Urban heat island area and heat source point identification method and system

Abstract

The invention discloses an urban heat island area and heat source point identification method and system, and the method comprises the steps: obtaining multi-source remote sensing data at the same transit time, carrying out the preprocessing of the multi-source remote sensing data, and obtaining the map auxiliary data of a research area; key parameter inversion is carried out on the preprocessed multi-source remote sensing data, wherein the key parameters comprise land surface temperature, vegetation indexes and impervious surface information; according to the inversed surface temperature, vegetation index and impervious surface information, calculating heat island intensity indexes, including global heat island intensity based on suburban reference temperature and local heat island intensity based on distance weighted neighborhood temperature; and carrying out heat island area division based on the global heat island intensity, and identifying a microscopic heat source point based on the local heat island intensity. According to the method, a complete technical chain of global-local collaborative heat island identification is constructed, and the defect that only'districts' can be identified and'point sources' are difficult to identify in previous researches is overcome.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of remote sensing environmental monitoring, and particularly relates to a method and system for identifying urban heat island areas and heat source points. BACKGROUND

[0002] Urban heat island effect refers to the phenomenon that the temperature in urban areas is significantly higher than that in the surrounding suburbs, which is one of the most typical local climate changes in the process of urbanization. The main causes of urban heat island effect are the replacement of natural land surface by high heat storage artificial materials such as asphalt and concrete, and the emission of huge heat generated by human activities. With the acceleration of global urbanization process, the UHI effect is increasingly intensified, and has caused a series of serious environmental and social problems. In terms of energy consumption and carbon emissions, in order to cope with high temperature, the energy consumption of air conditioners and refrigeration equipment has increased dramatically, further exacerbating energy demand and greenhouse gas emissions, forming a vicious cycle. In terms of air quality, high temperature environment can accelerate photochemical reaction, promote the generation of secondary pollutants such as ozone (O3) and fine particulate matter (PM2.5), and seriously endanger public health. In terms of public health, extreme high temperature is the main environmental risk factor leading to heat-related diseases and deaths, and poses a serious threat to vulnerable groups such as the elderly, children and outdoor workers. In terms of ecological system, it changes the hydrological cycle and phenology cycle of urban areas, affects the habitat of animals and plants, and leads to the decline of biodiversity. Therefore, it is of great practical significance to accurately and efficiently monitor and trace the urban heat island effect, in order to improve the urban living environment, promote low-carbon city construction, protect public health and develop scientific climate adaptation strategies.

[0003] Urban heat island effect is a major environmental problem in the process of urbanization. The existing technology mainly has the following limitations: first, the data source is single and the representation is not comprehensive. Most studies rely on single thermal infrared data (such as Landsat, MODIS) for land surface temperature (LST) inversion and UHI analysis, which cannot simultaneously obtain high-precision land cover information (especially at night city structure), resulting in subjective suburban benchmark temperature demarcation, insufficient impervious surface extraction accuracy, and UHI intensity calculation bias. Second, the recognition scale is single and the data precision is insufficient. Traditional methods mostly use Landsat, MODIS, Sentinel-3 and other satellites' thermal infrared bands to obtain land surface temperature through single-window / single-channel, atmospheric correction and other algorithms, and identify heat island areas through statistical methods, which cannot identify specific and local heat source points (such as single buildings, factories, transportation hubs); and the spatial resolution is limited, often difficult to depict small-scale urban heat island distribution, and only rely on a single sensor, which is difficult to distinguish the background difference of heat island. Third, the neighborhood algorithm is rough and the physical mechanism is not reflected. The existing local heat island index mostly uses moving average method with fixed window size to calculate neighborhood reference temperature, without considering distance attenuation effect, and without combining with urban physical properties such as normalized difference vegetation index (NDVI) and impervious surface index (ISA).

[0004] In summary, the current monitoring technology of urban heat island effect has many limitations such as data source shortage, single recognition scale and rough algorithm model. SUMMARY

[0005] The purpose of the present application is to overcome the shortcomings of the prior art and provide a method for identifying urban heat island areas and heat source points based on multi-source data fusion. This method aims to take advantage of the unique thermal infrared, multispectral and micro-light three sensors of SDGSAT-1 satellite, and through the calculation of global heat island intensity (Urban Heat Island Intensity, UHII) based on suburban benchmark temperature and local urban heat island intensity (Distance-Weighted Neighborhood Urban Heat Island Intensity, DW-UHII) based on distance-weighted neighborhood temperature, to realize the identification and tracing of macro heat island area to micro heat source point, and to provide scientific and reliable decision basis for urban heat environment governance.

[0006] To solve the above technical problems, the present application provides the following technical solutions: A method for identifying urban heat island areas and heat source points, comprising the following steps: Step 1: Obtain multi-source remote sensing data at the same transit time and pre-process the multi-source remote sensing data respectively, and obtain map auxiliary data of the study area; Step 2: Perform key parameter inversion on the preprocessed multi-source remote sensing data, including land surface temperature, vegetation index, and impervious surface information; Step 3: Calculate the heat island intensity index based on the retrieved surface temperature, vegetation index and impermeable surface information, including the global heat island intensity based on the suburban baseline temperature and the local heat island intensity based on the distance-weighted neighborhood temperature. Step 4: Divide the heat island region based on the global heat island intensity, and identify micro-heat source points based on the local heat island intensity.

[0007] Furthermore, the multi-source remote sensing data in step one includes at least thermal infrared data, multispectral data, and nighttime low-light data; the map-aided data includes administrative boundary vector data of the study area, road network and street block division data, and point of interest (POI) data.

[0008] Furthermore, in step two, the vegetation index is the normalized vegetation index (NDBI). The inversion of the impervious surface information includes calculating the normalized building index (NDBI) and calculating the impervious surface index (ISA) based on the NDBI and the normalized vegetation index.

[0009] Furthermore, calculating the global heat island intensity UHII based on the suburban baseline temperature includes: The nighttime low-light data is used to set thresholds to divide built-up areas into built-up areas and unbuilt-up areas; Calculate the average temperature of the non-built-up area pixels as the reference temperature for the suburbs; The global heat island intensity is calculated pixel-by-pixel based on the pixel temperature and the suburban reference temperature interpolation.

[0010] Furthermore, the formula for calculating the global heat island intensity pixel by pixel is:

[0011] in, For a pixel ( i,j) The global heat island intensity, For a pixel ( i,j) temperature, The reference temperature for the suburbs is [temperature value].

[0012] Furthermore, calculating the local heat island intensity based on distance-weighted neighborhood temperature includes: For each pixel, the search neighborhood radius is adaptively determined based on the spatial gradient of its surface temperature; Calculate the distance weight and feature similarity weight of the pixels in the search neighborhood relative to the center pixel, wherein the feature similarity weight is calculated based on the difference between the vegetation index and the impermeable surface index ISA; Based on the distance weight and the similarity weight of ground features, the neighborhood weighted average temperature is calculated; The local heat island intensity is calculated pixel-by-pixel based on the difference between the surface temperature and the neighborhood weighted average temperature.

[0013] Furthermore, the formula for calculating the intensity of the local heat island per pixel is:

[0014] in, For pixels The global heat island intensity, Pixel Surface temperature; For pixels Searching neighborhood radius The pixel-weighted average temperature within the cell; The calculation formula is as follows:

[0015] in, To search the neighborhood radius Neighborhood pixels within, For neighboring pixels Distance weights; For neighboring pixels Weights of similar features For neighboring pixels The surface temperature.

[0016] Furthermore, the formulas for calculating distance weight and similar feature weight are as follows:

[0017]

[0018] in, It refers to the neighboring pixels With the target pixel The Euclidean distance between them; It is a small constant that avoids division by zero; It is the decay index; and These are the weighting coefficients. For target pixel Normalized Difference Vegetation Index (NDVI) For neighboring pixels Normalized Difference Vegetation Index (NDVI) For target pixel Impermeability index, For neighboring pixels Impermeability index.

[0019] Furthermore, the identification of the microscopic heat source points is achieved by spatially overlaying and analyzing the local heat island intensity raster layer and the point of interest (POI) data, and then selecting POI points whose local heat island intensity values ​​are greater than a set threshold as heat source points.

[0020] On the other hand, the present invention provides a system for identifying urban heat island zones and heat source points, comprising: Data acquisition module: It is used to acquire multi-source remote sensing data at the same transit time and preprocess the multi-source remote sensing data respectively, while acquiring map auxiliary data of the study area; Parameter inversion module: It is used to invert key parameters of the preprocessed multi-source remote sensing data, including land surface temperature, vegetation index and impervious surface information. Heat Island Intensity Calculation Module: It is used to calculate heat island intensity indices based on the inverted surface temperature, vegetation index and impervious surface information, including global heat island intensity based on suburban baseline temperature and local heat island intensity based on distance-weighted neighborhood temperature. The result generation module is used to divide the heat island area based on the global heat island intensity and to identify micro heat source points based on the local heat island intensity.

[0021] Compared with the prior art, the present invention has the following beneficial effects: This invention is the first to systematically and collaboratively utilize data from three sensors (TIS, GIU, and MII) of the SDGSAT-1, organically combining thermal anomalies, vegetation cover, and nighttime urban structure information. It achieves integrated extraction of LST, vegetation index, and built-up area mask, compensating for the limitations of single sensors and improving the spatial accuracy and semantic interpretability of heat island identification. Compared to traditional fixed-window mean methods, this method employs a gradient-adaptive multi-scale window, taking into account both local details and the overall background. It integrates NDVI / ISA to form similar feature weights, avoiding background temperature interference from heterogeneous features, and introduces a distance attenuation function, which better aligns with the physical process of heat diffusion. This invention constructs a complete technical chain for global-local collaborative heat island identification, overcoming the limitation of previous studies that could only identify "areas" but not "point sources." It achieves a leap from heat island phenomenon perception to heat source attribution, generating a heat source traceability list and greatly enhancing the practical value of the technology. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0023] Figure 1This is a flowchart illustrating the method implementation of an embodiment of the present invention; Figure 2 This is a thermal map of the surface temperature of City W, as described in an embodiment of the present invention. Figure 3 This is an embodiment of the present invention: an LST grid overlaid street vector map. Figure 4 This is a histogram of the average surface temperature of a street in City W, according to an embodiment of the present invention. Figure 5 This is a grading diagram of the average surface temperature of a street in City W, according to an embodiment of the present invention. Figure 6 This is the extraction process of the impermeable surface mask in an embodiment of the present invention; Figure 7 This is a schematic diagram illustrating the calculation results of the global urban heat island intensity based on the suburban baseline temperature and the local urban heat island intensity based on the distance-weighted neighborhood temperature, according to an embodiment of the present invention. Figure 8 This is a heat island intensity level distribution map of City W, according to an embodiment of the present invention. Figure 9 This is a schematic diagram showing the area distribution of different heat island levels according to an embodiment of the present invention; Figure 10 This is a schematic diagram of the heat source identification process in an embodiment of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0025] Example 1 The following description, in conjunction with the accompanying drawings, further illustrates this embodiment.

[0026] like Figure 1 As shown, this embodiment provides a method for identifying urban heat island zones and heat source points, including the following steps: Step 1: Acquire multi-source remote sensing data at the same transit time and preprocess the multi-source remote sensing data respectively, while acquiring map auxiliary data of the study area; Step 2: Perform key parameter inversion on the preprocessed multi-source remote sensing data, including land surface temperature, vegetation index, and impervious surface information; Step 3: Calculate the heat island intensity index based on the retrieved surface temperature, vegetation index and impermeable surface information, including the global heat island intensity based on the suburban baseline temperature and the local heat island intensity based on the distance-weighted neighborhood temperature. Step 4: Divide the heat island region based on the global heat island intensity, and identify micro-heat source points based on the local heat island intensity.

[0027] In this embodiment, thermal infrared (TIS), multispectral (GIU), and low-light (MII) datasets from the same period were downloaded from the SDGSAT-1 data open system. All data were imaged on May 14, 2024. Simultaneously, data quality was carefully considered, selecting products with cloud cover less than 5%. For TIS products, summer nighttime hours were preferred, as the thermal radiation characteristics of ground features are more pronounced at this time, helping to distinguish temperature differences between urban and suburban areas. The study area selected in this embodiment is City W, and the TIS product imaging time was 2:07 AM on May 14, 2024. The data quality met the experimental requirements. The following detailed description of the invention, using the example dataset, further illustrates this invention: In step one, the map auxiliary data includes administrative boundary vector data of the study area, road network and street division data, and points of interest (POIs) from the Amap (Gaode Maps) platform. Preprocessing operations included radiometric calibration and atmospheric correction of TIS and GIU data to obtain surface radiance and reflectance data; radiometric calibration and high dynamic range (HDR) fusion processing of MII low-light data to enhance image signal-to-noise ratio and dynamic range; processing all data to the same geographic coordinate system and spatial resolution to achieve accurate geometric registration; and mosaicking and cropping the data to the study area. The radiometric calibration formula is as follows:

[0028] In the formula L Radiance; DN For pixel digital values; Gain (Gain value) ,Bias (Bias value) is the calibration parameter.

[0029] The scaling factors are shown in Table 1.

[0030] Table 1. Radiometric Calibration Coefficients of SDGSAT-1 Satellite

[0031] In step two, the preprocessed image from step one is subjected to NDVI (Normalized Difference Vegetation Index), NDBI (Normalized Difference Built-up Index), ISA (Impervious Surface Area), and LST (Land Surface Temperature), and the reflectance of the corresponding wavelength is substituted into the formula.

[0032]

[0033] In the formula, NIR This refers to the reflectance in the near-infrared band of SDGSAT-1 multispectral data (the same applies below). RED The reflectivity refers to the reflectivity in the red light band.

[0034]

[0035] In the formula, SWIR It refers to the reflectance of the short-wave infrared band in SDGSAT-1 multispectral data.

[0036] Calculated NDVI and NDBI get ISA The calculation formula is as follows:

[0037] Range of values ​​[ [1,1], the higher the value, the greater the degree of impermeability of the ground surface.

[0038] LST can be obtained by Planck's equation and single-window algorithm. The B3 band (11.72 μm) is located in the central region of the atmospheric window (10-13 μm), and is least affected by water vapor. The B3 band has the best radiometric calibration performance, reaching 0.361 K. It is recommended to use the single-window algorithm to retrieve the surface temperature.

[0039] Planck's equation converts radiance into the satellite's effective on-orbit temperature or brightness temperature, and the calculation formula is as follows:

[0040] In the formula, T Light temperature (K) h It is Planck's constant (6.626 × 10⁻⁶). -34 J·s), c It is the speed of light (2.9979 × 10⁻⁶). 8 m / s), kIt is the Boltzmann constant (1.3806 × 10⁻⁶). -23 J / K), λ It is the center wavelength (μm). L wavelength Spectral radiance (W / m) at 2 / sr / μm).

[0041] The single-window algorithm is used to invert the LST, and the calculation formula is as follows:

[0042] T The brightness temperature (K) retrieved from the satellite; λ It is the band equivalent center wavelength (µm). =14388μm K is a combination constant of Planck's constant and the wavelength factor; It is the surface emissivity.

[0043] like Figure 2 As shown; the LST raster map is overlaid with the vector boundaries of the administrative districts and blocks of City W, as shown. Figure 3 As shown; the average surface temperature of each block was calculated using the "zonal statistics" tool in ArcGIS Pro. The histogram of average surface temperature statistics for each block in City W is shown below. Figure 4 As shown, KMeans clustering was used to cluster the average surface temperature of each block (an unsupervised classification algorithm used to automatically divide samples into K clusters, where samples within a cluster are similar and samples between clusters are highly different). The clustering results are as follows: 1078 blocks with low temperatures (20.3℃~27.1℃); 2323 blocks with relatively low temperatures (27.1℃~28.7℃); 2349 blocks with relatively high temperatures (28.7℃~30.3℃); and 831 blocks with high temperatures (30.3℃~34.4℃). The spatial distribution of the clustering results is shown below. Figure 5 As shown.

[0044] The LST inversion value of surface temperature in W City ranges from 9.06 to 44.12℃, with an average temperature of 28.61℃. High-temperature neighborhoods are mainly distributed in the core urban area within the second ring road, and there are a large number of neighborhoods with temperatures ranging from 27℃ to 30℃.

[0045] The built-up area of ​​City W in 2024 was obtained from the statistical yearbook as 925.97 square kilometers. Based on the cumulative summation of pixel areas at different brightness values, the extraction threshold was determined to be 2.815 × 10⁻⁴. In the GIS platform, the raster is converted to polygons, the area of ​​the polygon features is calculated, and field fusion is performed based on the brightness value. Polygon features are then filtered by attribute according to the extraction threshold and fused to obtain an impermeable surface mask. The specific operation process is as follows: Figure 6 As shown.

[0046] In step three, the suburban baseline temperature refers to the average surface temperature of all "non-built-up area" pixels extracted using the impermeable mask generated in step two, which is then used as the suburban baseline temperature for that area. Then, the UHII is calculated pixel-by-pixel using the following formula:

[0047] In the formula, For pixels (i,j) The global heat island intensity, For pixels (i,j) temperature, The reference temperature for the suburbs is [temperature value].

[0048] Furthermore, the DW-UHII calculation process in step three is as follows: (1) Calculate the temperature gradient; Applying Sobel filtering to LST yields the temperature gradient magnitude:

[0049] In the formula, It refers to the position in the image. The gradient value at the given location; x and y represent the horizontal and vertical directions, respectively.

[0050] (2) Determine the threshold interval and select the search neighborhood radius In high gradient areas (where temperature changes rapidly, such as urban-rural fringe areas or areas near rivers), small radii are used to avoid interference caused by crossing different land features; in low gradient areas (where temperature is uniform, such as large impermeable areas or vegetated areas), large radii are used to improve statistical stability. Utilizing the distribution characteristics of the gradient, two thresholds are set: g1 (60th quantile) and g2 (85th quantile). The LST range obtained from inversion varies greatly across different cities, images, and seasons. Directly setting fixed thresholds can lead to threshold failure in certain situations. Using gradient quantile thresholds is a normalized adaptive boundary method, determined solely by the gradient distribution pattern, and has better versatility. g1 (60th quantile) is slightly higher than the median (50%), ensuring that at least half the area is considered a temperature-stable zone, enhancing the smoothness of background temperature estimation; g2 (85th quantile) corresponds to the upper tail distribution of the gradient (the first 15% of pixels), concentrated in the boundary areas of drastic temperature changes, avoiding blurring caused by large windows "averaging across land features".

[0051] The radius selection rule is as follows:

[0052] In the formula, To search for the neighborhood radius, , For search neighborhood radius values ​​of different gradient ranges: =10 pixels (300m), corresponding to a city block; =20 pixels (600m), corresponding to urban functional zones; =30 pixels (900m), corresponding to urban clusters.

[0053] (3) Calculate the distance weight Considering that the influence of neighborhood temperature on target pixel temperature decreases with spatial distance, the neighborhood pixel temperature is inversely distance-weighted, and the calculation formula is as follows:

[0054] In the formula, For neighboring pixels Distance weights, It refers to the neighboring pixels With the target pixel The Euclidean distance between them; It is a small constant that avoids division by zero; It is the decay exponent, which is set to 2 here.

[0055] (4) Introduce similarity weights for land features Since temperatures are more comparable between identical or similar land features, and the temperature transfer effect between different types of land features is weaker, this invention introduces indices such as NDVI and ISA to measure the structural similarity of land features between pixels, thereby adjusting the contribution of neighborhood temperature to the central pixel. In this way, the estimation of background temperature is more consistent with physical processes and underlying surface characteristics. The formula for calculating the weight of similar land features is as follows:

[0056] In the formula, For neighboring pixels Weights of similar features; and Here, the weighting coefficients are used to control the contribution of different indices to similarity; all are set to 1. For target pixel Normalized Difference Vegetation Index (NDVI) For neighboring pixels Normalized Difference Vegetation Index (NDVI) For target pixel Impermeability index, For neighboring pixels Impermeability index.

[0057] (5) Calculate DW-UHII index per pixel The intensity index of the local heat island, taking into account distance attenuation, is calculated on a per-pixel basis. The calculation formula is as follows:

[0058] In the formula, D For pixels The intensity of the local heat island Pixel Surface temperature; For pixels Searching neighborhood radius The pixel-weighted average temperature within the cell; The calculation formula is as follows:

[0059] in, To search the neighborhood radius Neighborhood pixels within, For neighboring pixels Distance weights; For neighboring pixels Weights of similar features For neighboring pixels The surface temperature.

[0060] In this embodiment, based on the LST inversion results calculated in step two and the impermeable surface mask, the average temperature of suburban pixels is obtained as 25.92℃. The suburban area is the city area of ​​W excluding the impermeable surface mask. Then, the global UHII and local DW-UHII index values ​​are calculated pixel by pixel. The index value is the temperature difference, so the unit is either ℃ or K. The results are as follows: Figure 7 As shown, the UHII range is -16.9 to 18.2℃ (K), and the DW-UHII range is -12.9 to 9.8℃ (K).

[0061] Furthermore, step four, heat island identification, involves visualizing and mapping the global UHII results obtained in step three, classifying them according to intensity values, and identifying contiguous heat island regions. Based on domestic and international standards, a heat island intensity classification system is established: UHII < -5℃ / K is an extremely cold region; -5℃ / K ≤ UHII < -2℃ / K is a cold region; -2℃ / K ≤ UHII < 2℃ / K is a neutral region; 2℃ / K ≤ UHII < 5℃ / K is a heat island region; and UHII ≥ 5℃ / K is a strong heat island region.

[0062] Microscopic heat source extraction refers to the spatial overlay analysis of the local DW-UHII raster layer and POI data obtained in step three. The DW-UHII raster values ​​are added to the POI point attributes using the "Extract Multivalues ​​to Point" tool in ArcGIS Pro. POI points with DW-UHII > 2K are identified as heat sources through attribute filtering. Finally, statistical analysis is performed to obtain a list of heat sources (including the name, type, address, latitude and longitude, and DW-UHII value of each heat source).

[0063] In this embodiment, the UHII calculated according to Step 3 is used for heat island area division. The high-value area of UHII is the area with high heat island intensity, that is, the temperature in this area is significantly higher than the suburban reference temperature. According to the intensity value, level division is carried out to identify the contiguous heat island areas. The results are as Figure 8 and Figure 9 shown.

[0064] The DW-UHII values calculated according to Step 3 are concentrated in the range of -2K to 2K. This part of the pixels accounts for 98.64% of the total area of City W. Its spatial distribution does not show the characteristic of high-value contiguous distribution like the global UHII value, but prominently shows the distribution of local heat source points (scattered pixels with high values). On the basis of the calculation of DW-UHII, POI data (obtained by crawling through the Gaode Map API, with a total of 38,416 valid data) is superimposed, and the raster values are extracted to the POI points; the POI points with DW-UHII > 2K are selected as heat source points, with a total of 601. Among them, the proportion of transportation service facilities is 54%, the proportion of residential areas is 25%, the proportion of industrial parks is 14%, and the proportion of business districts is 7%. The heat source point identification process is as Figure 10 shown.

[0065] These 601 significant local heat source points represent areas with abnormally prominent temperatures above the urban background temperature, which are typical "heat island cores". The proportion of transportation service facilities is the highest, exceeding half. This type of POI includes surface parking lots, bus hubs, logistics yards, gas stations, etc., and usually has the following characteristics: large-area hardening, impervious coverage (concrete, asphalt), vehicle emissions and mechanical heat sources, lack of vegetation coverage and cooling buffers; and due to the specific heat capacity and thermal conductivity characteristics of materials such as asphalt and concrete, it stores heat rapidly during the day and releases heat slowly at night. According to the heat source list generated in this embodiment, the transportation facility type POIs with high DW-UHII values are all supporting parking lots for large factories or industrial parks. 25% of the residential area heat source points are mostly high-density communities with dense buildings and high plot ratios, forming "heat retention"; and the daily energy consumption of residents (such as air conditioners, cooking, lighting, etc.) increases local heat emissions. 14% of the industrial park heat source points and 7% of the business district heat source points are mainly related to business types, factory building structures and scales, etc.

[0066] Embodiment 2 This embodiment provides an identification system for urban heat island areas and heat source points, including: Data acquisition module: It is used to acquire multi-source remote sensing data at the same transit time and preprocess the multi-source remote sensing data respectively, and at the same time acquire map auxiliary data of the research area; Parameter inversion module: It is used to invert key parameters of the preprocessed multi-source remote sensing data, including land surface temperature, vegetation index and impervious surface information. Heat Island Intensity Calculation Module: It is used to calculate heat island intensity indices based on the inverted surface temperature, vegetation index and impervious surface information, including global heat island intensity based on suburban baseline temperature and local heat island intensity based on distance-weighted neighborhood temperature. The result generation module is used to divide the heat island area based on the global heat island intensity and to identify micro heat source points based on the local heat island intensity.

[0067] It should be understood that any parts not described in detail in this specification belong to the prior art.

[0068] It should be understood that the above description of the preferred embodiments is quite detailed, but this should not be construed as limiting the scope of protection of this invention. It is neither necessary nor possible to exhaustively describe all possible implementations. Those skilled in the art, guided by this invention, can make substitutions or modifications without departing from the scope of the claims, all of which fall within the scope of protection of this invention. The scope of protection of this invention should be determined by the appended claims.

Claims

1. A method for identifying urban heat island zones and heat source points, characterized in that, Includes the following steps: Step 1: Acquire multi-source remote sensing data at the same transit time and preprocess the multi-source remote sensing data respectively, while acquiring map auxiliary data of the study area; Step 2: Perform key parameter inversion on the preprocessed multi-source remote sensing data, including land surface temperature, vegetation index, and impervious surface information; Step 3: Calculate the heat island intensity index based on the retrieved surface temperature, vegetation index and impermeable surface information, including the global heat island intensity based on the suburban baseline temperature and the local heat island intensity based on the distance-weighted neighborhood temperature. Step 4: Divide the heat island region based on the global heat island intensity, and identify micro-heat source points based on the local heat island intensity.

2. The method for identifying urban heat island zones and heat source points according to claim 1, characterized in that, The multi-source remote sensing data in step one includes at least thermal infrared data, multispectral data, and nighttime low-light data; the map-aided data includes administrative boundary vector data of the study area, road network and street block division data, and point of interest (POI) data.

3. The method for identifying urban heat island zones and heat source points according to claim 1, characterized in that, In step two, the vegetation index is the normalized vegetation index. The inversion of the impervious surface information includes calculating the normalized building index NDBI and calculating the impervious surface index ISA based on the normalized building index NDBI and the normalized vegetation index.

4. The method for identifying urban heat island zones and heat source points according to claim 1, characterized in that, Calculating the global heat island intensity (UHII) based on a suburban baseline temperature includes: The nighttime low-light data is used to set thresholds to divide built-up areas into built-up areas and unbuilt-up areas; Calculate the average temperature of the non-built-up area pixels as the reference temperature for the suburbs; The global heat island intensity is calculated pixel-by-pixel based on the pixel temperature and the suburban reference temperature interpolation.

5. The method for identifying urban heat island zones and heat source points according to claim 4, characterized in that, The formula for calculating the global heat island intensity pixel by pixel is: in, For a pixel ( i,j) The global heat island intensity, For a pixel ( i,j) temperature, The reference temperature for the suburbs is [temperature value].

6. The method for identifying urban heat island zones and heat source points according to claim 1, characterized in that, Calculating the intensity of a local heat island based on distance-weighted neighborhood temperature includes: For each pixel, the search neighborhood radius is adaptively determined based on the spatial gradient of its surface temperature; Calculate the distance weight and feature similarity weight of the pixels in the search neighborhood relative to the center pixel, wherein the feature similarity weight is calculated based on the difference between the vegetation index and the impermeable surface index ISA; Based on the distance weight and the similarity weight of ground features, the neighborhood weighted average temperature is calculated; The local heat island intensity is calculated pixel-by-pixel based on the difference between the surface temperature and the neighborhood weighted average temperature.

7. The method for identifying urban heat island zones and heat source points according to claim 6, characterized in that, The formula for calculating the intensity of the local heat island per pixel is: in, For pixels The global heat island intensity, Pixel Surface temperature; For pixels Searching neighborhood radius The pixel-weighted average temperature within the cell; The calculation formula is as follows: in, To search the neighborhood radius Neighborhood pixels within, For neighboring pixels Distance weights; For neighboring pixels Weights of similar features For neighboring pixels The surface temperature.

8. The method for identifying urban heat island zones and heat source points according to claim 7, characterized in that, The formulas for calculating distance weight and similar feature weight are as follows: in, It refers to the neighboring pixels With the target pixel The Euclidean distance between them; It is a small constant that avoids division by zero; It is the decay index; and These are the weighting coefficients. For target pixel Normalized Difference Vegetation Index (NDVI) For neighboring pixels Normalized Difference Vegetation Index (NDVI) For target pixel Impermeability index, For neighboring pixels Impermeability index.

9. The method for identifying urban heat island zones and heat source points according to claim 7, characterized in that, Identifying the microscopic heat source points involves spatially overlaying and analyzing the local heat island intensity raster layer with the point of interest (POI) data, and then selecting POI points whose local heat island intensity values ​​are greater than a set threshold as heat source points.

10. A system for identifying urban heat island zones and heat source points, characterized in that, include: Data acquisition module: It is used to acquire multi-source remote sensing data at the same transit time and preprocess the multi-source remote sensing data respectively, while acquiring map auxiliary data of the study area; Parameter inversion module: It is used to invert key parameters of the preprocessed multi-source remote sensing data, including land surface temperature, vegetation index and impervious surface information. Heat Island Intensity Calculation Module: It is used to calculate heat island intensity indices based on the inverted surface temperature, vegetation index and impervious surface information, including global heat island intensity based on suburban baseline temperature and local heat island intensity based on distance-weighted neighborhood temperature. Result generation module: It is used to divide the heat island area based on the global heat island intensity and to identify micro heat source points based on the local heat island intensity; The urban heat island zone and heat source identification system is used to perform the steps in the urban heat island zone and heat source identification method according to any one of claims 1-9.

Images