Satellite remote sensing recognition method and system for provincial rural black and odorous water body
Through multi-source satellite remote sensing data processing and water body index model, the automatic identification and extraction of rural black and odorous water bodies is realized, solving the problem of low monitoring efficiency of rural black and odorous water bodies, and improving monitoring efficiency and management effect.
Patent Information
- Application Number
- CN202410965845.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-18
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-07-18
AI Technical Summary
The monitoring efficiency of rural black and odorous water bodies is low, time-consuming and labor-intensive, and easy to miss inspections, and cannot carry out efficient dynamic monitoring, which affects the unified supervision effect of provincial environmental management departments.
Using multi-source satellite remote sensing data, by acquiring and processing satellite images, digital elevation models and urban built-in area data, a water extraction model is constructed using near-infrared single-band index, normalized water index and shadow water index, and a suitable extraction threshold is set. , extract and crop water distribution data, combine BOI and NDBWI index to perform raster image algebraic operation, realize automatic identification and extraction of black and odorous water bodies.
It greatly improves the inspection efficiency of rural black and odorous water bodies, saves manpower and material resources, and realizes full coverage and periodic monitoring of black and odorous water bodies within the province, providing spatial distribution data of black and odorous water bodies points for rural water environment management. Easy to conduct on-site verification and governance.
Smart Images

Figure CN118968328A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of environmental remote sensing, and in particular to a method and system for identifying rural black and odorous water bodies in a provincial scope via satellite remote sensing. Background Art
[0002] Black and odorous water pollution is a serious water environment problem that affects the quality of life of urban and rural residents, and directly affects the residents' sense of happiness in life. At present, the treatment of black and odorous water bodies has become the main content of environmental protection work. In the actual work of water environment management, black and odorous water bodies are divided into urban black and odorous water bodies and rural black and odorous water bodies, which are respectively under the responsibility of different government management departments. The treatment of urban black and odorous water bodies started earlier, and the scope of investigation is smaller. Manual on-site verification and other means can be adopted, supplemented by water quality testing, satellite and drone remote sensing and other means for accurate identification and positioning. Compared with urban built-up areas, rural areas are large in scope, water bodies are scattered and numerous, and investigation is more difficult.
[0003] In the management practice of most areas, the monitoring of black and odorous water bodies in rural areas is still mainly carried out by manual investigation, supplemented by certain water quality monitoring methods, measuring parameters such as ammonia nitrogen, dissolved oxygen, and transparency of water bodies, and referring to relevant guidelines and standards. After the black and odorous water bodies are determined, they are reported by management agencies at all levels step by step, and a unified ledger is established. This method of investigation is inefficient, time-consuming and labor-intensive, and easy to miss. It is impossible to carry out efficient dynamic monitoring, and new technologies and methods are urgently needed. In the past decade, with the popularization and application of high-spatial-resolution satellite data, relevant research on remote sensing monitoring of black and odorous water bodies in rural areas has been gradually carried out. Since black and odorous water bodies and normal water bodies differ in many aspects such as color, odor, and substances contained, their spectral characteristics also have certain differences. By analyzing the differences in the spectra of black and odorous water bodies and normal water bodies in relevant bands and comparing them, starting from the indicators that may cause differences between the two from visible light to near-infrared bands, by constructing different black and odorous water remote sensing identification models such as difference and ratio, and then determining the corresponding thresholds through image feature analysis and field investigation, the distribution range of rural black and odorous water bodies can be extracted. In response to the above problems, the inventors proposed a provincial rural black and odorous water body satellite remote sensing identification method and system for solving the above problems. Summary of the invention
[0004] In order to solve the problem that in the practice of rural water environment management in most areas, the investigation of rural black and odorous water bodies is mainly carried out by investigators on-site verification, questionnaire surveys and water quality monitoring of some water bodies, which has low investigation efficiency, long cycle, time-consuming and labor-intensive. Local investigators investigate and report on their own, which easily miss some black and odorous water bodies, affecting the unified supervision effect of provincial environmental management departments. It is urgent to adopt new technical means to improve work efficiency and supervision effect. Based on the needs of the actual management of black and odorous water bodies in rural areas at the provincial level, the present invention uses multi-source satellite remote sensing data to develop a method and process for identifying and extracting black and odorous water bodies in rural areas within the province, and develops a remote sensing identification system for black and odorous water bodies in rural areas, which can greatly improve the efficiency of related work; the purpose of the present invention is to provide a satellite remote sensing identification method and system for black and odorous water bodies in rural areas within the province.
[0005] In order to solve the above technical problems, the present invention adopts the following technical solution: a method for identifying black and odorous water bodies in rural areas within a province by satellite remote sensing, comprising the following steps:
[0006] S1. Data acquisition: obtain the latest original multi-source satellite remote sensing image data or relevant calibrated data products of the monitoring area, and obtain the digital elevation model data (DEM) and urban built-up area data of the area;
[0007] S2. Data processing: The acquired satellite images are processed according to the relevant procedures, including radiation calibration, atmospheric correction, orthorectification, geometric correction and image fusion, and finally a geometrically corrected high spatial resolution reflectance data product is produced;
[0008] S3, Data image mosaic: DEM data is mosaicked, and the slope of the monitoring area is calculated using it to generate a slope raster image, which is then resampled to keep its spatial resolution and reflectivity data consistent;
[0009] S4. Obtain water distribution data: Based on the above data products, a water extraction model is constructed using the near infrared single band index (NIR), normalized difference water index (NDWI), shadow water index (SWI) and slope grid. According to the relevant knowledge base, appropriate extraction thresholds are set for different regions and different satellite data sources to extract water and obtain water distribution data.
[0010] S5. Data clipping: clip the above water body distribution data using the vector data of urban built-up areas within the monitoring range to obtain water body distribution data in rural areas within the monitoring range;
[0011] S6, obtain index data products: Based on the data products produced in S2, use water index models such as BOI and NDBWI to perform raster image algebraic operations to obtain corresponding index data products;
[0012] S7, obtaining classified grids: according to the prior knowledge base, setting appropriate extraction thresholds for different regional ranges and different remote sensing data sources, performing binary classification on the above-mentioned raster data products based on the set thresholds, and obtaining classified grids;
[0013] S8. Preliminary acquisition of the distribution of black and odorous water bodies: The rural water body distribution data obtained in S5 and the classification grid obtained in S7 are superimposed and an “AND” operation is performed. The result is the preliminary distribution range grid of suspected black and odorous water bodies in rural areas of the monitoring area;
[0014] S9. Obtain the final distribution of black and odorous water bodies: Overlay the above-mentioned suspected black and odorous water body distribution raster data in rural areas and the reflectance data product obtained in S2 for visual interpretation, exclude suspected artificial aquaculture, fishing ponds and other water bodies based on their spatial distribution characteristics, and obtain the final distribution range of suspected black and odorous water bodies in rural areas.
[0015] The original data of multi-source satellite remote sensing images in S1 include domestic Gaofen-2, Gaofen-1, Gaofen-6 or other remote sensing image data with similar spatial resolution, and the spatial resolution of the panchromatic band should be better than 2 meters; the calibrated data products of multi-source satellite remote sensing images include Sentinel2B satellite L1B or L2A image products;
[0016] The DEM data in S1 uses SRTM 30-meter elevation data, and the urban built-up area uses the GUB (Global Urban Boundary) dataset released by the relevant research team of Tsinghua University;
[0017] The radiation calibration in S2 adopts the following formula: L(λ)=gain(λ)*DN(λ)+offset(λ), where L(λ) is the radiation calibration result of the λ band, DN(λ) is the original pixel value of the λ band, gain(λ) and offset(λ) are the gain and offset values of the λ band of the satellite remote sensing sensor used, which can be obtained from the website of the China Resources Satellite Application Center or relevant satellite remote sensing data providers; atmospheric correction adopts the Flaash model or relative radiation correction model of the ENVI remote sensing software, and orthorectification adopts the gdal_translate command line tool of the GDAL library; image fusion adopts the NNdiffusePansharpen algorithm or the pan_sharpen command line tool of the GDAL library;
[0018] The geometric precision correction in S2 uses Sentinel L1B or L2A data as the reference image and uses the arosics library for batch automatic processing. In order to achieve faster processing speed and higher correction accuracy, its parameters are set as follows:
[0019] (1) max_shift is set to 100;
[0020] (2) For 1-meter resolution images, grid_res is set to 800 and window_size is set to 512;
[0021] (3) For 2-meter resolution images, grid_size is set to 200 and window_size is set to 256;
[0022] The following formula is used to calculate the slope based on DEM in S3:
[0023] slope=ATAN(√([dz / dx]2+[dz / dy]2))*57.29578;
[0024] dx is the distance difference between the two pixels in the x-axis direction, dy is the distance difference between the two pixels in the y-axis direction, and dz is the calculated result of the elevation difference between the two pixels in degrees;
[0025] The water index used in S4 is calculated according to the following formula:
[0026] NIR=R(nir);
[0027] NDWI=(R(green)-R(nir)) / (R(green)+R(nir));
[0028] SWI=R(blue)+R(green)-R(nir);
[0029] In the formula, R(blue) is the reflectance of the blue band, R(green) is the reflectance of the green band, and R(nir) is the reflectance of the near-infrared band;
[0030] The water extraction threshold described in S4 is determined based on a pre-established threshold lookup table. The relevant values in the water threshold lookup table are determined based on the selected satellite and the actual situation of the monitoring area, and are determined based on the relevant characteristics of the image histogram and the field investigation. The recommended values are as follows: GF1 and GF2 satellites can be set as follows: NIR<1800; NDWI<-0.50; SWI>-500; slope<20 GF1B, C, and D satellites can be set as follows: NIR<1500; NDWI<-0.60; SWI>-1000; slope<20;
[0031] The water body extraction described in S4 is to perform an “AND” operation on the classification results of the three indexes NIR, NDWI, and SWI and the calculation results of SLOPE.
[0032] The calculation of NDBWI and BOI black and odorous water index described in S6 is as follows:
[0033] NDBWI=(R(green)-R(red)) / (R(green)+R(red));
[0034] BOI=(R(green)-R(red)) / (R(blue)+R(green)+R(red));
[0035] In the formula, R(blue) is the reflectivity of the blue band, R(green) is the reflectivity of the green band, and R(red) is the reflectivity of the red band;
[0036] The black and odorous water extraction threshold value in S7 is determined according to a preset black and odorous water extraction threshold value lookup table. The relevant values in the black and odorous water threshold value lookup table are determined according to the selected satellite and the field survey of normal water bodies and black and odorous water bodies in the monitoring area. The recommended values are as follows:
[0037] GF1 and GF2 satellites can be set: NDBWI<0.050; BOI<0.054;
[0038] GF1B, C, D satellites can be set as follows: NDBWI<0.026; BOI<0.028.
[0039] A system for a method for identifying black and odorous water bodies in rural areas within a province by satellite remote sensing, comprising the following steps:
[0040] Step 1: Satellite image raw data decompression module: decompress the raw image to the target folder according to the set raw image folder and decompression target folder;
[0041] Step 2: Satellite image index establishment module: Extract satellite image metadata (acquisition time, date, cloud cover, etc.) and the coverage of each image (upper left corner longitude and latitude, upper right corner longitude and latitude, lower right corner longitude and latitude, lower left corner longitude and latitude), establish spatial index for high-resolution image data, and store the index file in shapefile format, with the coordinate system used by the original satellite image data;
[0042] Step 3: Image processing module: automatically processes satellite images, including radiation calibration, atmospheric correction, orthorectification, image fusion, and geometric correction (precise geometric correction based on existing reference images);
[0043] Step 4: Water body extraction module: automatically extract water bodies, using the method model described in S4, and automatically extract the distribution range of water bodies in the monitoring area based on the corrected image according to the set threshold;
[0044] Step 5: suspected black and smelly water extraction module: automatically extract suspected black and smelly water in rural areas, and use the black and smelly water calculation model described in S6 to automatically extract suspected black and smelly water according to the set threshold;
[0045] Step 6: Navigation QR code generation module: Call the map software related API to automatically generate Amap and Baidu map navigation QR codes according to the longitude and latitude of the center of the suspected black and odorous water patch. Use the corresponding mobile navigation App to scan the QR code to plan the route, making it convenient to quickly reach the location during on-site verification.
[0046] Compared with the prior art, the present invention has the following beneficial effects:
[0047] 1. The present invention utilizes domestic high-resolution satellite images, foreign Sentinel2B satellite and other multi-source satellite remote sensing image data, combined with on-site field measurements of normal water bodies and black and odorous water bodies, spectral characteristics, dissolved oxygen, ammonia nitrogen and transparency and other water quality parameters, and uses NIR, NDWI, SWI index and slope data to build a remote sensing recognition model for water bodies in rural areas within the province. Then, the BOI, NDBWI and other indexes are used to build a suspected black and odorous water body extraction model in rural areas. Combined with the relevant data of field measurements of water bodies in the monitoring area, the extraction thresholds of water bodies and black and odorous water bodies are determined, and the automatic recognition and extraction of black and odorous water bodies in rural areas are realized, and a related software system is developed;
[0048] 2. The relevant technical methods and systems of the present invention make full use of the natural advantages of remote sensing technology, and can conduct full coverage and periodic monitoring of black and odorous water bodies in rural areas within the province, provide spatial distribution data of black and odorous water body locations for rural water environment management, facilitate on-site verification and treatment, greatly improve the efficiency of black and odorous water body detection, and save manpower and material resources. BRIEF DESCRIPTION OF THE DRAWINGS
[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0050] Figure 1 It is a schematic diagram of the overall technical process of the present invention;
[0051] Figure 2 This is a schematic diagram of the spectrum curve of a rural turbid river according to the present invention;
[0052] Figure 3 This is a schematic diagram of the spectrum curve of black and smelly pond water in the present invention;
[0053] Figure 4This is a schematic diagram of the initial interface of the software system of the present invention;
[0054] Figure 5 This is a schematic diagram of water distribution in rural areas of city B in an embodiment of the present invention;
[0055] Figure 6 This is a distribution map of suspected black and smelly water bodies in the rural area of city B in an embodiment of the present invention. DETAILED DESCRIPTION
[0056] 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.
[0057] like Figure 1-6 As shown, the present invention provides a method for identifying black and odorous water bodies in rural areas within a province by satellite remote sensing, comprising the following steps:
[0058] S1. Data acquisition: obtain the latest original multi-source satellite remote sensing image data or relevant calibrated data products of the monitoring area, and obtain the digital elevation model data (DEM) and urban built-up area data of the area;
[0059] S2. Data processing: The acquired satellite images are processed according to the relevant procedures, including radiation calibration, atmospheric correction, orthorectification, geometric correction and image fusion, and finally a geometrically corrected high spatial resolution reflectance data product is produced;
[0060] S3, Data image mosaic: DEM data is mosaicked, and the slope of the monitoring area is calculated using it to generate a slope raster image, which is then resampled to keep its spatial resolution and reflectivity data consistent;
[0061] S4. Obtain water distribution data: Based on the above data products, a water extraction model is constructed using near infrared single band index (NIR), normalized difference water index (NDWI), shadow water index (SWI) and slope grid. According to the relevant knowledge base, appropriate extraction thresholds are set for different regions and different satellite data sources to extract water and obtain water distribution data.
[0062] S5. Data clipping: clip the above water body distribution data using the vector data of urban built-up areas within the monitoring range to obtain water body distribution data in rural areas within the monitoring range;
[0063] S6, obtain index data products: Based on the data products produced in S2, use water index models such as BOI and NDBWI to perform raster image algebraic operations to obtain corresponding index data products;
[0064] S7, obtaining classified grids: according to the prior knowledge base, setting appropriate extraction thresholds for different regional ranges and different remote sensing data sources, performing binary classification on the above-mentioned raster data products based on the set thresholds, and obtaining classified grids;
[0065] S8. Preliminary acquisition of the distribution of black and odorous water bodies: The rural water body distribution data obtained in S5 and the classification grid obtained in S7 are superimposed and an “AND” operation is performed. The result is the preliminary distribution range grid of suspected black and odorous water bodies in rural areas of the monitoring area;
[0066] S9. Obtain the final distribution of black and odorous water bodies: Overlay the above-mentioned suspected black and odorous water body distribution raster data in rural areas and the reflectance data product obtained in S2 for visual interpretation, exclude suspected artificial aquaculture, fishing ponds and other water bodies based on their spatial distribution characteristics, and obtain the final distribution range of suspected black and odorous water bodies in rural areas.
[0067] The original data of multi-source satellite remote sensing images in S1 include domestic Gaofen-2, Gaofen-1, Gaofen-6 or other remote sensing image data with similar spatial resolution, and the spatial resolution of the panchromatic band should be better than 2 meters; the calibrated data products of multi-source satellite remote sensing images include Sentinel2B satellite L1B or L2A image products;
[0068] The DEM data in S1 uses SRTM 30-meter elevation data, and the urban built-up area uses the GUB (Global Urban Boundary) dataset released by the relevant research team of Tsinghua University.
[0069] The radiation calibration in S2 adopts the following formula: (λ) = gain(λ)*DN(λ)+offset(λ), where L(λ) is the radiation calibration result of the λ band, DN(λ) is the original pixel value of the λ band, gain(λ) and offset(λ) are the gain and offset values of the λ band of the satellite remote sensing sensor used, which can be obtained from the website of the China Resources Satellite Application Center or relevant satellite remote sensing data providers; the atmospheric correction adopts the Flaash model or the relative radiation correction model of the ENVI remote sensing software, and the orthorectification adopts the gdal_translate command line tool of the GDAL library; the image fusion adopts the NNdiffusePansharpen algorithm or the pan_sharpen command line tool of the GDAL library.
[0070] The geometric precision correction in S2 uses Sentinel L1B or L2A data as the reference image and uses the arosics library for batch automatic processing. In order to achieve faster processing speed and higher correction accuracy, its parameters are set as follows:
[0071] (1) max_shift is set to 100;
[0072] (2) For 1-meter resolution images, grid_res is set to 800 and window_size is set to 512;
[0073] (3) For 2-meter resolution images, grid_size is set to 200 and window_size is set to 256.
[0074] The following formula is used to calculate the slope based on DEM in S3:
[0075] slope=ATAN(√([dz / dx]2+[dz / dy]2))*57.29578;
[0076] dx is the distance difference between the two pixels in the x-axis direction, dy is the distance difference between the two pixels in the y-axis direction, and dz is the calculated result of the elevation difference between the two pixels in degrees.
[0077] The water index used in S4 is calculated according to the following formula:
[0078] NIR=R(nir);
[0079] NDWI=(R(green)-R(nir)) / (R(green)+R(nir));
[0080] SWI=R(blue)+R(green)-R(nir);
[0081] In the formula, R(blue) is the reflectivity of the blue band, R(green) is the reflectivity of the green band, and R(nir) is the reflectivity of the near-infrared band.
[0082] The water body extraction threshold described in S4 is determined according to a pre-established threshold lookup table. The relevant values in the water body threshold lookup table are determined according to the actual situation of the selected satellite and the monitoring area, and are determined according to the relevant characteristics of the image histogram and the field investigation. The recommended values are as follows: GF1 and GF2 satellites can be set: NIR<1800; NDWI<-0.50; SWI>-500; slope<20 GF1B, C, and D satellites can be set: NIR<1500; NDWI<-0.60; SWI>-1000; slope<20.
[0083] The water body extraction described in S4 is to perform an “AND” operation on the classification results of the three indexes NIR, NDWI, and SWI and the calculation results of SLOPE.
[0084] The calculation of NDBWI and BOI black and odorous water index described in S6 is as follows:
[0085] NDBWI=(R(green)-R(red)) / (R(green)+R(red));
[0086] BOI=(R(green)-R(red)) / (R(blue)+R(green)+R(red));
[0087] In the formula, R(blue) is the reflectivity of the blue band, R(green) is the reflectivity of the green band, and R(red) is the reflectivity of the red band.
[0088] The black and odorous water extraction threshold value in S7 is determined according to a preset black and odorous water extraction threshold value lookup table. The relevant values in the black and odorous water threshold value lookup table are determined according to the selected satellite and the field survey of normal water bodies and black and odorous water bodies in the monitoring area. The recommended values are as follows:
[0089] GF1 and GF2 satellites can be set: NDBWI<0.050; BOI<0.054;
[0090] GF1B, C, D satellites can be set: NDBWI<0.026; BOI<0.028;
[0091] The reflectance image is classified into two values according to the above conditions, and the pixel value that meets the judgment formula is set to 1, and the pixel value that does not meet the judgment formula is set to 0.
[0092] A system for a method for identifying black and odorous water bodies in rural areas within a province by satellite remote sensing, characterized in that it comprises the following steps:
[0093] Step 1: Satellite image raw data decompression module: decompress the raw image to the target folder according to the set raw image folder and decompression target folder;
[0094] Step 2: Satellite image index establishment module: Extract satellite image metadata (acquisition time, date, cloud cover, etc.) and the coverage of each image (upper left corner longitude and latitude, upper right corner longitude and latitude, lower right corner longitude and latitude, lower left corner longitude and latitude), establish spatial index for high-resolution image data, and store the index file in shapefile format, with the coordinate system used by the original satellite image data;
[0095] Step 3: Image processing module: automatically processes satellite images, including radiation calibration, atmospheric correction, orthorectification, image fusion, and geometric correction (precise geometric correction based on existing reference images);
[0096] Step 4: Water body extraction module: automatically extract water bodies, using the method model described in S4, and automatically extract the distribution range of water bodies in the monitoring area based on the corrected image according to the set threshold;
[0097] Step 5: suspected black and smelly water extraction module: automatically extract suspected black and smelly water in rural areas, and use the black and smelly water calculation model described in S6 to automatically extract suspected black and smelly water according to the set threshold;
[0098] Step 6: Navigation QR code generation module: Call the map software related API to automatically generate Amap and Baidu map navigation QR codes according to the longitude and latitude of the center of the suspected black and odorous water body patch. Use the corresponding mobile navigation app to scan the QR code to plan the route, which is convenient for quick arrival at the location during on-site inspection
[0099] Example: Select cities A, B, C and D in a province as monitoring areas, and conduct remote sensing investigation on black and odorous water bodies in rural areas of these four cities. The specific process is as follows:
[0100] Step 1: Collect remote sensing data and other data. Domestic high-resolution satellite remote sensing data (GF-1, GF-1B, GF-1C, GF-1D, GF-6) were collected. The spatial resolution was 8 meters for multispectral and 2 meters for panchromatic. Sentinel2B satellite data was collected for geometric correction. The following auxiliary data were collected and sorted: (1) administrative boundary data of a province; (2) global urban range data released by the relevant team of Tsinghua University; (3) SRTM 30-meter dem data of a province; (4) location data of black and odorous water bodies in the above four cities that have been confirmed and registered, etc. The software system developed by the present invention was used to establish a spatial index for satellite data, and the satellite data was screened according to administrative regions, cloud cover, image acquisition time, etc., and more than 200 high-resolution images and more than 30 Sentinel2B images were screened.
[0101] Step 2: Automatically process the high-resolution satellite image data using the software system provided by the present invention, specifically including radiation calibration, atmospheric correction, orthorectification, image fusion and geometric precision correction. The radiation calibration parameters use the calibration parameters of satellite sensors published by the China Resources Satellite Application Center. The atmospheric correction calls the FLAASH model of the ENVI software. The average altitude in the model is obtained from the SRTM 30-meter elevation data. The region is selected as rural. The orthorectification calls the gdal_translate function of the GDAL library. The DEM used in the orthorectification is the SRTM 30-meter elevation data. The image fusion calls the NNDiffusePansharpen tool of the ENVI software. The geometric precision correction calls the COREG_LOCAL function of the arosics library. The grid_res parameter is set to 200, the window_size parameter is set to 256, the max_shift is set to 100, and the reflectivity image product output format is set to Gtiff.
[0102] Step 3: Based on the SRTM 30-meter DEM data, the slope is calculated according to the formula slope = ATAN (√ ([dz / dx]2 + [dz / dy]2)) * 57.29578), the slope raster data is generated, and it is resampled to a 2-meter spatial resolution.
[0103] Step 4: Using the software system provided by the present invention, based on the reflectance image product, calculate water indexes such as NIR, NDWI, SWI as shown below to generate water index raster data;
[0104] NIR=R(nir);
[0105] NDWI=(R(green)-R(nir)) / (R(green)+R(nir));
[0106] SWI=R(blue)+R(green)-R(nir);
[0107] Step 5: Using the software system provided by the present invention, based on the above water index and slope grid data, automatically align and perform binary classification, and set the following thresholds:
[0108] GF1: NIR<1800; NDWI<-0.50; SWI>-500; slope<20;
[0109] GF1B, C, D satellite settings: NIR<1500; NDWI<-0.60; SWI>-1000;
[0110] slope<20;
[0111] The pixel value that satisfies the judgment formula is set to 1, and the pixel value that does not satisfy the judgment formula is set to 0.
[0112] Step 6: Using the software system provided by the present invention, the numpy related functions are called to perform “and” operations on the above classification results to generate a water body distribution grid image. After rough statistics, more than 6,000 water body patches of the types of rivers, ponds and ditches were extracted from the four cities. After excluding the urban built-up areas, there are about 5,000 water bodies in the rural areas of the four cities.
[0113] Step 7: Based on the reflectance product produced in step 2, calculate the NDBWI and BOI black and odorous water index as follows to generate the corresponding result grid.
[0114] NDBWI=(R(green)-R(red)) / (R(green)+R(red));
[0115] BOI=(R(green)-R(red)) / (R(blue)+R(green)+R(red));
[0116] Step 8: Perform binary classification on the result grid generated in step 7 according to the following thresholds to obtain the classification result grid image:
[0117] GF1 satellite can be set as follows: NDBWI<0.050; BOI<0.054;
[0118] GF1B, C, D satellites can be set: NDBWI<0.026; BOI<0.028;
[0119] Step 9: Overlay the above classification result raster image with the water body distribution raster obtained in step 6 to obtain the distribution raster image of suspected black and odorous water bodies in the four cities.
[0120] Step 10: Use the built-up area data of the four cities to cut the above-mentioned suspected black and odorous water body distribution grid. The suspected black and odorous water body patches outside the urban built-up areas are the suspected black and odorous water bodies in the rural areas of the four cities.
[0121] Step 11: Load the raster images of suspected black and odorous water bodies in rural areas of the above four cities into ArcGIS software, and overlay the Tiandi Map WMTS image and image annotation layer, and interpret them according to the following spatial distribution characteristics. Those that meet the following characteristics are not black and odorous water bodies:
[0122] (1) The shape of the water body is a regular quadrilateral;
[0123] (2) There is a guardhouse near the water body;
[0124] (3) The water bodies are distributed in a continuous area, which is obviously a breeding area;
[0125] (4) The water body is located in parks, scenic spots, etc.;
[0126] Step 12: Using the software system provided by the present invention, call the relevant functions of the fiona and rasterio libraries to vectorize the above suspected raster, obtain the distribution vector data of suspected black and odorous water bodies in rural areas of the four cities, and edit it using ArcGIS software to calculate the coordinates of the center point of each water body polygon. A total of 100 suspected black and odorous water bodies were identified in the four cities.
[0127] Step 13: Using the software system provided by the invention, call the Amap API to generate the Amap Gaohang QR code according to the coordinates of the above-mentioned central store, and scan it with the Amap mobile APP or WeChat to plan a route to the suspected black and smelly water body.
[0128] At this point, the remote sensing identification of suspected black and smelly water bodies in rural areas of four cities including City A in a certain province has been completed.
[0129] This embodiment describes the main features and basic process of the present invention. It should be understood by those skilled in the art that the present invention is not limited by the above embodiment. The above embodiment and description are only for explaining the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.
[0130] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.
Claims
1. A satellite remote sensing identification method for black and odorous water bodies in rural areas within a province, characterized in that: The following steps are involved: S1. Data acquisition: Acquire the latest original multi-source satellite remote sensing image data or relevant calibrated data products of the monitoring area, and obtain the digital elevation model data and urban built-up area data of the area; S2. Data processing: The acquired satellite images are processed according to the relevant procedures, including radiation calibration, atmospheric correction, orthorectification, geometric correction and image fusion, and finally a geometrically corrected high spatial resolution reflectance data product is produced; S3, Data image mosaic: DEM data is mosaicked, and the slope of the monitoring area is calculated using it to generate a slope raster image, which is then resampled to keep its spatial resolution and reflectivity data consistent; S4. Obtain water distribution data: Based on the above data products, a water extraction model is constructed using near-infrared single-band index, normalized water index, shadow water index and slope grid. According to the relevant knowledge base, appropriate extraction thresholds are set for different regions and different satellite data sources to extract water and obtain water distribution data. S5. Data clipping: clip the above water body distribution data using the vector data of urban built-up areas within the monitoring range to obtain water body distribution data in rural areas within the monitoring range; S6, obtain index data products: Based on the data products produced in S2, use water index models such as BOI and NDBWI to perform raster image algebraic operations to obtain corresponding index data products; S7, obtaining classified grids: according to the prior knowledge base, setting appropriate extraction thresholds for different regional ranges and different remote sensing data sources, performing binary classification on the above-mentioned raster data products based on the set thresholds, and obtaining classified grids; S8. Preliminary acquisition of the distribution of black and odorous water bodies: The rural water body distribution data obtained in S5 and the classification grid obtained in S7 are superimposed and an "AND" operation is performed. The result is the preliminary distribution range grid of suspected black and odorous water bodies in rural areas of the monitoring area; S9. Obtain the final distribution of black and odorous water bodies: Overlay the above-mentioned suspected black and odorous water body distribution raster data in rural areas and the reflectance data product obtained in S2 for visual interpretation, exclude suspected artificial aquaculture, fishing ponds and other water bodies based on their spatial distribution characteristics, and obtain the final distribution range of suspected black and odorous water bodies in rural areas.
2. A method for identifying black and odorous water bodies in rural areas by satellite remote sensing in a provincial scope as claimed in claim 1, characterized in that: The original data of multi-source satellite remote sensing images in S1 include domestic Gaofen-2, Gaofen-1, Gaofen-6 or other remote sensing image data with similar spatial resolution, and the spatial resolution of the panchromatic band should be better than 2 meters; the calibrated data products of multi-source satellite remote sensing images include Sentinel2B satellite L1B or L2A image products; The DEM data in S1 uses SRTM 30-meter elevation data, and the urban built-up area uses the GUB dataset released by the relevant research team of Tsinghua University.
3. A method and system for identifying black and odorous water bodies in rural areas in a provincial scope by satellite remote sensing as claimed in claim 1, characterized in that: The radiation calibration in S2 adopts the following formula: L(λ)=gain(λ)*DN(λ)+offset(λ), where L(λ) is the radiation calibration result of the λ band, DN(λ) is the original pixel value of the ( band, gain(λ) and offset(λ) are the gain and offset values of the ( band of the satellite remote sensing sensor used, which can be obtained from the website of the China Resources Satellite Application Center or relevant satellite remote sensing data providers; atmospheric correction adopts the Flaash model or relative radiation correction model of the ENVI remote sensing software, and orthorectification adopts the gdal_translate command line tool of the GDAL library; image fusion adopts the NNdiffusePansharpen algorithm or the pan_sharpen command line tool of the GDAL library; The geometric precision correction in S2 uses Sentinel L1B or L2A data as the reference image and uses the arosics library for batch automatic processing. In order to achieve faster processing speed and higher correction accuracy, its parameters are set as follows: (1) max_shift is set to 100; (2) For 1-meter resolution images, grid_res is set to 800 and window_size is set to 512; (3) For 2-meter resolution images, grid_size is set to 200 and window_size is set to 256.
4. A method and system for identifying black and odorous water bodies in rural areas in a provincial scope by satellite remote sensing as claimed in claim 1, characterized in that: The following formula is used to calculate the slope based on DEM in S3: slope=ATAN(√([dz / dx]2+[dz / dy]2))*57.29578; dx is the distance difference between the two pixels in the x-axis direction, dy is the distance difference between the two pixels in the y-axis direction, and dz is the calculated result of the elevation difference between the two pixels in degrees.
5. A method and system for identifying black and odorous water bodies in rural areas in a provincial scope by satellite remote sensing as claimed in claim 1, characterized in that: The water index used in S4 is calculated according to the following formula: NIR=R(nir); NDWI=(R(green)-R(nir)) / (R(green)+R(nir)); SWI=R(blue)+R(green)-R(nir); In the formula, R(blue) is the reflectivity of the blue band, R(green) is the reflectivity of the green band, and R(nir) is the reflectivity of the near-infrared band.
6. A method and system for identifying black and odorous water bodies in rural areas by satellite remote sensing in a provincial scope as claimed in claim 1, characterized in that: The water body extraction threshold described in S4 is determined according to a pre-established threshold lookup table. The relevant values in the water body threshold lookup table are determined according to the actual situation of the selected satellite and the monitoring area, and are determined according to the relevant characteristics of the image histogram and the field investigation. The recommended values are as follows: GF1 and GF2 satellites can be set: NIR<1800; NDWI<-0.50; SWI>-500; slope<20 GF1B, C, and D satellites can be set: NIR<1500; NDWI<-0.60; SWI>-1000; slope<20.
7. A method and system for identifying black and odorous water bodies in rural areas in a provincial scope by satellite remote sensing as claimed in claim 1, characterized in that: The water body extraction described in S4 is to perform an "AND" operation on the classification results of the three indexes NIR, NDWI, and SWI and the calculation results of SLOPE.
8. A method and system for identifying black and odorous water bodies in rural areas by satellite remote sensing in a provincial scope as claimed in claim 1, characterized in that: The calculation of NDBWI and BOI black and odorous water index described in S6 is as follows: NDBWI=(R(green)-R(red)) / (R(green)+R(red)); BOI=(R(green)-R(red)) / (R(blue)+R(green)+R(red)); In the formula, R(blue) is the reflectivity of the blue band, R(green) is the reflectivity of the green band, and R(red) is the reflectivity of the red band.
9. A method and system for identifying black and odorous water bodies in rural areas by satellite remote sensing in a provincial scope as claimed in claim 1, characterized in that: The black and odorous water extraction threshold value in S7 is determined according to a preset black and odorous water extraction threshold value lookup table. The relevant values in the black and odorous water threshold value lookup table are determined according to the selected satellite and the field survey of normal water bodies and black and odorous water bodies in the monitoring area. The recommended values are as follows: GF1 and GF2 satellites can be set: NDBWI<0.050; BOI<0.054; GF1B, C, D satellites can be set: NDBWI<0.026; BOI<0.028; The reflectance image is classified into two values according to the above conditions, and the pixel value that meets the judgment formula is set to 1, and the pixel value that does not meet the judgment formula is set to 0.
10. The system used in the method for identifying black and odorous water bodies in rural areas in a provincial scope by satellite remote sensing as claimed in claim 1, characterized in that: The following steps are involved: Step 1: Satellite image raw data decompression module: decompress the raw image to the target folder according to the set raw image folder and decompression target folder; Step 2: Satellite image index establishment module: Extract satellite image metadata and the coverage of each image, establish a spatial index for the high-resolution image data, and store the index file in shapefile format, with the coordinate system being the coordinate system used by the original satellite image data; Step 3: Image processing module: automatically processes satellite images, including radiation calibration, atmospheric correction, orthorectification, image fusion, and geometric correction; Step 4: Water body extraction module: automatically extract water bodies, use the method model described in S4, and automatically extract the distribution range of water bodies in the monitoring area based on the corrected image according to the set threshold; Step 5: suspected black and smelly water extraction module: automatically extract suspected black and smelly water in rural areas, and use the black and smelly water calculation model described in S6 to automatically extract suspected black and smelly water according to the set threshold; Step 6: Navigation QR code generation module: Call the map software related API to automatically generate the map software's QR code based on the longitude and latitude of the center of the suspected black and odorous water patch. Use the corresponding mobile navigation App to scan the QR code to plan the route, making it convenient to quickly reach the location during on-site inspection.
Citation Information
Patent Citations
Remote sensing image water extraction method and device based on RGBW characteristic space diagram cutting algorithm
CN105631903A
Method for distinguishing black and odorous water body based on feature fusion of remote sensing image
CN114359699A
Rural suspected black and odorous water body identification method based on high-resolution image
CN116704349A
Landslide hidden danger comprehensive remote sensing identification method, system, equipment and medium
CN116994156A
Rural black and odorous water body monitoring method based on satellite remote sensing
CN117710831A
Cited By
Black and odorous water body identification method and system based on mobile phone photographing
CN119723356A
A method and system for identifying black and odorous water bodies based on mobile phone photography
CN119723356B
