CN109993237B - Water body rapid extraction method and system based on high-resolution satellite optical remote sensing data - Google Patents
Water body rapid extraction method and system based on high-resolution satellite optical remote sensing data Download PDFInfo
- Publication number
- CN109993237B CN109993237B CN201910299477.7A CN201910299477A CN109993237B CN 109993237 B CN109993237 B CN 109993237B CN 201910299477 A CN201910299477 A CN 201910299477A CN 109993237 B CN109993237 B CN 109993237B
- Authority
- CN
- China
- Prior art keywords
- remote sensing
- water body
- sensing data
- optical remote
- resolution satellite
- 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.)
- Expired - Fee Related
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/27—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands using photo-electric detection ; circuits for computing concentration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Biochemistry (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Pathology (AREA)
- Immunology (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Image Processing (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a water body rapid extraction method based on high-resolution satellite optical remote sensing data, which comprises the following steps: acquiring optical remote sensing data and digital elevation model data of a high-resolution satellite; carrying out radiometric calibration on the high-resolution satellite optical remote sensing data; atmospheric correction processing is carried out on the radiometric calibrated optical remote sensing data of the high-resolution satellite; performing orthorectification on the high-resolution satellite optical remote sensing data after atmospheric rectification processing; carrying out spectrum transformation processing on the high-resolution satellite optical remote sensing data after the incidence correction to obtain a spectrum calculation result; carrying out color transformation processing on the high-resolution satellite optical remote sensing data after the spectrum transformation processing; obtaining a water body classification result for the high-resolution satellite optical remote sensing data after the color transformation processing; extracting the height difference of the grid from the nearest neighbor water system based on the digital height model data; correcting the water body classification result based on the elevation difference of the grid distance nearest neighbor water system; and realizing regional water body mapping based on the corrected water body classification result.
Description
Technical Field
The invention relates to the field of water body monitoring, in particular to a water body rapid extraction method and system based on high-resolution satellite optical remote sensing data.
Background
Surface water as an important carrier in land water circulation is a key indicator of regional water balance. The change of the surface water body is comprehensively influenced by climate change and human activities, and the formation, disappearance, expansion and contraction of the land water body and the change of the ecological environment caused by the land water body reflect the climate change condition and the human activity condition of a certain area and even the whole world. With the development of the economic society and the influence of human activities, severe problems of great reduction, area shrinkage, functional degradation and the like of land water appear, and the dynamic monitoring of water bodies and water environment supervision in areas need to be carried out urgently. The satellite remote sensing is an effective means for dynamically monitoring the land surface water body, the land surface water body real-time dynamic monitoring application based on the satellite remote sensing can reveal the influence rule of natural factors and human activities on the water area, has extremely important significance for rational development, utilization and protection of the land surface water area, and has irreplaceable important value for timely mastering the water quantity balance of the area, the sustainable utilization of the water resource of the area, the monitoring and forecasting of flood/drought disasters, the disaster reduction, the disaster assessment and the like.
The remote sensing technology has the capability of monitoring the dynamic change of the earth surface environment in a large range in time and quickly, and is an important technical means for monitoring the dynamic change of the large-range land surface water area. At present, the water body monitoring method based on optical satellite remote sensing mainly comprises the following steps: (1) manual sketching method; the manual drawing method is troublesome and hard, and large-scale business monitoring application is difficult to realize. (2) A single band threshold method; the single-band threshold method realizes the extraction of the water body mainly according to the characteristic that the water body has low reflectivity in the near infrared band and is easy to distinguish from other ground objects. However, the single-band threshold method is easily affected by mountain shadows, so that a large error exists in water body extraction. (3) Inter-spectrum relationship method; and performing combined operation on the wave bands to realize the distinguishing of the water body and other ground objects. The inter-spectrum relation method is better applied to typical lake water bodies, and large uncertainty still exists in large-scale water body monitoring application. (4) A water body index method; and carrying out normalized difference processing by using the wave bands to highlight the water body information in the image. Representative water body indices are normalized differential water body index (NDWI) as proposed by mcfeetts in 1996 and MNDWI index as proposed by xu autumn et al in 2005. In the extraction of a large-scale water body, the water body index method still faces the problem that the water body index threshold is difficult to determine. (5) The image classification method mainly comprises a supervised classification method and an unsupervised classification method. The unsupervised classification method needs to combine the regional geographic environment characteristics to perform the merging processing of the classification results. The supervised classification method needs to manually select training samples and train classification models so as to realize the extraction of the water body. The image classification method also faces the problems of insufficient water body samples and inaccurate classification in the large-scale water body monitoring application.
In recent years, with the development and progress of the satellite remote sensing technology in China, a large number of domestic satellites are launched and the satellite data acquisition is increasingly convenient, so that the possibility of the large-scale water body monitoring business in China is provided. However, the spectral resolution of the currently-produced high-grade satellite 1 and 2 is low, and a certain gap exists between the spectral resolution of the overseas Landsat8 optical satellite and the Sentinil 2 optical satellite. Particularly, the domestic high-resolution satellites 1 and 2 are not provided with short-wave infrared bands (1.3-3.0 mu m), so that the domestic high-resolution satellites have obvious defects in water body extraction. At present, great uncertainty still exists in the rapid and accurate extraction of a large-scale water body based on domestic high-resolution satellite data, and in the existing model method, no method which can effectively solve the rapid extraction of the large-scale water body based on the high-resolution satellite optical data still exists, and particularly, the accurate and rapid extraction of the large-scale water body by using the high-resolution optical data with only 4 wave bands still remains a difficult point to be solved urgently in practical application.
Disclosure of Invention
In order to solve the technical problems, the invention provides a rapid water body extraction method based on high-resolution satellite optical remote sensing data, which comprises the following steps:
acquiring altitude satellite optical remote sensing data and digital elevation model data of a research area;
carrying out radiometric calibration on the high-resolution satellite optical remote sensing data;
atmospheric correction processing is carried out on the high-resolution satellite optical remote sensing data subjected to radiometric calibration;
performing orthorectification on the high-resolution satellite optical remote sensing data subjected to atmospheric rectification processing;
carrying out spectrum transformation processing on the high-resolution satellite optical remote sensing data subjected to the orthometric correction to obtain a spectrum calculation result of the approximate short wave infrared band characteristic of the high-resolution satellite optical remote sensing data;
carrying out color transformation processing on the high-resolution satellite optical remote sensing data subjected to the spectrum transformation processing;
carrying out water body classification extraction on the high-resolution satellite optical remote sensing data subjected to color transformation processing to obtain a water body classification result;
extracting the height difference of a grid in a research area from a nearest water system based on the digital height model data;
correcting the water body classification result based on the elevation difference of the grid to the nearest neighbor water system;
and realizing regional water body mapping based on the corrected water body classification result.
Preferably, the radiometric calibration of the high-resolution satellite optical remote sensing data comprises:
converting the high-resolution satellite optical remote sensing data into radiance;
wherein the radiance is obtained by:
Lλ＝Gain*DN；
wherein Gain is a scaling coefficient (W/(m)2Sr μm)), DN is the observed value of the satellite-borne sensor, LλIs the radiance (W/(m)2·sr·μm))。
Preferably, the method further comprises the following steps:
converting the radiance into zenith reflectance;
wherein the zenith reflectivity is obtained by the following formula:
in the formula, ρTOA，λIs zenith reflectivity, pi is circumferential ratio, d is distance between day and earth, ESUNλIs the average solar radiation value (W/(m) of the wave band2·μm))，θsThe zenith angle of the sun.
Preferably, the performing the orthorectification on the high-resolution satellite optical remote sensing data after the atmospheric correction processing includes:
and performing geometric distortion correction, inclination correction and projection difference correction on the high-resolution satellite optical remote sensing data subjected to atmospheric correction processing.
Preferably, the spectrum transformation processing is performed on the optical remote sensing data of the high-resolution satellite after the ortho-rectification, and the spectrum calculation result of the approximate short-wave infrared band feature of the optical remote sensing data of the high-resolution satellite includes:
obtaining a spectrum calculation result of the near short wave infrared band characteristic through the following formula:
ρSWIR＝(1.0-NDWI)·(ρR+ρNIR)/2.0；
where ρ isSWIRThe result of the spectral calculation, rho, being characteristic of the near-short-wave infrared bandRSurface reflectance values, rho, for the red band of high-resolution satellitesNIRThe surface reflectance value of the near infrared band of the high-resolution satellite is shown, and the NDWI is a normalized difference water body index.
Preferably, the normalized difference water body index is obtained by the following formula:
NDWI＝(ρG-ρNIR)/(ρG+ρNIR)；
where ρ isGAnd the earth surface reflectance value is the green band of the high-resolution satellite.
Preferably, the color transformation processing of the high-resolution satellite optical remote sensing data after the spectrum transformation processing comprises:
respectively obtaining high-resolution remote sensing data of the high-resolution satellite based on the spectrum calculation result of the approximate short wave infrared band characteristic to perform color transformation, so as to obtain lightness, saturation and chroma;
wherein the lightness is obtained by the following formula:
V＝max(ρSWIR，ρNIR，ρR)；
the saturation is obtained by the following formula:
the chroma is obtained by the following formula:
preferably, the extracting of the water body classification of the high-resolution satellite optical remote sensing data after the color transformation processing is performed, and the obtaining of the water body classification result includes:
respectively obtaining a water body sample point and a non-water body sample point;
sampling the reflectivities of a blue wave band, a green wave band, a red wave band and an approximate short wave infrared band which respectively correspond to the water body sample point and the water body sample point to obtain a training sample set;
training the training sample set by a machine learning method to obtain an image classifier;
and classifying the high-resolution satellite optical remote sensing data based on the image classifier to obtain a water body classification result.
Preferably, the extracting the elevation difference of the grid from the nearest water system within the study area based on the digital elevation model data comprises:
cutting and projection conversion are carried out on the digital elevation model data;
hydrologic analysis is carried out on the digital elevation model data subjected to cutting and projection conversion, and river network data in a research area are extracted;
and extracting the height difference of the grid in the research area from the nearest water system based on the river network data and the digital height model data.
The invention provides a water body rapid extraction system based on high-resolution satellite optical remote sensing data in a second aspect, which comprises:
the acquisition module is used for acquiring the altitude satellite optical remote sensing data and the digital elevation model data of a research area;
the radiometric calibration module is used for radiometric calibration of the high-resolution satellite optical remote sensing data;
the correction processing module is used for carrying out atmospheric correction processing on the high-resolution satellite optical remote sensing data subjected to radiometric calibration;
the ortho-rectification module is used for carrying out ortho-rectification on the high-resolution satellite optical remote sensing data after atmospheric rectification;
the spectrum transformation processing module is used for carrying out spectrum transformation processing on the high-resolution satellite optical remote sensing data subjected to the ortho-rectification to obtain a spectrum calculation result of the approximate short-wave infrared band characteristic of the high-resolution satellite optical remote sensing data;
the color transformation processing module is used for carrying out color transformation processing on the high-resolution satellite optical remote sensing data subjected to the spectrum transformation processing;
the water body classification result extraction module is used for extracting water body classification from the high-resolution satellite optical remote sensing data subjected to color transformation to obtain a water body classification result;
the calculation module extracts the height difference of the grid in the research area from the nearest neighbor water system based on the digital height model data;
the correction module corrects the water body classification result based on the elevation difference of the grid to the nearest neighbor water system;
and the regional water body drawing module is used for realizing regional water body drawing based on the corrected water body classification result.
The invention has the following beneficial effects:
the technical scheme of the invention fully utilizes the spectral band information of the high-resolution satellite, generates a new wave band similar to the characteristics of water and land spectrums through spectrum transformation, converts three color channels consisting of the new wave band of the spectrum transformation, a near infrared wave band and a red light wave band into a chromaticity space, a lightness space and a saturation space based on color transformation, provides a water body classification criterion without regional water body sample information, eliminates false water body pattern spots caused by mountain shadow based on the height difference of a grid from the nearest water system as an important criterion, improves the extraction precision, and can serve the application requirements of the application fields in the aspects of regional water body dynamic monitoring, regional water resource dynamic analysis and evaluation, regional drought monitoring and the like.
Drawings
The following describes embodiments of the present invention in further detail with reference to the accompanying drawings.
Fig. 1 shows a flow chart of a water body fast extraction method based on high-resolution satellite optical remote sensing data according to a first embodiment of the present invention;
FIG. 2 is a tree diagram of a water body classification method based on chromaticity and lightness characteristics in the present embodiment;
fig. 3 shows a structural block diagram of a water body rapid extraction system based on high-resolution satellite optical remote sensing data according to a first embodiment of the present invention;
fig. 4 shows an example of water extraction in the middle of Anhui province based on high-resolution satellite optical images.
Detailed Description
In order to more clearly illustrate the invention, the invention is further described below with reference to preferred embodiments and the accompanying drawings. Similar parts in the figures are denoted by the same reference numerals. It is to be understood by persons skilled in the art that the following detailed description is illustrative and not restrictive, and is not to be taken as limiting the scope of the invention.
In order to solve the problems in the background art, fig. 1 shows a flowchart of a method for rapidly extracting a water body based on high-resolution satellite optical remote sensing data according to a first embodiment of the present invention, as shown in fig. 1, the method includes the following steps:
s1, acquiring altitude satellite optical remote sensing data and digital elevation model data of a research area;
s2, carrying out radiometric calibration on the high-resolution satellite optical remote sensing data;
s3, performing atmospheric correction processing on the optical remote sensing data of the high-resolution satellite subjected to radiometric calibration;
s4, performing orthorectification on the high-resolution satellite optical remote sensing data subjected to the atmospheric rectification processing;
s5, performing spectrum transformation processing on the high-resolution satellite optical remote sensing data after the ortho-rectification to obtain a spectrum calculation result of the approximate short wave infrared band characteristic of the high-resolution satellite optical remote sensing data;
s6, carrying out color transformation processing on the high-resolution satellite optical remote sensing data after the spectrum transformation processing;
s7, extracting the high-resolution satellite optical remote sensing data subjected to color conversion to obtain a water body classification result;
s8, extracting the height difference between the grids in the research area and the nearest water system based on the digital height model data;
s9, correcting the water body classification result based on the height difference of the grid to the nearest water system;
and S10, realizing regional water body mapping based on the corrected water body classification result.
Specifically, in S1, high-resolution satellite optical remote sensing data of a research area is acquired according to research needs, where cloud coverage is lower than 30%, where the high-resolution satellite optical remote sensing data may be acquired through a chinese resource satellite application center, the high-resolution satellite may select a 16-meter multispectral camera or an 8-meter panchromatic multispectral camera observation data, and in this embodiment, the data acquired by the 16-meter multispectral camera is taken as an example, and the data elevation simulation data may acquire Digital Elevation Model (DEM) data of a corresponding coverage research area according to a range of the research area.
Specifically, S2 specifically includes: converting the high-resolution satellite optical remote sensing data into radiance;
further, the method also comprises the following steps: converting the radiance into zenith reflectivity;
it should be noted that radiometric calibration is to convert the obtained optical remote sensing data of the high-resolution satellite into radiance or zenith reflectivity, first, a calibration coefficient of the high-resolution satellite needs to be obtained, which can be specifically obtained from a central website of a domestic resource satellite application, and then, according to a sensor type of the obtained data, radiometric calibration is performed on the optical remote sensing data of the high-resolution satellite, and the radiometric calibration can be converted into radiance or zenith reflectivity, where the radiance is obtained by the following formula:
Lλ＝Gain*DN；
wherein Gain is a scaling coefficient (W/(m)2Sr μm)), DN is the observed value of the satellite-borne sensor, LλIs the radiance (W/(m)2·sr·μm))。
Radiance may be further converted to zenith reflectance, calculated as follows:
in the formula, ρTOA，λIs zenith reflectivity, pi is circumferential ratio, d is distance between day and earth, ESUNλIs the average solar radiation value (W/(m) of the wave band2·μm))，θsThe zenith angle of the sun.
Specifically, in S3, the atmospheric correction processing of the high-resolution satellite optical remote sensing data is performed to eliminate the influence of the atmospheric absorption and scattering on the surface reflectivity, eliminate the radiation error caused by the atmospheric influence, and feed back the reflectivity of the ground object.
In the embodiment, ENVI/FLAASH is mainly adopted to carry out atmospheric correction processing on the optical remote sensing data of the high-resolution satellite, firstly, unit conversion is carried out on the optical remote sensing data of the high-resolution satellite according to the FLAASH input requirement, the data storage format is converted into BIL, secondly, parameters such as sensor height, pixel size, an atmospheric model and an aerosol model are set according to file information of the optical remote sensing data head of the high-resolution satellite, and finally, atmospheric correction processing is executed, a correction result is converted into 0-1, and the step S3 can be skipped for quick and simple extraction of regional large-range water bodies.
Specifically, the S4 includes: and performing geometric distortion correction, inclination correction and projection difference correction on the high-resolution satellite optical remote sensing data subjected to atmospheric correction processing.
The method comprises the steps of correcting geometric distortion of high-resolution satellite optical remote sensing data subjected to atmospheric correction, simultaneously correcting inclination and projection difference of an image, resampling the image into an orthoscopic image, wherein the high-resolution satellite optical remote sensing data comprises an RPC file, and performing orthoscopic correction by using an orthoscopic correction tool based on the RPC in ENVI software.
Specifically, in S5, the multi-spectral bands of the high-resolution satellite sensor include blue (0.45-0.52 μm), green (0.52-0.59 μm), red (0.63-0.69 μm) and near infrared(0.77-0.89 μm) four wave bands, compared with the similar foreign optical satellites Landsat8 and Sentinel2, the domestic high-resolution satellite optical remote sensing data lack the short wave infrared band (SWIR), the short wave near infrared spectrum range is 1.3-3.0 μm, the reflectivity of the water body in the short wave infrared band is lower, the reflectivity of the soil in the spectrum band is higher, the short wave infrared band is an important spectrum band for distinguishing soil and water bodies, because the domestic high-resolution satellite lacks the short wave infrared band, the improved water body index MNDWI cannot be calculated, the high-resolution satellite has defects in the water body extraction analysis in the large area, and in this way, the embodiment provides a spectrum conversion method, the wave band similar to the short wave infrared band characteristic is obtained through the spectrum calculation, and the spectrum calculation result rho of the near short wave infrared band characteristic is obtained through the spectrum calculationSWIRAdding the data into the original wave band of the optical remote sensing data of the high-resolution satellite, wherein the wave spectrum calculation result of the approximate short-wave infrared wave band characteristic is obtained through the following formula:
ρSWIR＝(1.0-NDWI)·(ρR+ρNIR)/2.0；
where ρ isSWIRThe result of the spectral calculation, rho, being characteristic of the near-short-wave infrared bandRSurface reflectance values (or celestial reflectance values) for the red band of high-resolution satellites, pNIRThe surface reflectance value of the near infrared band of the high-resolution satellite is shown, and the NDWI is a normalized difference water body index.
Further, the normalized difference water body index is obtained by the following formula:
NDWI＝(ρG-ρNIR)/(ρG+ρNIR)；
where ρ isGAnd the earth surface reflectance value is the green band of the high-resolution satellite.
In the embodiment, a new waveband similar to the short-wave infrared waveband characteristic is constructed based on the red, green and near infrared wavebands, the reflectivity of the new waveband in a water body is lower than that of the near infrared waveband, but the reflectivity of the new waveband in soil and vegetation areas is enhanced and is obviously greater than that of the soil and vegetation in the near infrared and red waveband.
Specifically, S6 further includes obtaining the high-resolution satellite high-resolution remote sensing data based on the spectrum calculation result of the approximate short-wave infrared band feature, and performing color transformation to obtain lightness, saturation, and chromaticity.
In the present embodiment, the color conversion is a process of converting an image in red, green, and blue color spaces into lightness, chroma, and saturation. The difference between water and non-water ground objects can be enhanced after the high-resolution satellite optical remote sensing data subjected to the spectrum transformation is subjected to the color transformation processing.
Firstly, selecting a spectrum calculation result rho of the near short wave infrared band characteristicSWIRTo select p as the red bandNIRAs green band, ρ is selectedRAs a blue light band, an RGB color space is formed, and then, a color change process is performed on the RGB space,
wherein the lightness is obtained by the following formula:
V＝max(ρSWIR，ρNIR，ρR)；
the saturation is obtained by the following formula:
the chroma is obtained by the following formula:
specifically, S7 further includes:
respectively obtaining a water body sample point and a non-water body sample point;
sampling the reflectivities of a blue wave band, a green wave band, a red wave band and an approximate short wave infrared band which respectively correspond to the water body sample point and the water body sample point to obtain a training sample set;
training the training sample set by a machine learning method to obtain an image classifier;
and classifying the high-resolution satellite optical remote sensing data based on the image classifier to obtain a water body classification result.
The step is to classify the regional water body and the non-water body according to the chroma and the lightness calculated by combining the color transformation, the color water body is highlighted in the chroma space, and the separation of the water body and the non-water body in the lightness space and the chroma space is clear. Specifically, in order to realize the automatic classification of the water body, 15 ten thousand sample points in different seasons and different areas of the water body of the Chinese land are collected, wherein 50% of the sample points are water body sample points, 50% of the sample points are non-water body sample points, and reflectivity values of all wave bands corresponding to the sample points are extracted. The method comprises the steps of training a training sample set by a machine learning method by adopting reflectivity information of blue, green, red and near infrared bands of a foreign Landsat8 optical satellite with 15 ten thousand sample points as the spectral ranges of the blue, green, red and near infrared bands of the foreign Landsat8 optical satellite are highly consistent with the spectral ranges of high-resolution satellites in the three spectral ranges, obtaining an image classifier, and finally classifying optical remote sensing data of the high-resolution satellite based on the image classifier to obtain a water body classification result.
It should be noted that the machine learning classification method adopted in this embodiment may include a classification regression tree method and a random forest method, and those skilled in the art should understand that the method is not limited to these two methods, and other machine learning classification methods should also fall within the scope of the present invention.
Further, fig. 2 shows a tree diagram of a water body classification method based on chromaticity and lightness features, specifically, the classification regression tree method is an intelligent decision tree classification method, and the principle thereof is to construct a machine learning method for model prediction from training data, and finally obtain a decision classification binary tree model method by recursively dividing data and fitting a prediction model in each partition; the random forest method is a classifier integration algorithm based on decision trees, wherein each tree depends on a random vector, and all vectors of the random forest are independently and identically distributed. The random forest is to randomize the column variables and row observations of the data set to generate a plurality of classification numbers, and finally summarize the classification tree results.
Specifically, S8 further includes: cutting and projection conversion are carried out on the digital elevation model data;
hydrologic analysis is carried out on the digital elevation model data subjected to cutting and projection conversion, and river network data in a research area are extracted;
and extracting the height difference of the grid in the research area from the nearest water system based on the river network data and the digital height model data.
In the present embodiment, the present step is intended to extract the height difference (HAND) of the mesh from the nearest water system using the digital elevation model. Firstly, based on the digital elevation model data acquired in S1, clipping and projection conversion are performed on the digital elevation model data; and secondly, performing hydrological analysis operation by using ARCGIS software to realize the extraction of the inland river network in the research area. Specific flow of the river network extraction operation based on ArcGIS can be referred to [ Tang Guoan, Yang Xin and other works of ArcGIS geographic information system spatial analysis experiment course, scientific publishing agency 2006 ]. And thirdly, calculating the elevation difference of the grid distance nearest neighbor water system by combining the extracted river network data and DEM data, and establishing a HAND calculation model based on ArcGIS model builder to realize the rapid calculation of the HAND index of the research area.
Specifically, in S9, due to the influence of mountain shadows, the water body classification result extracted in S7 has a pseudo water body, and needs to be further removed in combination with the topographic features of the research area. S9 eliminates the pseudo water body based on the height difference of the grid to the nearest water system based on S7 and S8. Setting the height difference of the grid from the nearest water system to be less than a certain threshold value as a reasonable range of water body extraction, wherein the threshold value can be set between 30 and 50. And eliminating the false water bodies exceeding the threshold, wherein the operation is to eliminate the existence of the false water bodies caused by mountain shadow in the research area. And carrying out classification post-processing operation on the proposed water body classification result, and carrying out clustering and filtering processing on the classified images. It should be noted that the threshold may be set by the operator, and the specific value is not limited in this embodiment.
Specifically, in S10, based on the water body classification result corrected in S9, the regional basic geographic information data is studied in combination, so as to realize rapid mapping and output of regional water bodies.
In summary, the method described in this embodiment fully utilizes spectral band information of a high-resolution satellite, generates a new band with similar water and land spectral features through spectral transformation, converts three color channels composed of the new band of the spectral transformation, a near-infrared band and a red band into chromaticity, lightness and saturation spaces based on color transformation, provides a water body classification criterion without regional water body sample information, eliminates pseudo water body pattern spots caused by mountain shadow based on the height difference of a grid from a nearest water system as an important criterion, improves extraction accuracy, and can serve application requirements in application fields of regional water body dynamic monitoring, regional water resource dynamic analysis and evaluation, regional drought monitoring and the like.
Fig. 3 shows a water body rapid extraction system based on high-resolution satellite optical remote sensing data according to a second embodiment of the present invention, as shown in fig. 3, the system includes:
the acquisition module is used for acquiring the altitude satellite optical remote sensing data and the digital elevation model data of a research area;
the radiometric calibration module is used for radiometric calibration of the high-resolution satellite optical remote sensing data;
the correction processing module is used for carrying out atmospheric correction processing on the high-resolution satellite optical remote sensing data subjected to radiometric calibration;
the ortho-rectification module is used for carrying out ortho-rectification on the high-resolution satellite optical remote sensing data after atmospheric rectification;
the spectrum transformation processing module is used for carrying out spectrum transformation processing on the high-resolution satellite optical remote sensing data subjected to the ortho-rectification to obtain a spectrum calculation result of the approximate short-wave infrared band characteristic of the high-resolution satellite optical remote sensing data;
the color transformation processing module is used for carrying out color transformation processing on the high-resolution satellite optical remote sensing data subjected to the spectrum transformation processing;
the water body classification result extraction module is used for extracting water body classification from the high-resolution satellite optical remote sensing data subjected to color transformation to obtain a water body classification result;
the calculation module extracts the height difference of the grid in the research area from the nearest neighbor water system based on the digital height model data;
the correction module corrects the water body classification result based on the elevation difference of the grid to the nearest neighbor water system;
and the regional water body drawing module is used for realizing regional water body drawing based on the corrected water body classification result.
The invention is further described by combining example verification, fig. 4 shows an optical image of a high-resolution satellite in the middle of Anhui province, the resolution is 16 meters, the water body distribution condition of the middle area of Anhui province is extracted by the technical scheme of the invention, and in order to verify the precision of the invention, example verification is carried out by combining the water body sample of a google earth acquisition test area and the distribution thereof, and the verification result shows that the total precision of water body extraction can reach 94.6 percent based on the method provided by the invention, the model method has higher precision and reasonable and credible result, and can serve the application requirements of the application fields of regional water body dynamic monitoring, regional water resource dynamic analysis and evaluation, regional drought monitoring and the like.
It should be understood that the above-mentioned embodiments of the present invention are only examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention, and it will be obvious to those skilled in the art that other variations or modifications may be made on the basis of the above description, and all embodiments may not be exhaustive, and all obvious variations or modifications may be included within the scope of the present invention.
Claims (7)
1. A water body rapid extraction method based on high-resolution satellite optical remote sensing data is characterized by comprising the following steps:
acquiring altitude satellite optical remote sensing data and digital elevation model data of a research area;
carrying out radiometric calibration on the high-resolution satellite optical remote sensing data;
atmospheric correction processing is carried out on the high-resolution satellite optical remote sensing data subjected to radiometric calibration;
performing orthorectification on the high-resolution satellite optical remote sensing data subjected to atmospheric rectification processing;
carrying out spectrum transformation processing on the high-resolution satellite optical remote sensing data subjected to the orthometric correction to obtain a spectrum calculation result of the approximate short wave infrared band characteristic of the high-resolution satellite optical remote sensing data;
the spectrum calculation result includes:
obtaining a spectrum calculation result of the near short wave infrared band characteristic through the following formula:
ρSWIR＝(1.0-NDWI)·(ρR+ρNIR)/2.0；
where ρ isSWIRThe result of the spectral calculation, rho, being characteristic of the near-short-wave infrared bandRSurface reflectance values, rho, for the red band of high-resolution satellitesNIRThe surface reflectance value of the near-infrared band of the high-resolution satellite is shown, and NDWI is a normalized difference water body index;
obtaining the normalized difference water body index by the following formula:
NDWI＝(ρG-ρNIR)/(ρG+ρNIR)；
where ρ isGThe surface reflectance value of the green wave band of the high-resolution satellite;
carrying out color transformation processing on the high-resolution satellite optical remote sensing data subjected to the spectrum transformation processing;
the color transformation process includes:
respectively obtaining high-resolution remote sensing data of the high-resolution satellite based on the spectrum calculation result of the approximate short wave infrared band characteristic to perform color transformation, so as to obtain lightness, saturation and chroma;
wherein the lightness is obtained by the following formula:
V＝max(ρSWIR，ρNIR，ρR)；
the saturation is obtained by the following formula:
the chroma is obtained by the following formula:
carrying out water body classification extraction on the high-resolution satellite optical remote sensing data subjected to color transformation processing to obtain a water body classification result;
extracting the height difference of a grid in a research area from a nearest water system based on the digital height model data;
correcting the water body classification result based on the elevation difference of the grid to the nearest neighbor water system;
and realizing regional water body mapping based on the corrected water body classification result.
2. The method of claim 1, wherein the radiometric calibration of the high-resolution satellite optical remote sensing data comprises:
converting the high-resolution satellite optical remote sensing data into radiance;
wherein the radiance is obtained by:
Lλ＝Gain*DN；
in the formula, Gain is a scaling coefficient, and the unit is: w/(m)2Sr μm), DN is the observed value of the satellite-borne sensor, LλFor radiance, the unit is: w/(m)2·sr·μm)。
3. The method of claim 2, further comprising:
converting the radiance into zenith reflectance;
wherein the zenith reflectivity is obtained by the following formula:
in the formula, ρTOA，λIs zenith reflectivity, pi is circumferential ratio, d is distance between day and earth, ESUNλIs the average solar radiation value of the wave band, and the unit is: w/(m)2·μm)，θsThe zenith angle of the sun.
4. The method of claim 1, wherein the orthorectifying the high-resolution satellite optical remote sensing data after the atmospheric rectifying process comprises:
and performing geometric distortion correction, inclination correction and projection difference correction on the high-resolution satellite optical remote sensing data subjected to atmospheric correction processing.
5. The method according to claim 1, wherein the extracting of the water body classification of the high-resolution satellite optical remote sensing data after the color transformation processing to obtain the water body classification result comprises:
respectively obtaining a water body sample point and a non-water body sample point;
sampling the reflectivities of a blue wave band, a green wave band, a red wave band and an approximate short wave infrared band which respectively correspond to the water body sample point and the water body sample point to obtain a training sample set;
training the training sample set by a machine learning method to obtain an image classifier;
and classifying the high-resolution satellite optical remote sensing data based on the image classifier to obtain a water body classification result.
6. The method according to claim 5, wherein extracting elevation differences for a grid to nearest water systems within an area of interest based on the digital elevation model data comprises:
cutting and projection conversion are carried out on the digital elevation model data;
hydrologic analysis is carried out on the digital elevation model data subjected to cutting and projection conversion, and river network data in a research area are extracted;
and extracting the height difference of the grid in the research area from the nearest water system based on the river network data and the digital height model data.
7. A water body rapid extraction system based on high-resolution satellite optical remote sensing data is characterized by comprising:
the acquisition module is used for acquiring the altitude satellite optical remote sensing data and the digital elevation model data of a research area;
the radiometric calibration module is used for radiometric calibration of the high-resolution satellite optical remote sensing data;
the correction processing module is used for carrying out atmospheric correction processing on the high-resolution satellite optical remote sensing data subjected to radiometric calibration;
the ortho-rectification module is used for carrying out ortho-rectification on the high-resolution satellite optical remote sensing data after atmospheric rectification;
the spectrum transformation processing module is used for carrying out spectrum transformation processing on the high-resolution satellite optical remote sensing data subjected to the ortho-rectification to obtain a spectrum calculation result of the approximate short-wave infrared band characteristic of the high-resolution satellite optical remote sensing data;
the spectrum calculation result includes:
obtaining a spectrum calculation result of the near short wave infrared band characteristic through the following formula:
ρSWIR＝(1.0-NDWI)·(ρR+ρNIR)/2.0；
where ρ isSWIRThe result of the spectral calculation, rho, being characteristic of the near-short-wave infrared bandRSurface reflectance values, rho, for the red band of high-resolution satellitesNIRThe surface reflectance value of the near-infrared band of the high-resolution satellite is shown, and NDWI is a normalized difference water body index;
obtaining the normalized difference water body index by the following formula:
NDWI＝(ρG-ρNIR)/(ρG+ρNIR)；
where ρ isGThe surface reflectance value of the green wave band of the high-resolution satellite;
the color transformation processing module is used for carrying out color transformation processing on the high-resolution satellite optical remote sensing data subjected to the spectrum transformation processing;
the color transformation process includes:
respectively obtaining high-resolution remote sensing data of the high-resolution satellite based on the spectrum calculation result of the approximate short wave infrared band characteristic to perform color transformation, so as to obtain lightness, saturation and chroma;
wherein the lightness is obtained by the following formula:
V＝max(ρSWIR，ρNIR，ρR)；
the saturation is obtained by the following formula:
the chroma is obtained by the following formula:
the water body classification result extraction module is used for extracting water body classification from the high-resolution satellite optical remote sensing data subjected to color transformation to obtain a water body classification result;
the calculation module extracts the height difference of the grid in the research area from the nearest neighbor water system based on the digital height model data;
the correction module corrects the water body classification result based on the elevation difference of the grid to the nearest neighbor water system;
and the regional water body drawing module is used for realizing regional water body drawing based on the corrected water body classification result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910299477.7A CN109993237B (en) | 2019-04-15 | 2019-04-15 | Water body rapid extraction method and system based on high-resolution satellite optical remote sensing data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910299477.7A CN109993237B (en) | 2019-04-15 | 2019-04-15 | Water body rapid extraction method and system based on high-resolution satellite optical remote sensing data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109993237A CN109993237A (en) | 2019-07-09 |
| CN109993237B true CN109993237B (en) | 2020-11-10 |
Family
ID=67133687
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910299477.7A Expired - Fee Related CN109993237B (en) | 2019-04-15 | 2019-04-15 | Water body rapid extraction method and system based on high-resolution satellite optical remote sensing data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109993237B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110703244B (en) * | 2019-09-05 | 2023-08-04 | 中国科学院遥感与数字地球研究所 | Method and device for identifying urban water body based on remote sensing data |
| CN111178186A (en) * | 2019-12-16 | 2020-05-19 | 广州地理研究所 | Rice extraction method, device and equipment based on sentinel remote sensing data |
| CN111652882B (en) * | 2020-07-07 | 2020-12-08 | 中国水利水电科学研究院 | Large-scale surface water product drawing precision evaluation method |
| CN111931709B (en) * | 2020-09-17 | 2021-01-05 | 航天宏图信息技术股份有限公司 | Water body extraction method and device for remote sensing image, electronic equipment and storage medium |
| CN112308024A (en) * | 2020-11-23 | 2021-02-02 | 中国水利水电科学研究院 | Water body information extraction method |
| CN112613545B (en) * | 2020-12-17 | 2021-08-31 | 中国农业科学院农业资源与农业区划研究所 | Method for extracting water body based on multispectral remote sensing data |
| CN112733745A (en) * | 2021-01-14 | 2021-04-30 | 北京师范大学 | Cultivated land image extraction method and system |
| CN113096114B (en) * | 2021-04-28 | 2022-07-01 | 成都理工大学 | High-resolution urban water body pattern spot remote sensing extraction method combining morphology and index |
| CN113240648A (en) * | 2021-05-18 | 2021-08-10 | 北京师范大学 | Vegetation growth monitoring and analyzing method and device of multi-temporal visible light image |
| CN113642191B (en) * | 2021-08-25 | 2022-03-22 | 中国水利水电科学研究院 | Short wave infrared-based remote sensing evapotranspiration model construction method |
| CN114112906B (en) * | 2021-10-12 | 2023-11-17 | 中通服咨询设计研究院有限公司 | Water body feature extraction system based on unmanned aerial vehicle low altitude remote sensing and local topography |
| CN114544452B (en) * | 2022-04-25 | 2022-07-26 | 自然资源部第二海洋研究所 | Multi-angle polarized water color remote sensor satellite atmosphere correction method |
| CN117054349B (en) * | 2023-10-11 | 2023-12-26 | 中国水利水电科学研究院 | Water network water quality pressure evaluation method based on remote sensing data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103927455A (en) * | 2014-04-24 | 2014-07-16 | 中国科学院遥感与数字地球研究所 | Land aerosol optical property retrieval method based on Gaofen-1 satellite |
| CN106127133A (en) * | 2016-06-16 | 2016-11-16 | 西安科技大学 | A kind of Clean water withdraw method based on HONGGUANG near infrared spectrum feature space |
| CN109308451A (en) * | 2018-08-09 | 2019-02-05 | 哈尔滨哈船导航技术有限公司 | A kind of high score data information extraction system and method |
-
2019
- 2019-04-15 CN CN201910299477.7A patent/CN109993237B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103927455A (en) * | 2014-04-24 | 2014-07-16 | 中国科学院遥感与数字地球研究所 | Land aerosol optical property retrieval method based on Gaofen-1 satellite |
| CN106127133A (en) * | 2016-06-16 | 2016-11-16 | 西安科技大学 | A kind of Clean water withdraw method based on HONGGUANG near infrared spectrum feature space |
| CN109308451A (en) * | 2018-08-09 | 2019-02-05 | 哈尔滨哈船导航技术有限公司 | A kind of high score data information extraction system and method |
Non-Patent Citations (1)
| Title |
|---|
| 高分系列卫星影像特征及其在太湖生态环境监测中的应用;侍昊等;《南京林业大学学报(自然科学版)》;20160627;第40卷(第6期);第63-68页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109993237A (en) | 2019-07-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109993237B (en) | Water body rapid extraction method and system based on high-resolution satellite optical remote sensing data | |
| Goodenough et al. | Processing Hyperion and ALI for forest classification | |
| Mucher et al. | Land cover characterization and change detection for environmental monitoring of pan-Europe | |
| Li et al. | Principal component analysis of stacked multi-temporal images for the monitoring of rapid urban expansion in the Pearl River Delta | |
| Apan | Land cover mapping for tropical forest rehabilitation planning using remotely-sensed data | |
| Chen et al. | An automated cloud detection method for daily NOAA-14 AVHRR data for Texas, USA | |
| Muramatsu et al. | Pattern decomposition method in the albedo space for Landsat TM and MSS data analysis | |
| Chen et al. | Mapping forest and their spatial–temporal changes from 2007 to 2015 in tropical hainan island by integrating ALOS/ALOS-2 L-Band SAR and landsat optical images | |
| de Colstoun et al. | Improving global scale land cover classifications with multi-directional POLDER data and a decision tree classifier | |
| Chen et al. | Aboveground biomass of salt-marsh vegetation in coastal wetlands: Sample expansion of in situ hyperspectral and Sentinel-2 data using a generative adversarial network | |
| CN114022783A (en) | Satellite image-based water and soil conservation ecological function remote sensing monitoring method and device | |
| CN108898070A (en) | A kind of high-spectrum remote-sensing extraction Mikania micrantha device and method based on unmanned aerial vehicle platform | |
| CN113379759A (en) | Automatic water body extraction method for optical remote sensing satellite image | |
| CN114778483A (en) | Method for correcting terrain shadow of remote sensing image near-infrared wave band for monitoring mountainous region | |
| Tole | An estimate of forest cover extent and change in Jamaica using Landsat MSS data | |
| Fu et al. | Forest cover classification with MODIS images in Northeastern Asia | |
| CN112329790B (en) | Quick extraction method for urban impervious surface information | |
| Solaimani et al. | Land use/cover change detection based on remote sensing data (A case study; Neka Basin) | |
| Bektas Balcik et al. | Determination of magnitude and direction of land use/land cover changes in Terkos Water Basin, Istanbul | |
| Elsharkawy et al. | Improvement in the detection of land cover classes using the Worldview-2 imagery | |
| Davaasuren et al. | Extent and health of mangroves in Lac Bay Bonaire using satellite data | |
| Anggari et al. | Preliminary study of the potential utilization of imagery from LAPAN-A4 satellites | |
| Whiteside et al. | Vegetation map for Magela Creek floodplain using WorldView-2 multispectral image data | |
| Guo et al. | China forest cover extraction based on Google Earth Engine | |
| Yeh et al. | Principal component analysis of stacked multi-temporal images for the monitoring of rapid urban expansion in the Pearl River Delta |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20201110 Termination date: 20210415 |