified sampling mechanism, the problems of training deviation and over-fitting caused by insufficient representativeness of a data training set are solved, and a surface temperature product which takes seasons as units and has a 10m resolution ratio and a complete space-time sequence is generated. Verification shows that the method can accurately compensate for surface temperature data loss caused by factors such as cloud cover, ensures space integrity and time continuity of the surface temperature data, truly reflects the surface temperature of a research area when a satellite passes, and effectively represents seasonal space-time evolution of an urban thermal environment. Besides, the invention further develops a GeoAI-based feature importance intelligent evaluation framework, landscape features strongly related to the surface temperature can be analyzed and screened, scientific support and reference are provided for accurate characterization, modeling and optimization of the urban thermal environment, and the method has important application value.

Description

Technical Field

[0001] The invention relates to the field of machine learning and remote sensing technology, and provides a method for intelligent downscaling seasonal characterization and analysis of surface temperature based on GeoAI machine learning. Background Art

[0002] Urban thermal environment refers to the reaction and transformation of thermal fields after the natural environment is transformed into urban landscape under the influence of human activities, that is, urban built-up areas. Land surface temperature (LST) is a comprehensive reflection of the interaction between the atmosphere and land objects, energy exchange, migration and transformation. The complete and continuous land surface temperature and the related urban spatial landscape characteristics are the basis for characterizing, studying and managing the urban thermal environment.

[0003] At present, the research and governance of urban thermal environment at the city scale are mainly based on the characterization and analysis of the urban heat island effect. The surface temperature is an important reference for the characterization and analysis of urban heat islands and thermal environments. It mainly relies on the temporal and spatial scale advantages of satellite remote sensing data to invert the surface temperature. The remote sensing inversion data is not affected by the region and has a complete spatial scale, but it is blocked by clouds and cannot provide complete temporal and spatial series data for the complete and accurate characterization of the urban thermal environment. At the same time, the urban spatial characteristics are limited by the lack of GIS information data, and it is impossible to accurately characterize the landscape factors that affect and shape the urban thermal environment; therefore, a method that can be used to accurately characterize, analyze and govern the urban thermal environment is very important. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a method for intelligent downscaling seasonal characterization and analysis of land surface temperature based on GeoAI machine learning. The method performs land surface temperature reconstruction and seasonal analysis of urban thermal environment based on satellite images and urban landscape characteristics, achieving the effect of seamless spatiotemporal connection of seasonal urban thermal environment characterization, and laying the foundation for urban thermal environment research. Compared with numerical simulation, the visual intelligent model based on machine learning and remote sensing visual images constructed by the present invention can obtain high spatial resolution land surface temperature products and urban landscape characteristics while ensuring accuracy.

[0005] The present invention comprises the following steps:

[0006] S1: Obtain the Landsat-8 surface temperature products within the study area on a daily basis, preprocess them, and screen effective daily surface temperature products;

[0007] S2: Obtain landscape characteristic index products related to urban land use and land cover (LULC) on an annual basis and preprocess them;

[0008] S3: For any daily surface temperature data in the valid daily surface temperature product, select the valid surface temperature pixels in the study area on that day, extract the urban land use and surface cover data, surface temperature data and Sentianl-2 satellite band data with consistent coordinates according to the geographic spatial location of the valid pixels, perform stratified random sampling, and construct a basic training sample set;

[0009] S4: Based on the basic training sample set, a multi-spatial scale U-Net model for estimating building and road probability indexes is trained to generate a smart city building intensity index and a smart road intensity index to characterize the distribution of urban space buildings and roads;

[0010] S5: Comprehensive landscape characteristic index product, based on the geographic spatial location of effective pixels, extracts comprehensive landscape characteristics consistent with the stratified random sampling spatial location to generate a comprehensive training sample set;

[0011] S6: Based on the comprehensive training sample set, the random forest model used to reconstruct the LST data is trained to reconstruct the surface temperature data of any pixel position within the study area on that day to generate a daily surface temperature product with a complete spatial sequence;

[0012] S7: Take the weighted average of the LST products of spring, summer, autumn and winter, and normalize them to generate the LST seasonal variation index;

[0013] S8: Based on the comprehensive training sample set, a random forest model for landscape feature importance analysis is trained to generate feature contribution analysis, screen important landscape feature index products for the urban heat island effect, and analyze and guide potential spatial governance solutions.

[0014] Furthermore, in step S1, the preprocessing includes: acquiring surface temperature images, date screening, quality screening, cropping of the study area and resampling.

[0015] Furthermore, step S1 specifically includes:

[0016] S1.1: With the center of the study area as the center of the circle, set a buffer zone with a radius of r according to the spatial scale of the study urban area. r should be greater than the maximum distance between the center of the circle and the urban boundary. Obtain the Landsat-8 surface temperature product through the GEE platform, and use the buffer zone as the study area to mask and crop all surface temperature products to the same spatial range as the study area;

[0017] S1.2: Extract the pixel quality band QA_Pixel of Landsat-8 products, analyze the fluctuation of daily surface temperature in the study area in different seasons and the degree of influence by clouds, further screen the quality of Landsat-8 surface temperature products extracted by mask, and obtain daily surface temperature images with qualified quality;

[0018] S1.3 For qualified daily surface temperature images, a bilinear sampling method is used to resample them to a resolution of 10 m to make them consistent with the spatial resolution of the Sentinal-2 satellite.

[0019] Furthermore, in step S1.2, the quality screening criteria are that the cloud coverage of a single surface temperature image is less than 20% and the surface temperature image covers more than 50% of the study area.

[0020] The formula for calculating the cloud coverage rate of a single image is as follows:

[0021]

[0022] Among them, w i represents the cloud coverage rate on the i-th day, that is, the percentage of missing values ​​in pixels on the i-th day. represents the number of valid pixels in the Landsat-8 surface temperature product of the i-th day after cropping, that is, the effective image coverage area, w ref Represents the number of pixels in the study area, that is, the area of ​​the study area.

[0023] The validity of a pixel at any spatial location is determined as follows: If the clouds Bit Mask and the shadow Bit Mask in QA_Pixel are both equal to 0, then If the cloud mask cloudsBitMask in QA_Pixel is equal to 1 or the shadow mask cloudShadowBitMask is equal to 1, then Where r and c represent the horizontal and vertical coordinates of the pixel position respectively.

[0024] Furthermore, in step S2, the landscape characteristic index products include: Sentinel satellite bands, remote sensing satellite index (RSEI vegetation index), elevation data, night light data, urban land use and surface cover data, building distribution outlines, and road outline data.

[0025] Furthermore, step S2 specifically includes:

[0026] S2.1: Through the GEE platform, the band data (B2, B3, B4, B8A, B8, B9, B11, B12) with a resolution of 10m are obtained from the Sentinel-2 satellite database as band index indicators, and the band data obtained by S2.1 are used to generate RSEI vegetation index products (NDVI, NDBI, NDWI) as basic data set channels 1-11;

[0027] S2.2: From the GEE platform, obtain the urban land use and land cover data (LULC) with a spatial resolution of 10m in the Dynamicworld data as the land use functional zoning factor index and as the 12th to 21st channels of the basic data set;

[0028] S2.3: From the GEE platform, obtain the SRTM elevation data (DEM) with a spatial resolution of 30 m in the Earthdata database as the physical morphological factor index and as the 22nd channel of the basic data set;

[0029] S2.4: From the GEE platform, obtain the VIIRS sensor spatial resolution 400 night light data (NLT) as an indicator of socioeconomic factors and as the 23rd channel of the basic data set;

[0030] S2.5: From the GEE platform, obtain the built-up area building volume data (BV) and building height data (BH) of the built-up area with a spatial resolution of 100m in the GHSL database as the physical morphological factors of the urban built-up area and as the 24th and 25th channels of the basic data set;

[0031] S2.6: Download the latest street data, building distribution data, street levels and building functional attributes of the study area from the OSM platform, screen the street data and building distribution data, extract the street and building distribution vector data, and perform vector-to-raster processing and morphological processing as the basic data for characterizing the physical morphological factors of the road distribution and building distribution in the urban built-up area, and use them to generate the physical morphological factors of the built-up area including the smart city building index and the smart road intensity index. These features will be used to generate the smart city building index (Built footprints) and the smart road intensity index (Road footprints) as the 26th and 27th channels of the comprehensive data set; and use the obtained street vector data to obtain the block outline of the study area.

[0032] S2.7 uses the data obtained from S2.5 and S2.6 to calculate and generate the physical morphology factors of the built-up areas; the physical morphology factors of the built-up areas include: building density (BD) of urban built-up areas, sky width (SVF) of urban built-up areas, and floor ratio (FAR) of urban built-up areas, which are channels 28-30 of the comprehensive data set.

[0033] S2.8 performs preprocessing operations on the image products obtained in S2.1-S2.7.

[0034] Furthermore, the band data obtained in step S2.1, the specific method of generating the RSEI vegetation index product is:

[0035] Using the band, the NDVI expression is generated as follows:

[0036]

[0037] Among them, NDVI stands for Normalized Difference Vegetation Index, NIR stands for near infrared band, i.e. band 8, and RED stands for red light band, i.e. band 4.

[0038] Using the band, the NDWI is generated and its expression is:

[0039]

[0040] Among them, NDWI stands for normalized water index, NIR stands for near infrared band, i.e. band 8, and GREEN stands for green light band, i.e. band 3.

[0041] Using the bands, the NDBI expression is generated as follows:

[0042]

[0043] Among them, NDBI stands for Normalized Building Index, NIR stands for near infrared band, i.e. band 8, and SWIR stands for shortwave red light band, i.e. band 11.

[0044] Further, in step S2.1, the preprocessing method is:

[0045] Including annual median calculation of periodic bands and RSEI index products to generate index products with consistent time series;

[0046] Furthermore, in step S2.6

[0047] The street data and building distribution data obtained from the OSM platform were screened using street levels and building functional attributes, and the street data and building distribution data that were consistent with the remote sensing image content were extracted. The obtained street and building distribution data were converted from vector to raster in QGIS. The rasterized street and building distribution data were morphologically processed to achieve the effect of noise reduction. The building distribution data after morphological processing is a binary image. The area value of the building distribution is 1, and the background area value is 0, which serves as the basic data of the physical morphological factors that characterize the building distribution in the urban built-up area. Similarly, the road distribution data after morphological processing is also a binary image, with the area value of the road distribution being 1 and the background area value being 0, which serves as the basic data of the physical morphological factors that characterize the road distribution in the urban built-up area.

[0048] Furthermore, the morphological processing in step S2.6 is as follows:

[0049] The rasterized street data is first binarized, then morphological dilation is performed to merge multi-directional roads, and then morphological skeleton extraction and dilation are performed to accurately depict the distribution of roads.

[0050] The rasterized building distribution data is first binarized and then subjected to partition morphological closing and filling operations.

[0051] Furthermore, in step S2.7

[0052] For building density (BD), the calculation formula is as follows:

[0053]

[0054] Among them, A b represents the total area of ​​all buildings in the block, A S Represents the total area of ​​the block.

[0055] For the urban built-up area floor ratio (FAR), the calculation formula is as follows:

[0056]

[0057] in, and are the building area and number of floors of the jth building in the i-th block, n represents the total number of buildings, s i Represents the block area of ​​the i-th block.

[0058] For the sky width (SVF) of urban built-up areas, considering that the complex spatial environment within the urban area has an impact on the absorption and emission of long-wave radiation, the building height data (BH) and the elevation data (DEM) of the study area are made into a digital surface model and input into the SAGA GIS tool. At the same time, the maximum search radius is set to 100m to obtain the sky width (SVF) of urban built-up areas.

[0059] In a three-dimensional urban environment, the SVF calculation formula for a point is as follows:

[0060]

[0061] Where β is the angle from the center point to the maximum obstacle height at the maximum distance equal to the constant search radius (R).

[0062] Furthermore, in step S2.7, the preprocessing includes downsampling all images and landscape index products with a spatial resolution greater than 10m to 10m spatial resolution using a bilinear sampling method to generate images and landscape index products with consistent spatial resolution. In addition, the RSEI index products of the study area are cropped to generate index products with consistent spatiotemporal resolution.

[0063] Furthermore, step S3 specifically includes:

[0064] S3.1 Use the daily surface temperature data of any day in the valid daily surface temperature product to select the pixel linear index coordinates of the valid surface temperature pixels in the study area on that day, and read the LULC data, valid surface temperature data and Sentianl-2 satellite band product data with consistent geographical location into the table to form the basic data set;

[0065] S3.2 uses the surface temperature data in the basic data set as the stratification standard, performs stratified sampling, sets an appropriate sampling interval z, iterates k times during sampling, and randomly selects m samples from each layer each time to construct a training sample set. Assuming that the value interval of LST is Z, the number of samples generated is M = Z / z*k*m. The smaller the z value, the smaller the sampling interval, and the fewer the number of samples, and vice versa.

[0066] S3.3 draws a probability distribution histogram of the extracted samples to determine whether the sample distribution is uniform. K and M are constantly adjusted to make the sample distribution approach a balanced distribution, which is used as the basic sample training set and as the input of the subsequent model.

[0067] Furthermore, step S4 specifically includes:

[0068] S4.1 constructs a multi-spatial-scale U-Net model based on convolutional neural network (CNN) to generate the smart city building index and smart road intensity index.

[0069] S4.2 is based on the basic sample training set, takes the geographic spatial position of each sample as the index, extracts the exponential feature image of the image area with a neighborhood pixel size of [64,64] as the sample, and constructs the CNN sample training set.

[0070] S4.3 The landscape feature index product of channel 1 to channel 25 in the CNN sample training set is used as x, and the road distribution data is used as y road Train the multi-spatial scale U-Net model. The model training effect is evaluated by calculating the model training accuracy. 2 Higher models generate smart road strength index.

[0071] S4.4 takes the landscape feature index product of channel 1 to channel 25 in the CNN sample training set as x and the building distribution data as y built Train the multi-spatial scale U-Net model. The model training effect is evaluated by calculating the model training accuracy. 2 Higher models generate smart road strength index.

[0072] Furthermore, the U-Net model described in S4.3 is set to have a depth of 2, a model classification category number of 2, a specified solver as an adaptive matrix, a maximum number of traversals of 30, a small batch size of 32 for each iterative training, an initial learning rate of 1e-5, a learning decay rate of 0.1, and a learning decay traversal number of 10. The mathematical relationship of the multi-spatial scale U-Net model is as follows:

[0073]

[0074] Where m CNN2road represents the multi-spatial-scale U-Net model for generating intelligent road strength index, represents the intelligent road strength index generated by the U-Net model, and x represents the data in the CNN sample training set.

[0075] Furthermore, the U-Net model described in S4.4 is set to have a depth of 2, a model classification category number of 2, a specified solver as an adaptive matrix, a maximum number of traversals of 30, a small batch size of 32 for each iterative training, an initial learning rate of 1e-5, a learning decay rate of 0.1, and a learning decay traversal number of 10. The mathematical relationship of the multi-spatial scale U-Net model is as follows:

[0076]

[0077] Where m CNN2built represents a multi-spatial-scale U-Net model for generating smart city building index, represents the smart city building index generated by the U-Net model, and x represents the data in the CNN sample training set.

[0078] Furthermore, step S5 specifically includes:

[0079] S5.1 By studying the pixel index coordinates of valid pixels in the region, the smart road intensity index and smart city building index with consistent geographical locations are superimposed on the basic data set to form a comprehensive data set;

[0080] S5.2 extracts the corresponding building and road feature indexes in the comprehensive data set through the geographic location index of the basic training set to generate a comprehensive training set to generate a daily surface temperature product with a complete spatiotemporal sequence.

[0081] Furthermore, step S6 specifically includes:

[0082] S6.1 Take the landscape characteristic index product of channels 1-30 in the comprehensive sample training set as X and LST as y LST Train a random forest model.

[0083] S6.2 Use the trained random forest model to reconstruct all pixels on any valid date of surface temperature in the study area to generate a daily surface temperature product with a complete spatiotemporal sequence.

[0084] Furthermore, in the random forest model of S6.1, the number of decision trees is set to 50 and the minimum leaf size is set to 5. The mathematical relationship is as follows:

[0085]

[0086] Where m RF2LST represents the random forest model for reconstructing land surface temperature, represents the surface temperature data reconstructed by the random forest model, and X represents the landscape characteristic index product in the comprehensive sample training set.

[0087] Furthermore, step S7 specifically includes:

[0088] S7.1 For all reconstructed surface temperature data in the same season, weighted average is performed based on the proportion of missing pixel values ​​caused by cloud cover and other reasons before interpolation to obtain the surface temperature data for that season;

[0089] S7.2 Normalize the surface temperature data of the four seasons;

[0090] S7.3 Calculate the local Moran index for the normalized land surface temperature data to obtain the LST seasonal variation index.

[0091] Furthermore, the formula for weighted average of surface temperature in the same season is as follows:

[0092]

[0093] in, represents the simulated LST value on the i-th day, w lst i represents the percentage of missing values ​​for that day, w lst i A value of 1 indicates complete absence, where i is any date containing a surface temperature. Represents the surface temperature of the kth season, where k is 1, 2, 3, or 4, corresponding to spring, summer, autumn, and winter.

[0094] Furthermore, the formula for normalizing the four seasons is as follows:

[0095]

[0096] Among them, LST k Represents the heat island index for a particular season, Represents the minimum value of the water body in the study area on that day, represents the 95th percentile value of the surface temperature of this day and night. Represents the normalized seasonal heat island index.

[0097] Further, step S8 specifically includes:

[0098] S8.1 uses the features in the comprehensive training set as x, and uses the corresponding pixel position index of x to extract the heat island index of spring, summer, autumn and winter as Construct a training set for feature seasonality estimation.

[0099] S8.2 trains random forest models for the heat island index in different seasons and analyzes the contribution of various landscape factors to the heat island in different seasons.

[0100] S8.3 Sort the feature contributions and screen the important landscape features associated with the urban heat island effect. Then, analyze and guide potential urban space optimization plans based on the physical meaning of the urban space mapped by the features.

[0101] Compared with the prior art, the advantages and positive effects of the present invention are as follows: the present invention can fully integrate and develop the temporal, spatial and spectral information advantages of open source satellite image data, urban land use and surface cover data and GIS database, and combine them with intelligent GeoAI algorithms to merge them into an urban surface temperature dataset of daily, monthly, seasonal and annual sequences with optimal temporal and spatial scales to characterize the seasonal variation characteristics of the urban thermal environment.

[0102] The present invention uses GeoAI intelligent perception to establish a mathematical logic model between urban landscape index and surface temperature, thereby obtaining urban-scale surface temperature downscaling data with complete spatiotemporal coverage and high spatial resolution. This method effectively solves the following three problems: First, the lack of urban spatial GIS information database leads to incomplete representation of urban spatial structure, which in turn affects the accurate assessment of urban thermal environment; second, the urban landscape characteristic index in remote sensing images is diverse and lacks standardization, making it difficult to accurately depict urban spatial structure and landscape characteristics; third, surface temperature data is easily affected by clouds and satellite operation cycles, resulting in incomplete spatiotemporal scales and low spatial resolution. Through the complete simulation of daily surface temperature, the present invention provides strong support for the accurate characterization and analysis of changes in urban thermal environment.

[0103] The present invention effectively overcomes the impact of clouds on urban thermal environment observations and ensures the spatiotemporal continuity of surface temperature data. Starting from machine learning intelligent perception, the urban spatial characteristics are intelligently and accurately characterized, and on this basis, high-resolution downscaling of surface temperature data is achieved. In addition, the present invention also provides a feature importance analysis function to help users deeply explore the inherent mechanism of the urban thermal environment, reveal key urban layout and spatial structure characteristics, provide a scientific basis for urban planning departments, and support the formulation of more targeted spatial optimization and governance strategies. The present invention makes full use of open source urban spatial information and remote sensing data to provide a highly universal and easy-to-promote solution for urban thermal environment research. BRIEF DESCRIPTION OF THE DRAWINGS

[0104] Figure 1 Flowchart of the GeoAI machine learning-based intelligent downscaling seasonal characterization and analysis method for surface temperature provided by the present invention.

[0105] Figure 2 Provide intelligent characterization products for downscaling urban landscape and surface temperature based on GeoAI.

[0106] Figure 3 It is a product for analyzing the contribution of urban landscape factors to summer heat islands and winter cold islands. DETAILED DESCRIPTION

[0107] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0108] The purpose of the present invention is to develop a method for seasonal characterization and analysis of land surface temperature based on GeoAI machine learning intelligent downscaling. The method can interpolate missing values ​​in the missing areas of land surface temperature affected by cloud factors through the available urban landscape characteristic index products and incomplete satellite land surface temperature inversion data, and downscale the land surface temperature products. In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0109] According to the present invention, to achieve this object, the present invention comprises the following steps:

[0110] S1: Obtain the Landsat-8 surface temperature product within the study area on a daily basis, preprocess it, and screen qualified daily surface temperature pixels;

[0111] Furthermore, step S1 specifically comprises:

[0112] This study takes the city center as the center and sets a buffer zone with a radius of 30 km according to the spatial scale of the city. The Landsat-8 surface temperature product is obtained through the GEE platform, and the buffer zone is used as the study area. All surface temperature products are masked and cropped to be consistent with the spatial range of the study area; further, the original Landsat-8 surface temperature product is further screened with a single image with a cloud coverage rate of less than 20%, and the image units covering more than 50% of the study area are further screened as the quality screening standard to obtain the daily surface temperature images of qualified dates. In this example, the qualified daily surface temperature images obtained are the 8th day, the 56th day, the 72nd day, the 88th day, the 200th day, the 232nd day, the 264th day, the 280th day, the 296th day, the 312th day and the 328th day; further, the daily surface temperature images with qualified quality are resampled to 10m resolution using the bilinear sampling method to make them consistent with the spatial resolution of the Sentinal-2 satellite.

[0113] Preferably, in step S1.2, the quality screening criteria are that the cloud coverage of a single surface temperature image is less than 20% and the surface temperature image covers more than 50% of the study area.

[0114] Furthermore, the calculation formula for the cloud coverage of a single image is as follows:

[0115]

[0116] Among them, w i represents the cloud coverage rate on the i-th day, that is, the percentage of missing values ​​in pixels on the i-th day. represents the number of valid pixels in the Landsat-8 surface temperature product of the i-th day after cropping, that is, the effective image coverage area, w ref Indicates the number of pixels in the study area, that is, the area of ​​the study area. In this example, i = 8, 56, 72, 88, 200, 232, 264, 280, 296, 312, 328.

[0117] The validity of any spatial location pixel is determined as follows:

[0118] If the cloud mask cloudsBitMask and shadow mask cloudShadowBitMask in QA_Pixel are both equal to 0, then If QA_Pixel cloudsBitMask is equal to 1 or shadow mask cloudShadowBitMask is equal to 1, then Where r and c represent the horizontal and vertical coordinates of the pixel position respectively.

[0119] S2: Obtain Sentianl-2 satellite band products, open source GHSL, Dynamic World and OSM landscape characteristic index products related to urban land use and land cover on an annual basis and preprocess them.

[0120] Furthermore, step S2 specifically includes: obtaining the band data (B2, B3, B4, B8A, B8, B9, B11, B12) of the Sentinel-2 satellite in 2018 through the Google Earth Engine (GEE) platform, and using the band data obtained by the Sentinel-2 satellite to generate RSEI vegetation index products (NDVI, NDBI, NDWI); obtaining the VIIRS sensor's 2018 night light data with a spatial resolution of 400m from the GEE platform, which can be used to measure the status of social and economic development; obtaining the SRTM elevation data with a spatial resolution of 30m from the Earthdata database from the GEE platform to reflect the changes in the terrain of the study area; and obtaining the 2018 night light data with a spatial resolution of 400m from the GEE platform. The GEE platform was used to obtain the 2018 urban land use and land cover data with a spatial resolution of 10m from the Dynamicworld database as indicators of land use function zoning factors; the GEE platform was used to obtain the built-up area building volume data (BV) and building height data (BH) with a spatial resolution of 100m from the GHSL database as indicators of physical morphological factors of urban built-up areas; the building and road outline vector data were obtained from the OSM platform, and vector-to-raster conversion and morphological processing were performed as basic data to characterize the physical morphological factors of urban built-up areas, which were used to generate the smart city building index (Built footprints) and smart road intensity index (Road footprints); the physical morphological factors of urban built-up areas were used to generate the physical morphological factors of built-up areas: building density (BD) in urban built-up areas, sky openness (SVF) in urban built-up areas, and floor ratio (FAR) in urban built-up areas. The obtained landscape characteristic index products were uniformly cropped to the size of the study area, and interpolated to a resolution of 10m by the nearest neighbor interpolation method.

[0121] Furthermore, the specific method for generating the RSEI vegetation index product is:

[0122] Using the band, the NDVI expression is generated as follows:

[0123]

[0124] Among them, NDVI stands for Normalized Difference Vegetation Index, NIR stands for near infrared band, i.e. band 8, and RED stands for red light band, i.e. band 4.

[0125] Using the band, the NDWI is generated and its expression is:

[0126]

[0127] Among them, NDWI stands for normalized water index, NIR stands for near infrared band, i.e. band 8, and GREEN stands for green light band, i.e. band 3.

[0128] Using the bands, the NDBI expression is generated as follows:

[0129]

[0130] Among them, NDBI stands for Normalized Building Index, NIR stands for near infrared band, i.e. band 8, and SWIR stands for shortwave red light band, i.e. band 11.

[0131] Furthermore, the specific method of morphological processing after converting the building and road outline vectors into raster is:

[0132] For the rasterized street data, we first binarize it, assigning all gray values ​​greater than or equal to 1 to 1; then we morphologically expand each pixel by 15 pixels to the surrounding area to merge multiple roads; then we perform morphological skeleton extraction and expand again with the same number of pixels to accurately depict the distribution of roads.

[0133] The rasterized building distribution data is first binarized, and all grayscale values ​​greater than or equal to 1 are assigned to 1; then, by finding connected areas, partition morphological closing and filling operations are performed. In this example, a total of 7523 connected areas are found.

[0134] For the building density (BD) of urban built-up areas, the calculation formula is as follows:

[0135] BD=A b / A S ,

[0136] Among them, A b represents the total area of ​​all buildings in the block, A S Represents the total area of ​​the block.

[0137] For the urban built-up area floor ratio (FAR), the calculation formula is as follows:

[0138]

[0139] Among them, A i and F i are the area and number of floors of the ith building, n represents the total number of buildings, and S represents the area of ​​the block.

[0140] For the sky width (SVF) of urban built-up areas, considering the complex spatial environment within a certain urban area, such as dense old residential areas or high-rise buildings, which have an impact on the absorption and emission of long-wave radiation, the building height data (BH) and the elevation data (DEM) of the study area are made into a digital surface model and input into the SAGA GIS tool. At the same time, the maximum search radius is set to 100m to obtain the sky width (SVF) of urban built-up areas.

[0141] In a three-dimensional urban environment, the SVF calculation formula for a point is as follows:

[0142]

[0143] where β is the angle from the center point to the maximum obstacle height at a maximum distance equal to the constant search radius R.

[0144] S3: For any daily surface temperature data in the valid daily surface temperature product, select the valid surface temperature pixels in the study area on that day, extract the LULC, surface temperature and Sentianl-2 satellite band information with consistent coordinates according to the geographic spatial position of the valid pixels, perform stratified random sampling, and construct a basic training sample set;

[0145] Furthermore, step S3 specifically includes: reading the LULC data, effective surface temperature data and Sentianl-2 satellite band product data with consistent geographical location into a table to form a basic data set through the pixel linear index coordinates of the effective pixels in the study area; taking the surface temperature data in the basic data set as the stratification standard, performing stratified sampling, setting a suitable sampling interval of 10, iterating 100 times during sampling, and extracting 20 samples in each layer each time.

[0146] S4: Based on the basic training sample set, a multi-spatial scale U-Net model for estimating building and road probability indexes is trained to generate a smart city building intensity index and a smart road intensity index to characterize the distribution of urban space buildings and roads;

[0147] Furthermore, in this example, step S4 specifically includes: constructing a multi-spatial scale U-Net model based on a convolutional neural network (CNN) to generate a smart city building index and a smart road strength index. Based on the basic sample training set, taking the geographic spatial position of each sample as an index, extracting the index feature image of the image area with a neighborhood pixel size of [64,64] as a sample, and constructing a CNN sample training set.

[0148] Furthermore, the landscape feature index product of channel 1 to channel 25 in the CNN sample training set is used as x, and the road distribution data is used as y roadTrain a multi-spatial scale U-Net model. In this example, the intelligent road strength index is generated by selecting a model with a training accuracy of 0.85. Channels 1 to 25 in the CNN sample training set are used as x, and building distribution data is used as y built Train a multi-spatial scale U-Net model. In this example, the smart city building index is generated by selecting a model with a training accuracy of 0.83.

[0149] Furthermore, the multi-spatial scale U-Net model for generating the intelligent road strength index is set to have a depth of 2, a model classification category of 2, a specified solver of an adaptive matrix, a maximum number of iterations of 1, a small batch size of 32 for each iteration training, an initial learning rate of 1e-5, a learning decay rate of 0.1, and a learning decay iteration number of 1. The mathematical relationship is as follows:

[0150]

[0151] Where m CNN2road represents the multi-spatial-scale U-Net model for generating intelligent road strength index, represents the intelligent road strength index generated by the U-Net model, and x represents the data in the CNN sample training set.

[0152] Furthermore, the multi-spatial scale U-Net model for generating the smart city building index is set to have a depth of 2, a model classification category of 2, a specified solver of an adaptive matrix, a maximum number of iterations of 1, a small batch size of 32 for each iteration training, an initial learning rate of 1e-5, a learning decay rate of 0.1, and a learning decay iteration number of 1. The mathematical relationship is as follows:

[0153]

[0154] Where m CNN2built represents a multi-spatial-scale U-Net model for generating smart city building index, represents the smart city building index generated by the U-Net model, and x represents the data in the CNN sample training set.

[0155] S5: Comprehensive landscape characteristic index product, based on the geographic spatial location of effective pixels, extracts comprehensive landscape characteristics consistent with the stratified random sampling spatial location to generate a comprehensive training sample set;

[0156] The landscape feature index products in step S5 include three categories. The first category is satellite band data, including eight bands of Sentinel-2, namely B2, B3, B4, B8A, B8, B9, B11 and B12; the second category is remote sensing satellite index data (RSEI index product), including normalized difference vegetation index (NDVI), normalized difference water index (NDWI) and normalized building index (NDBI); the third category is land cover data, including nine different types of land cover data; the fourth category is social-physical data, including the generation of smart city building index (Built footprints), smart road intensity index (Road footprints), built-up area building volume data (BV), built-up area building height data (BH), urban built-up area building density (BD), urban built-up area sky openness (SVF) and urban built-up area floor ratio (FAR), elevation data (DEM) describing terrain undulations, and night light data (NTL) describing social and economic development.

[0157] Furthermore, in this example, step S5 specifically includes: by studying the pixel index coordinates of the valid pixels in the area, superimposing the smart road intensity index and smart city building index with the same geographical location to the basic data set to form a comprehensive data set. Furthermore, by using the geographical location index of the basic training set, the corresponding building and road feature indexes in the comprehensive data set are extracted to generate a comprehensive training set to generate a daily surface temperature product with a complete spatiotemporal sequence.

[0158] S6: Based on the comprehensive training sample set, the random forest model used to reconstruct the LST data is trained to reconstruct the surface temperature data of any pixel position within the study area on that day to generate a daily surface temperature product with a complete spatial sequence;

[0159] Further, in this example, step S6 specifically includes: using the landscape characteristic index product of channels 1-30 in the comprehensive sample training set as X, LST as y LST Train the random forest model. In this example, a random forest model with an rmse of 0.003 and an r-sequare of 0.956 is used to reconstruct all pixels on any valid date of land surface temperature in the study area to generate a daily land surface temperature product with a complete spatiotemporal sequence.

[0160] Furthermore, in the random forest model of S6.1, the number of decision trees is set to 50 and the minimum leaf size is set to 5. The mathematical relationship is as follows:

[0161]

[0162] Where m RF2LST represents the random forest model for reconstructing land surface temperature, represents the surface temperature data reconstructed by the random forest model, and X represents the landscape characteristic index product in the comprehensive sample training set.

[0163] S7: Take the weighted average of the LST products of spring, summer, autumn and winter, and normalize them to generate the LST seasonal variation index.

[0164] Furthermore, in this example, step S7 specifically includes: for the reconstructed surface temperature data of several days in the same season, weighted average is performed according to the proportion of pixel missing values ​​caused by cloud cover and other reasons before interpolation, so as to obtain the surface temperature data of the season. In this example, Zhengzhou is taken as the study area, and according to the local climate, the 60th to 150th day is classified as spring, the 151st to 243rd day is classified as summer, the 244th to 334th day is classified as autumn, and the 335th to 365th day and the 1st to 60th day are classified as winter.

[0165] The surface temperature data of the four seasons were normalized. The local Moran index of the normalized surface temperature data was calculated using ArcGIS software to obtain the LST seasonal variation index.

[0166] Furthermore, the formula for weighted average of surface temperature in the same season is as follows:

[0167]

[0168] in, represents the simulated LST value on the i-th day, w i represents the percentage of missing values ​​for that day, w i Equal to 1 means complete absence, where i=8, 56, 72, 88, 200, 232, 264, 280, 296, 312, 328. Represents the surface temperature of the kth season, where k is 1, 2, 3, or 4, corresponding to spring, summer, autumn, and winter.

[0169] Furthermore, the formula for normalizing the four seasons is as follows:

[0170] y lst =(lst-lst water ) / (lst 95％Percentile -lst water ),

[0171] Among them, lst represents the surface temperature data of a specific season, lst water Represents the minimum value of the water body in the study area on that day, lst 95％Percentile represents the 95th percentile value of the surface temperature of this day and night, y lstRepresents the normalized seasonal heat island index.

[0172] S8: Based on the comprehensive training sample set, a random forest model for landscape feature importance analysis is trained to generate feature contribution analysis, screen important landscape feature index products for the urban heat island effect, and analyze and guide potential spatial governance solutions.

[0173] Furthermore, in this example, step S8 specifically includes using the features in the comprehensive training set as x, using the pixel position index corresponding to x, and extracting the heat island index of the four seasons of spring, summer, autumn, and winter as y i , construct a characteristic seasonal evaluation training set. For the heat island index of different seasons, train the random forest model and analyze the contribution of each landscape factor to the heat island in different seasons. Sort the characteristic contribution, screen the important landscape features associated with the heat island effect, and analyze and guide the potential urban space optimization plan according to the physical meaning of the urban space mapped by the characteristics.

[0174] Furthermore, in this example, the contribution of various landscape factors to summer heat islands and winter cold islands is analyzed as follows: Figure 3 As shown in the figure, the land cover factor and social physical factor are significantly higher than the spectral and vegetation index factors in both summer and winter environments. In the summer heat island environment, the land cover factor (35.95%) and social physical factor (33.44%) contribute more significantly to UTE than the spectral data (25.21%) or vegetation index (5.40%). In the winter cold environment, the land cover factor (36.00%) and social physical factor (41.82%) contribute more significantly to UTE than the spectral data (13.86%) or vegetation index (8.32%).

[0175] Depend on Figure 3 It can be seen that among the types of social physical factors, the contribution of elevation DEM and urban openness SVF to the heat island and cold island effects is very significant, followed by the night light index NTL in summer and the FAR of buildings in winter. This shows that when optimizing urban space, it is necessary to comprehensively consider the terrain elevation of the region to rationally plan the number, density and height of buildings in the urban area to alleviate the urban heat island effect. At the same time, in terms of capturing the impact of landscape factors on LST, the intelligent simulation building and road index generated by GeoAI is usually better than the traditional remote sensing vegetation index.

Claims

1. The intelligent downscaling seasonal characterization and analysis method of surface temperature based on GeoAI machine learning is characterized by: The following steps are involved: S1: Obtain the Landsat-8 surface temperature products within the study area on a daily basis, preprocess them, and screen effective daily surface temperature products; S2: Obtain landscape characteristic index products related to urban land use and land cover on an annual basis and preprocess them; S3: for the daily surface temperature data of any day in the effective daily surface temperature product, screen the effective surface temperature pixels in the study area on that day, extract the urban land use and surface cover data, surface temperature data and Sentianl-2 satellite band data with consistent coordinates according to the geographic spatial position of the effective pixels, perform stratified random sampling, and construct a basic training sample set; S4: Based on the basic training sample set, a multi-spatial scale U-Net model for estimating building and road probability index is trained to generate a smart city building index and a smart road intensity index to characterize the distribution of urban space buildings and roads; S5: Comprehensive landscape characteristic index product, based on the geographic spatial location of effective pixels, extracts comprehensive landscape characteristics consistent with the stratified random sampling spatial location to generate a comprehensive training sample set; S6: Based on the comprehensive training sample set, a random forest model for reconstructing land surface temperature data is trained to reconstruct the land surface temperature data at any pixel position within the study area on that day, and generate a daily land surface temperature product with a complete spatial sequence; S7: Take the weighted average of the surface temperature products of spring, summer, autumn and winter, and normalize them to generate the surface temperature seasonal change index; S8: Based on the comprehensive training sample set, a random forest model for landscape feature importance analysis is trained to generate feature contribution analysis, screen landscape feature index products that are important for the urban heat island effect, and analyze and guide potential spatial governance solutions.

2. The method for intelligent downscaling seasonal characterization and analysis of land surface temperature based on GeoAI machine learning according to claim 1 is characterized in that In step S1, the preprocessing includes: acquiring surface temperature images, date screening, quality screening, cropping of the study area and resampling; step S1 specifically includes: S1.1: With the center of the study area as the center of the circle, set a buffer zone with a radius of r according to the spatial scale of the study urban area. r should be greater than the maximum distance between the center of the circle and the urban boundary. Obtain the Landsat-8 surface temperature product through the GEE platform, and use the buffer zone as the study area to mask and crop all surface temperature products to be consistent with the spatial range of the study area. S1.2: Extract the pixel quality band QA_Pixel of Landsat-8 products, analyze the fluctuation of daily surface temperature in the study area in different seasons and the degree of influence by clouds, further screen the quality of Landsat-8 surface temperature products extracted by mask, and obtain daily surface temperature images with qualified quality; S1.3 For qualified daily surface temperature images, a bilinear sampling method is used to resample them to a resolution of 10 m to make them consistent with the spatial resolution of the Sentinal-2 satellite.

3. The method for intelligent downscaling seasonal characterization and analysis of land surface temperature based on GeoAI machine learning according to claim 2 is characterized in that In step S1.2, the quality screening criteria are that the cloud coverage of a single surface temperature image is less than 20% and the surface temperature image covers more than 50% of the study area; The calculation formula for the cloud coverage rate of a single image is as follows: Among them, w i represents the cloud coverage rate on the i-th day, that is, the percentage of missing values ​​in pixels on the i-th day. represents the number of valid pixels in the Landsat-8 surface temperature product of the i-th day after cropping, that is, the effective image coverage area, w ref Indicates the number of pixels in the study area, that is, the area of ​​the study area; the criterion for determining the validity of a pixel at any spatial position is: if the cloud mask and shadow mask in QA_Pixel are both equal to 0, then If the cloud mask in QA_Pixel is equal to 1 or the shadow mask is equal to 1, then Where r and c represent the horizontal and vertical coordinates of the pixel position respectively.

4. The method for intelligent downscaling seasonal characterization and analysis of land surface temperature based on GeoAI machine learning according to claim 1 is characterized in that The landscape feature index products in step S2 include: Sentinel satellite bands, remote sensing satellite index, elevation data, night light data, urban land use and surface cover data, building distribution outlines, and road outline data; step S2 specifically also includes: S2.1: Obtain band data of B2, B3, B4, B8A, B8, B9, B11 and B12 with a resolution of 10m from the Sentinel-2 satellite database through the GEE platform as band index indicators, and use the obtained band data to calculate and generate remote sensing satellite index products of normalized vegetation index, normalized water index and normalized building index; the band index indicators and the remote sensing satellite index products are used as channels 1-11 of the basic data set; S2.2: From the GEE platform, obtain the land use and land cover data with a spatial resolution of 10m in the Dynamicworld data as the land use functional zoning factor index and as the 12th to 21st channels of the basic data set; S2.3: From the GEE platform, obtain the Earthdata database with a spatial resolution of 30 SRTM elevation data (DEM) as the physical morphological factor index and the 22nd channel of the basic data set; S2.4: From the GEE platform, obtain the VIIRS sensor spatial resolution 400 night light data (NLT) as an indicator of socioeconomic factors and as the 23rd channel of the basic data set; S2.5: From the GEE platform, obtain the built-up area building volume data and building height data of the built-up area with a spatial resolution of 100m in the GHSL database as the physical morphological factors of the urban built-up area and as the 24th and 25th channels of the basic data set; S2.6: Download the latest street data, building distribution data, street grade and building function attribute data of the study area from the OSM platform, screen the street data and building distribution data, extract the street and building distribution vector data, and perform vector-to-raster conversion and morphological processing as the basic data for characterizing the physical morphological factors of the road distribution and building distribution in the urban built-up area, and use them to generate the physical morphological factors of the built-up area including the smart city building index and the smart road intensity index, which are used as the 26th and 27th channels of the comprehensive data set; and use the obtained street vector data to obtain the block outline of the study area; S2.7: Using the data obtained in S2.5 and S2.6, calculate and generate the physical form factor of the built-up area; the physical form factor of the built-up area includes: building density of urban built-up area, sky openness of urban built-up area and floor ratio of urban built-up area, as the 28th to 30th channels of the comprehensive data set; in the urban three-dimensional environment, the calculation formula of the sky openness of the urban built-up area at a certain point is: Where SVF is the sky openness of the urban built-up area, β is the angle θ from the center point to the maximum obstacle height at the maximum distance equal to the constant search radius R; S2.8: Perform preprocessing operations on the image products obtained in S2.1-S2.

7.

5. The method for intelligent downscaling seasonal characterization and analysis of land surface temperature based on GeoAI machine learning according to claim 1 is characterized in that The step S3 specifically includes: S3.1 Use the daily surface temperature data of any day in the valid daily surface temperature product to select the pixel linear index coordinates of the valid surface temperature pixels in the study area on that day, and read the urban land use and surface cover data, valid surface temperature data and Sentianl-2 satellite band product data with consistent geographical locations into the table to form the basic data set; S3.2 uses the surface temperature data in the basic data set as the stratification standard, performs stratified sampling, sets an appropriate sampling interval z, iterates k times during sampling, and randomly selects m samples in each layer each time to construct a training sample set; S3.3 draws the probability distribution histogram of the extracted samples to determine whether the sample distribution is uniform, and continuously adjusts k and m to make the sample distribution approach a balanced distribution, which is used as the basic sample training set and as the input of subsequent models.

6. The method for intelligent representation and seasonal analysis of land surface temperature based on GeoAI machine learning according to claim 1 is characterized in that The step S4 specifically includes: S4.1 builds a multi-spatial-scale U-Net model based on convolutional neural networks to generate smart city building index and smart road intensity index; S4.2 Based on the basic sample training set, the geographic spatial position of each sample is used as an index, and the exponential feature image of the image area with a neighborhood pixel size of [64,64] is extracted as a sample to construct a CNN sample training set; S4.3 The landscape feature index product of channel 1 to channel 25 in the CNN sample training set is used as x, and the road distribution data is used as y road Train the multi-spatial scale U-Net model, evaluate the model training effect by calculating the model training accuracy, and use the MSE with small R 2 Higher models generate an intelligent road strength index; S4.4 takes the landscape feature index product of channel 1 to channel 25 in the CNN sample training set as x and the building distribution data as y built Train the multi-spatial scale U-Net model, evaluate the model training effect by calculating the model training accuracy, and use the MSE with small R 2 Higher models generate smart city building index; The mathematical relationship of the multi-spatial scale U-Net model in S4.3 is: Where m CNN2road represents the multi-spatial-scale U-Net model for generating intelligent road strength index, represents the intelligent road strength index generated by the U-Net model, and x represents the data in the CNN sample training set; The mathematical relationship of the S4.4 multi-spatial scale U-Net model is: Where m CNN2built represents a multi-spatial-scale U-Net model for generating smart city building index, represents the smart city building index generated by the U-Net model, and x represents the data in the CNN sample training set.

7. The method for intelligent downscaling seasonal characterization and analysis of land surface temperature based on GeoAI machine learning according to claim 1 is characterized in that The step S5 specifically includes: S5.1 By studying the pixel index coordinates of valid pixels in the region, the smart road intensity index and smart city building index with consistent geographical locations are superimposed on the basic data set to form a comprehensive data set; S5.2 extracts the corresponding smart city building index and smart road intensity index in the comprehensive data set through the geographic location index of the basic training set to generate a comprehensive training set to generate a daily surface temperature product with a complete spatiotemporal sequence.

8. The method for intelligent downscaling seasonal characterization and analysis of land surface temperature based on GeoAI machine learning according to claim 1 is characterized in that The step S6 specifically includes: S6.1 Take the landscape characteristic index product of channels 1-30 in the comprehensive sample training set as X and the surface temperature as y LST Train the random forest model; the mathematical relationship of the random forest model is: Where m RF2LST represents the random forest model for reconstructing land surface temperature, represents the surface temperature data reconstructed by the random forest model, and X represents the landscape characteristic index product in the comprehensive sample training set; S6.2 Use the trained random forest model to reconstruct all pixels on any valid date of surface temperature in the study area to generate a daily surface temperature product with a complete spatiotemporal sequence.

9. The method for intelligent downscaling seasonal characterization and analysis of land surface temperature based on GeoAI machine learning according to claim 1 is characterized in that The step S7 specifically includes: S7.1 For all reconstructed surface temperature data in the same season, the surface temperature data for that season are obtained by weighting the proportion of missing pixel values ​​due to cloud cover before interpolation; S7.2 normalizes the surface temperature data of the four seasons in turn; S7.3 Calculate the local Moran index of the normalized surface temperature data to obtain the seasonal variation index of the surface temperature; the formula for normalizing the four seasons is: Among them, LST k Represents the heat island index for a particular season, Represents the minimum value of the water body in the study area on that day, represents the 95th percentile value of the surface temperature of this day and night. Represents the normalized seasonal heat island index.

10. The method for intelligent downscaling seasonal characterization and analysis of land surface temperature based on GeoAI machine learning according to claim 1 is characterized in that The step S8 specifically includes: S8.1 uses the features in the comprehensive training set as x, and uses the corresponding pixel position index of x to extract the heat island index of spring, summer, autumn and winter as Construct feature seasonality evaluation training set; S8.2 trains random forest models for heat island indices in different seasons and analyzes the contribution of various landscape factors to the heat island in different seasons; S8.3 Sort the feature contributions and screen the important landscape features associated with the urban heat island effect. Then, based on the physical meaning of the urban space mapped by the features, analyze and guide potential urban space optimization plans.

Images