CN111914896A - Urban local climate zone classification method based on multi-source data - Google Patents

Urban local climate zone classification method based on multi-source data Download PDF

Info

Publication number
CN111914896A
CN111914896A CN202010606871.3A CN202010606871A CN111914896A CN 111914896 A CN111914896 A CN 111914896A CN 202010606871 A CN202010606871 A CN 202010606871A CN 111914896 A CN111914896 A CN 111914896A
Authority
CN
China
Prior art keywords
surface fraction
building
local
lcz
source data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202010606871.3A
Other languages
Chinese (zh)
Inventor
杨英宝
胡佳
潘鑫
章勇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hohai University HHU
Original Assignee
Hohai University HHU
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hohai University HHU filed Critical Hohai University HHU
Priority to CN202010606871.3A priority Critical patent/CN111914896A/en
Publication of CN111914896A publication Critical patent/CN111914896A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/243—Classification techniques relating to the number of classes
    • G06F18/24323—Tree-organised classifiers
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/25—Fusion techniques

Landscapes

  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a city local climate zone classification method based on multi-source data, which comprises the following steps: acquiring a top-grade first image and urban 3D building data in a research area range, and preprocessing the images; extracting parameters for local climate zoning, including building height, building surface fraction, normalized vegetation index, vegetation coverage, permeable surface fraction, water surface fraction and impermeable surface fraction, establishing regular grids of different scales, and extracting LCZ zoning parameters of corresponding spatial scales respectively based on the regular grids of different scales; step three, establishing a random forest classification model; step four, based on the established random forest classification model, performing local climate zone classification of different scales; and fifthly, selecting the optimal local climate partition scale through visual interpretation and quantitative precision evaluation. The invention effectively combines the advantages of two local climate zoning methods, and improves the precision and the high efficiency of the local climate zoning.

Description

Urban local climate zone classification method based on multi-source data
Technical Field
The invention belongs to a climate zone classification method, and particularly relates to a city local climate zone classification method based on multi-source data.
Background
Rapid urbanization leads to changes in ground cover, city geometry and city structure, thus increasing city heat absorption and changing local climate. The urban heat island is one of the consequences caused by urbanization, and refers to a phenomenon that the urban temperature is higher than the suburban temperature. The traditional urban heat island strength research mainly focuses on urban and rural temperature difference calculation, temperature difference calculation between a permeable surface and an impermeable surface, or temperature difference calculation between a vegetation area and a non-vegetation area, and the traditional urban heat island strength research has the defects of lack of unified standards and strong subjectivity. Therefore, if a global uniform urban land division system can be established, quantitative calculation research of the urban heat island can be effectively realized. The concept provides a standard framework for the classification of urban ground surfaces according to urban morphology and ground surface attributes, and is a new trend of the future urban heat island phenomenon research.
The premise of urban heat island research based on the LCZ concept is correct classification of LCZ, and the two methods used most currently are the WUDAPT method (world city database and access portal tool) and the GIS-based method. The WUDAPPT method is a remote sensing-based method, and utilizes Landsat data and training samples selected from Google Earth to perform LCZ classification based on a random forest classifier. The method is widely applied due to high efficiency, but the input data of the method is intermediate-resolution Landsat data, and the input data lack building height information of three-dimensional forms of urban buildings, so that the method is not high in classification precision and is proved to be more suitable for local climate subareas of large areas. On the contrary, the GIS-based method performs LCZ classification by using parameters extracted from various high-precision urban GIS vector data and a threshold value division method, although the LCZ classification precision is high, a large amount of urban GIS data are required to be input, and the data are usually difficult to acquire because many urban GIS data are not open to the public. Secondly, compared with a random forest classifier of the WUDAPT method, the method based on the GIS has certain subjectivity depending on threshold division. In addition, many parameters for LCZ classification are difficult to quantitatively calculate, such as artificial heat output, so that more parameters need to be proposed for assisting LCZ classification, and the accuracy and efficiency of the existing local climate partitioning are further improved.
Disclosure of Invention
The purpose of the invention is as follows: in order to overcome the defects in the prior art, the invention aims to provide a multisource data-based city local climate zone classification method capable of realizing high-efficiency and high-precision LCZ (lower control panel) zones.
The technical scheme is as follows: the invention relates to a city local climate zone classification method based on multi-source data, which comprises the following steps:
acquiring a top-grade first image and urban 3D building data in a research area range, and preprocessing the images;
extracting parameters for local climate zoning, including building height, building surface fraction, normalized vegetation index, vegetation coverage, permeable surface fraction, water surface fraction and impermeable surface fraction, establishing regular grids of different scales, and extracting LCZ zoning parameters of corresponding spatial scales respectively based on the regular grids of different scales;
step three, establishing a random forest classification model;
step four, based on the established random forest classification model, performing local climate zone classification of different scales;
and fifthly, selecting the optimal local climate partition scale through visual interpretation and quantitative precision evaluation.
Further, the preprocessing in the first step comprises the radiation correction, the geometric correction, the image fusion and the image cutting of the high-resolution first image, the radiation correction and the geometric correction are carried out on the multispectral and panchromatic images of the high-resolution first image, the image fusion and the image cutting are carried out on the two images, and the 3D building vector data of the city are cut to the size of a research area.
Further, in the second step, the building height, the building surface score, the normalized vegetation index, the vegetation coverage, the permeable surface score, the water surface score and the impermeable surface score are respectively obtained through the following formulas:
Figure RE-GDA0002641849110000021
Figure RE-GDA0002641849110000022
Figure RE-GDA0002641849110000023
Figure RE-GDA0002641849110000024
Figure RE-GDA0002641849110000025
Figure RE-GDA0002641849110000026
ISF＝1-BSF-PSF-WSF
where BH is the building height, n is the number of buildings of an LCZ grid, BSiIs the floor area of a building, BHiIs the building height of a building; BSF is the building surface fraction, namely the building density, and is a parameter for judging the density or sparseness of a building area, and the larger BSF is, the more dense the building is, the higher the building density of the climate area is; ssiteIs the total area of an LCZ grid; NDVI is the normalized vegetation index, ρNIRIs the reflectivity, p, of the near infrared bandRIs the reflectance of the red band; fcIs the vegetation coverage, NDVI is the NDVI value of each pixel, NDVImaxIs the maximum value of NDVI in the whole image, NDVIminDivide to the minimum value of NDVI in the whole image; PSF is the permeable surface fraction, permeable surface mainly comprises vegetation and water, the invention considers water as single ground object type, therefore, permeable waterIn the area of research, only vegetation areas were considered. S _ per is the sum of the areas of the regions with NDVI values greater than 0.3; WSF is the water surface fraction, and S _ water is the sum of all water areas in the grid unit; ISF is the moisture impervious surface fraction.
In the third step, the random forest classification model comprises sample attributes, the number of classification trees and the number of characteristics at each node, the sample attributes are parameters extracted in the second step and used for LCZ classification, and the number of the classification trees and the value of the number of the characteristics at each node are changed in a circulating mode. And calculating a decision coefficient and a root mean square error through the test samples, evaluating the precision of different models, and selecting the group of parameters with the highest precision to establish a random forest classification model. The calculation formula for determining the coefficients and the root mean square error is as follows:
Figure RE-GDA0002641849110000031
Figure RE-GDA0002641849110000032
wherein R is2Is the coefficient of determination, LCZi' is the result of the partitioning of the test, LCZiIs the original partitioning result, LCZavgIs the average of the original partition results. m is the total pixel count of the image, LCZi' is the result of the partitioning of the test, LCZiIs the original partitioning result. Determining the coefficient R2The correlation between the partitioned images after classification based on the sample data and the test data is shown, the value range is between 0 and 1, and R is2The larger the result, the closer the test result is to the original partition result, the better the classification effect is, otherwise, the worse is. The RMSE is used to detect the consistency between the partition result of the test and the original partition result. A smaller RMSE indicates a higher degree of consistency between the two.
And in the fourth step, local climate subareas with different scales are divided based on the local climate subarea classification parameters established in the second step and the random forest model established in the third step, so that local climate subarea graphs (120-480 m) with 6 different scales are obtained.
And in the fifth step, selecting the optimal partition dimension of the local climate partition through visual interpretation and precision evaluation. Visual interpretation is to compare the zoning effect of different scales from the spatial distribution of local climate zones of different scales, and the accuracy evaluation is to quantitatively evaluate through overall accuracy and Kappa coefficient.
Has the advantages that: compared with the prior art, the invention has the following remarkable characteristics:
1. the advantages and the limitations of two common local climate zoning methods at present are effectively combined, and the precision and the efficiency of the local climate zoning are improved from two aspects: on one hand, the high-resolution first-order image and high-precision urban GIS data representing urban three-dimensional building forms are used, so that the precision of local climate zoning is fundamentally improved; on the other hand, a random forest classifier is used, so that the subjectivity and uncertainty caused by a threshold value method are overcome for a classification result, and the classification process is more efficient;
2. aiming at the problem that parameters which are more used for local climate subareas at present are difficult to directly and quantitatively estimate, three new parameters are provided for assisting the local climate subareas, wherein the three new parameters are respectively a normalized vegetation index, vegetation coverage and water body surface fraction, and the three parameters are also suitable for research areas with numerous other vegetation and numerous water bodies; 3. the method has the characteristics of reproducible technical route, wide applicable objects and the like, and can provide reference for the space-time change of the urban heat island and the relief research of the urban heat island; therefore, the invention has important practical significance in both theory and practice.
Drawings
FIG. 1 is a schematic diagram of the present invention;
FIG. 2 is a plot of the location of an investigation region of the present invention;
FIG. 3 is a comparison graph of the effect of the first high-resolution images before and after fusion, wherein a is before fusion and b is after fusion;
FIG. 4 is a 3D vector data plot of a study area building of the present invention, a being a two-dimensional plan view and b being a three-dimensional perspective view;
FIG. 5 is a spatial distribution diagram of the results of regional extraction of the local climate in the research area of the present invention, where a is the building height, b is the building surface fraction, c is the vegetation index, d is the vegetation coverage, e is the permeable surface fraction, f is the impermeable surface fraction, and g is the water surface fraction;
FIG. 6 is a diagram of the evaluation results of the random forest models in the research area of the present invention, where a is the decision coefficient for different numbers of decision trees, b is the root mean square error for different numbers of decision trees, and c is the decision coefficient and root mean square error for different numbers of features;
FIG. 7 is a plot of the results of the local climate zones at different scales for the study area of the present invention, a is 120X 120, b is 180X 180, c is 240X 240, d is 270X 270, e is 360X 360, and f is 480X 480;
FIG. 8 is the accuracy of the local climate zones at different scales of the study area of the present invention.
Detailed Description
Referring to fig. 1-2, the main urban area of Nanjing city is taken as a research area, and the method specifically comprises the following steps:
firstly, acquiring a top-grade first image and urban 3D building data in a research area range, and preprocessing. The remote sensing image obtained in the example is a high-resolution one-number image of 2016, 6, month and 17 days, the image comprises a panchromatic image and a multispectral image, and the adopted preprocessing mainly comprises radiation correction, geometric correction, image fusion and image cutting. In the ENVI software, the multispectral image and the panchromatic image of the high-resolution first-order image are respectively subjected to radiation correction and geometric correction, the multispectral image and the panchromatic image are subjected to image fusion, the resolution after the fusion is 2 meters, as shown in figure 3, and the images are cut to the size of a research area range. In addition, the city 3D building vector data used in this example, which was 2016, was also clipped to the study area range size, as in fig. 4.
Step two, extracting parameters with different scales for local climate partition based on the regular grid, and the specific steps are as follows:
(1) and extracting a normalized vegetation index NDVI and a normalized water body index NDWI in ENVI software through a wave band calculation tool based on the fused high-grade first image, and extracting a vegetation area and a water body according to a threshold value method.
Normalized vegetation index:
Figure RE-GDA0002641849110000051
in the formula, ρNIRAnd ρRRespectively, the reflectivities of the near infrared and red bands.
Normalizing the water body index:
Figure RE-GDA0002641849110000052
in the formula, B2And B4The green band and the near infrared band of the top-ranked first image are respectively.
(2) Respectively converting vegetation areas, water areas and NDVI values into vector formats in ArcGIS; and establishing a regular grid (120-480 m), and respectively associating the building 3D vector data, the NDVI, the extracted vegetation area, the water body and the like. Through the field calculator tool in the attribute table, 7 parameters for the local climate zone are calculated: building Height (BH), Building Surface Fraction (BSF), water permeability fraction (PSF) and water impermeability fraction (ISF), normalized vegetation index (NDVI), vegetation coverage (F)c) And Water Surface Fraction (WSF).
Building height:
Figure RE-GDA0002641849110000053
where BH is the building height, n is the number of buildings of an LCZ grid, BSiIs the floor area of a building, BHiIs the building height of a building.
Building surface fraction:
Figure RE-GDA0002641849110000061
wherein BSF is the building surface score, SsiteIs the total area of an LCZ grid.
Normalized vegetation index:
Figure RE-GDA0002641849110000062
where NDVI is the normalized vegetation index, ρNIRIs the reflectivity, p, of the near infrared bandRIs the reflectance of the red band.
Vegetation coverage:
Figure RE-GDA0002641849110000063
wherein, FcIs the vegetation coverage, NDVI is the NDVI value of each pixel, NDVImaxIs the maximum value of NDVI in the whole image, NDVIminIs the minimum value of NDVI in the entire image.
Water permeability fraction:
Figure RE-GDA0002641849110000064
where PSF is the permeable surface fraction and S _ per is the sum of the areas of the regions where the NDVI value is greater than 0.3.
Water surface fraction:
Figure RE-GDA0002641849110000065
wherein, WSF is the water surface fraction, and S _ water is the sum of all water areas in the grid unit.
Fraction of impervious surface:
ISF＝1-BSF-PSF-WSF
wherein ISF is the impervious surface fraction calculated from the building surface fraction, the permeable surface fraction, and the water surface fraction.
(3) Establishing 6 grids (120 meters, 180 meters, 240 meters, 270 meters, 360 meters and 480 meters) with different scales, and respectively extracting corresponding LCZ partition parameters based on the grids. Finally, the parameters of the 6 scales are respectively converted into a grid format, and the 7 extracted parameters are distributed in a space manner as shown in figure 5(120 meters scale).
And step three, establishing a random forest model. And the random forest classification model comprises sample attributes, the number of classification trees and the number of features at each node, the sample attributes are parameters for LCZ classification extracted in the step two, and the number of the classification trees and the value of the number of the features at each node are changed in a circulating mode. And calculating a decision coefficient and a root mean square error through the test samples, evaluating the precision of different models, and selecting the group of parameters with the highest precision to establish a random forest classification model. Determining a coefficient:
Figure RE-GDA0002641849110000071
wherein R is2Is the coefficient of determination, LCZi' is the result of the partitioning of the test, LCZiIs the original partitioning result, LCZavgIs the average of the original partition results.
Root mean square error:
Figure RE-GDA0002641849110000072
wherein m is the total pixel count of the image, LCZi' is the result of the partitioning of the test, LCZiIs the original partitioning result.
It can be seen that when the number of decision trees is greater than 180, R2Both RMSE and RMSE tend to be substantially stable and remain around the optimal value, as in fig. 6, thus setting the number of decision trees in the random forest classification model to 180. Secondly, when the number of decision trees is fixed, it can be seen that R is when the number of selected features is 52The maximum is reached, the RMSE is minimum, and the effect of the classification model is best, so that the optimal parameters of the random forest classification model are determined.
And step four, classifying local climate areas with different scales. And based on the random forest classification model established in the last step, performing random forest classification of different scales to obtain local climate zone classification results of 6 scales, wherein the local climate zone classification results are 120 meters, 180 meters, 240 meters, 270 meters, 360 meters and 480 meters respectively.
And fifthly, selecting the optimal partition dimension of the local climate partition through visual interpretation and precision evaluation. First, the partitioning effect is compared from the local climate zone spatial distribution of different scales, as in fig. 7, it can be seen that the larger the scale, the coarser the partitioning result, and much detail information is ignored, especially at the scale of 480 m. The small water bodies in the urban areas are not separated, and the large water body areas tend to be reduced in range compared with other scales. Secondly, the accuracy of the local climate subareas of different scales is quantitatively evaluated through the overall accuracy and the Kappa coefficient, as shown in fig. 8, the classification accuracy tends to increase first and decrease later with the increase of the subarea scale, the local climate subarea accuracy is higher under the scales of 240m and 270m, and the subarea accuracy is obviously reduced after reaching a certain degree (more than or equal to 360 m). The overall precision of the local climate subareas under the scales of 240m and 270m is high, the overall precision and the Kappa coefficient are both above 92% and 0.91, and the difference between the overall precision and the Kappa coefficient is only 0.22% and 0.002% respectively when the scales of 240m and 270m are large, so that the subarea effects of the two scales are considered to be relatively close, but in order to show more detailed information, 240m is selected as the optimal scale of the local climate subareas of the research area.

Claims (10)

1. A city local climate zone classification method based on multi-source data is characterized by comprising the following steps:
acquiring a top-grade first image and urban 3D building data in a research area range, and preprocessing the images;
extracting parameters for local climate zoning, including building height, building surface fraction, normalized vegetation index, vegetation coverage, permeable surface fraction, water surface fraction and impermeable surface fraction, establishing regular grids of different scales, and extracting LCZ zoning parameters corresponding to the space scales respectively based on the regular grids of different scales;
step three, establishing a random forest classification model;
step four, based on the established random forest classification model, performing local climate zone classification of different scales;
and fifthly, selecting the optimal local climate partition scale through visual interpretation and quantitative precision evaluation.
2. The method for classifying the local urban climate zone based on the multi-source data according to claim 1, wherein the method comprises the following steps: in the first step, the preprocessing comprises the radiation correction, the geometric correction, the image fusion and the image cutting of the top-grade first image, the radiation correction and the geometric correction are carried out on the multispectral and panchromatic images of the top-grade first image, the image fusion and the image cutting are carried out on the two images, and the 3D building vector data of the city are cut to the size of a research area.
3. The method for classifying the local climatic zones of the city based on the multi-source data according to the claim 1, wherein in the second step, the building height, the building surface fraction, the normalized vegetation index, the vegetation coverage, the permeable surface fraction, the water surface fraction and the impermeable surface fraction are respectively obtained by the following formulas:
Figure FDA0002559347960000011
Figure FDA0002559347960000012
Figure FDA0002559347960000013
Figure FDA0002559347960000014
Figure FDA0002559347960000015
Figure FDA0002559347960000016
ISF＝1-BSF-PSF-WSF
where BH is the building height, n is the number of buildings of an LCZ grid, BSiIs the floor area of a building, BHiIs the building height of a building; BSF is the building surface fraction, SsiteIs the total area of an LCZ grid; NDVI is the normalized vegetation index, ρNIRIs the reflectivity, p, of the near infrared bandRIs the reflectance of the red band; fcIs the vegetation coverage, NDVI is the NDVI value of each pixel, NDVImaxIs the maximum value of NDVI in the whole image, NDVIminDivide to the minimum value of NDVI in the whole image; PSF is the permeable surface fraction, S _ per is the sum of the areas of the regions with NDVI values greater than 0.3; WSF is the water surface fraction, and S _ water is the sum of all water areas in the grid unit; ISF is the moisture impervious surface fraction.
4. The method for classifying the local urban climate zone based on the multi-source data according to claim 1, wherein the method comprises the following steps: in the third step, the random forest classification model comprises sample attributes, the number of classification trees and the number of characteristics at each node, the sample attributes are parameters extracted in the second step and used for LCZ classification, and the values of the number of classification trees and the number of characteristics at each node are changed in a circulating mode to obtain an optimal classification result.
5. The method for classifying the local urban climate zone based on the multi-source data according to claim 4, wherein the method comprises the following steps: and calculating a decision coefficient and a root mean square error through the test samples, evaluating the precision of different models, and selecting the group of parameters with the highest precision to establish a random forest classification model.
6. The method for classifying the local urban climate zone based on the multi-source data according to claim 5, wherein the calculation formula of the decision coefficient is as follows:
Figure FDA0002559347960000021
wherein R is2Is the coefficient of determination, LCZi' is the result of the partitioning of the test, LCZiIs the original partitioning result, LCZavgIs the average of the original partition results.
7. The method for classifying the local urban climate zone based on the multi-source data according to claim 5, wherein the root mean square error is calculated according to the following formula:
Figure FDA0002559347960000022
wherein m is the total pixel count of the image, LCZi' is the result of the partitioning of the test, LCZiIs the original partitioning result.
8. The method for classifying the local urban climate zone based on the multi-source data according to claim 1, wherein the method comprises the following steps: in the fourth step, the dimension is 120-480 meters.
9. The method for classifying the local urban climate zone based on the multi-source data according to claim 1, wherein the method comprises the following steps: in the fifth step, the visual interpretation is to compare the zoning effect of different scales from the spatial distribution of the local climate zones of different scales.
10. The method for classifying the local urban climate zone based on the multi-source data according to claim 1, wherein the method comprises the following steps: in the fifth step, the quantitative precision evaluation is to quantitatively evaluate the precision of the local climate subareas with different scales through the overall precision and the Kappa coefficient.
CN202010606871.3A 2020-06-29 2020-06-29 Urban local climate zone classification method based on multi-source data Pending CN111914896A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010606871.3A CN111914896A (en) 2020-06-29 2020-06-29 Urban local climate zone classification method based on multi-source data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010606871.3A CN111914896A (en) 2020-06-29 2020-06-29 Urban local climate zone classification method based on multi-source data

Publications (1)

Publication Number Publication Date
CN111914896A true CN111914896A (en) 2020-11-10

Family

ID=73226722

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010606871.3A Pending CN111914896A (en) 2020-06-29 2020-06-29 Urban local climate zone classification method based on multi-source data

Country Status (1)

Country Link
CN (1) CN111914896A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113157834A (en) * 2021-02-24 2021-07-23 中国科学院空天信息创新研究院 Drawing method and device for city local climate partition classification
CN113610708A (en) * 2021-07-28 2021-11-05 国家卫星气象中心(国家空间天气监测预警中心) Mapping method and device for passive satellite remote sensing flood information

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110990511A (en) * 2019-11-29 2020-04-10 南京信息工程大学 Local climate zone classification method considering two-dimensional and three-dimensional forms of city

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110990511A (en) * 2019-11-29 2020-04-10 南京信息工程大学 Local climate zone classification method considering two-dimensional and three-dimensional forms of city

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
JIA HU等: "Analysis of the Spatial and Temporal Variations of Land Surface Temperature Based on Local Climate Zones: A Case Study in Nanjing, China", 《IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113157834A (en) * 2021-02-24 2021-07-23 中国科学院空天信息创新研究院 Drawing method and device for city local climate partition classification
CN113157834B (en) * 2021-02-24 2023-10-24 中国科学院空天信息创新研究院 Drawing method and device for urban local climate partition classification
CN113610708A (en) * 2021-07-28 2021-11-05 国家卫星气象中心(国家空间天气监测预警中心) Mapping method and device for passive satellite remote sensing flood information
CN113610708B (en) * 2021-07-28 2023-11-17 国家卫星气象中心(国家空间天气监测预警中心) Imaging method and device for passive microwave remote sensing flood information

Similar Documents

Publication Publication Date Title
CN105677890B (en) A kind of green amount numerical map production in city and display methods
US20230039554A1 (en) Tree crown extraction method based on unmanned aerial vehicle multi-source remote sensing
CN108984803B (en) Method and system for spatializing crop yield
CN104751478A (en) Object-oriented building change detection method based on multi-feature fusion
CN108549858B (en) Quantitative evaluation method for urban heat island effect
CN106547853A (en) Forestry big data building method based on a figure
CN110263717A (en) It is a kind of incorporate streetscape image land used status determine method
TW201022708A (en) Method of change detection for building models
CN114758252B (en) Image-based distributed photovoltaic roof resource segmentation and extraction method and system
CN103364781A (en) Remote sensing data and geographical information system-based grainfield ground reference point screening method
CN110188927A (en) A kind of analysis method, device and the storage medium of urban heat island Dominated Factors
CN115564926B (en) Three-dimensional patch model construction method based on image building structure learning
CN111047695A (en) Method for extracting height spatial information and contour line of urban group
CN115238584B (en) Population distribution identification method based on multi-source big data
CN108898096B (en) High-resolution image-oriented information rapid and accurate extraction method
CN111582575B (en) Method for identifying urban thermal environment formation development leading factors under multiple space-time scales
Leonowicz et al. Automated reduction of visual complexity in small-scale relief shading
CN111914896A (en) Urban local climate zone classification method based on multi-source data
CN110070038B (en) Method for measuring and calculating urban green land rate based on urban greening coverage area
CN101980317A (en) Method for predicting traffic flow extracted by improved C-V model-based remote sensing image road network
CN116151610B (en) Method for simulating risk exposure space of disaster-bearing body on underlying surface of heterogeneous city
CN112819066A (en) Res-UNet single tree species classification technology
CN104463971A (en) Method for creating green-degree spatial arrangement curve used for evaluating urban landscaping three-dimensional layout
CN116486289A (en) Gas pipeline high-consequence area identification method driven by multi-source data and knowledge
CN116129262A (en) Cultivated land suitability evaluation method and system for suitable mechanized transformation

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination