CN108537795A - A kind of mountain stream information extracting method - Google Patents
A kind of mountain stream information extracting method Download PDFInfo
- Publication number
- CN108537795A CN108537795A CN201810366283.XA CN201810366283A CN108537795A CN 108537795 A CN108537795 A CN 108537795A CN 201810366283 A CN201810366283 A CN 201810366283A CN 108537795 A CN108537795 A CN 108537795A
- Authority
- CN
- China
- Prior art keywords
- area
- image
- research area
- research
- mountain stream
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
- G06T2207/10036—Multispectral image; Hyperspectral image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
- G06T2207/30184—Infrastructure
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of mountain stream information extracting methods, it includes：To 8 OLI remote sensing image datas information of Landsat Landsat, to carry out include radiation calibration, atmospheric correction, inlay and cutting is handled；Calculate complex spectrum threshold value water body index；According to research area's dem data extraction research area's massif shade；According to research area's dem data extraction research area network of waterways, river buffering area is established；Calculate research area's land surface temperature；Calculate research area's surface albedo；Calculate research area normalization building index；To studying the land surface temperature in area, surface albedo and normalization building index figure layer resampling are standardized the building information of debug extraction；Using the closed operation filtering operation in mathematical morphology filter, the tiny river aperture to extracting research area is eliminated and gap filling；The prior art is solved for the technical problems such as mountain stream information extraction accuracy is to be improved.
Description
Technical field
The invention belongs to river information extraction technology more particularly to a kind of mountain stream information extracting methods.
Background technology
Surface water resources is human survival and irreplaceable one of the resource of social development, for the mankind, crops and life
State is most important.The interference of generally existing shade and building causes accidentally to carry in traditional mountain stream information extracting method, carries more
And small water-body leakage such as puies forward at the various problems in the network of waterways.These problems affect the guarantor of mountainous regions investigation and ecological environment
Shield.Therefore, the mountain stream information extracting method for establishing precise and high efficiency be very there is an urgent need for.
Currently, the data that mountain stream information extraction uses mainly have：The TM of Landsat (Landsat), ETM+ and
OLI, the remotely-sensed datas such as high-resolution satellite.Although the higher advantage of the remotely-sensed datas data resolution ratio such as high-resolution satellite,
There is also being difficult to obtain, data volume is big, takes time the problems such as long, of high cost, efficiency is low.Landsat images are due to its data
It obtains relatively easily, cost is relatively low, and application range is relatively extensive.
Common mountain stream information extraction takes method mainly by softwares such as ENVI and ArcGIS, using single band
Threshold method, multiband spectrum-photometric method, water body index method, the methods of novel water body index carry out mountain stream letter to remote sensing image
Breath extracts.
Above method has certain limitation, and without systematically excluding the interference of mountainous region shade and building, mountain area is thin
Small river is asked a question in the presence of leakage, and extraction accuracy is to be improved.
Invention content
The technical problem to be solved by the present invention is to：A kind of mountain stream information extracting method is provided, to solve the prior art
For mountain stream information extraction using the softwares such as ENVI and ArcGIS, using single band threshold method, multiband spectrum-photometric method,
Water body index method, the methods of novel water body index extract remote sensing image progress mountain stream information existing without system
The skills such as ground excludes the interference of mountainous region shade and building, and the tiny river in mountain area is asked a question in the presence of leakage, and extraction accuracy is to be improved
Art problem.
Technical solution of the present invention：
A kind of mountain stream information extracting method, it includes：
Step 1 carries out including radiation calibration, air school to Landsat Landsat 8OLI remote sensing image datas information
Just, it inlays and is handled with cutting；
Step 2 calculates complex spectrum threshold value water body index；
Step 3 studies area's massif shade according to research area's dem data extraction；
Step 4 studies the area network of waterways according to research area's dem data extraction, establishes river buffering area；
Step 5 calculates research area's land surface temperature；
Step 6 calculates research area's surface albedo；
Step 7 calculates research area normalization building index；
Step 8, the land surface temperature to studying area, surface albedo and normalization building index figure layer resampling carry out
Standardization, the building information of debug extraction；
Step 9, using the closed operation filtering operation in mathematical morphology filter, it is small to the tiny river for extracting research area
Hole is eliminated and gap filling.
Remote sensing image data information is carried out including that the method for radiation calibration and atmospheric correction is described in step 1：Radiation is fixed
Calibration method is：Convert the gray value of the remote sensing image data of acquisition to the reflectivity of actual physical meaning；Atmospheric correction
Method is：Scaled values are reduced to earth's surface real information, and ensure true recovery ground-object spectrum information；After atmospheric correction
Image is inlayed and is cut to determine the range in research area.
Described in step 2 calculate complex spectrum threshold value water body index formula be：
In formula：B2 is blue wave band, and b3 is green light band, and b5 is near infrared band, and b6 is shortwave Thermal infrared bands 1, b7
For shortwave Thermal infrared bands 2；α i (i=1,2,3,4,5) are respectively the ratio for the mean value for counting each wave band brightness value (DN) in river.
It is according to the method for research area's dem data extraction research area's massif shade described in step 3：The resolution ratio of 30 meters of selection
Dem data calculates illumination based on imaginary lighting source using ArcGIS softwares to each grid cell of elevation grid map
Value, shadow region when analog image is shot, in conjunction with Shaded histograms threshold value, to extract practical shadow region, profit
Practical shadow region when area's filming image is studied with research area vector boundary file acquisition.
According to research area's dem data extraction research area network of waterways described in step 4, the method for establishing river buffering area is：According to
To determine the position in river, binding area remote sensing image, which is established, studies area river buffering area for research area's dem data extraction network of waterways,
Ensure that river buffering area all covers the mountain stream for studying area's image.
The method for studying area's land surface temperature is calculated described in step 5 is：Research area's land surface temperature passes through air school
Correction method is calculated, and the radiation for calculating the Land surface emissivity and its image wave band 10 (Band10) of downloading image first is bright
Degree is inquired by the official websites NASA and obtains atmospheric profile parameter, and synthermal lower black body radiation brightness is calculated, and image is downloaded in then inverting
Practical surface temperature.
Described in step 6 calculate research area's surface albedo formula be
Albedo=0.356 × b2+0.130 × b4+0.373 × b5+0.085 × b6+0.072 × b7-0.0018
In formula：B2 is research area's image blue wave band, and b4 is research area's image red spectral band, and b5 is that research area's image is closely red
Wave section, b6 are research area's image shortwave Thermal infrared bands 1, and b7 is research area's image shortwave Thermal infrared bands 2.
Described in step 7 calculate research area normalization building index formula be：
In formula：B5 is research area's image near infrared band, b6 is research area's image shortwave Thermal infrared bands 1.
The land surface temperature in pair research area described in step 8, surface albedo and normalization building index figure layer resampling
It is standardized, the method for the building information of debug extraction is：Area's land surface temperature, the earth's surface reflection of light will be studied
Rate and normalization building index figure layer resampling, standardization, respectively according to 50 of the histogram of different figure layers and selection
Training sample threshold value isolates potential construction zone, the figure layer of three potential construction zones is intersected, really
Fixed practical building distributed areas, and the removal error extraction from the mountain stream after the terrestrial object information for excluding to be distributed without the network of waterways
Architecture information.
Advantageous effect of the present invention：
The present invention is based on DEM to simulate shade on the spot, can effectively exclude the influence of massif shade；LST(Land Surface
Temperature), Albedo, NDBI (Normalized Difference Build-up Index) can exclude to build very well
Build object interference；Complex spectrum threshold value water body index (CSTWI) is built, water model is extracted relative to tradition, on the one hand can have
Effect, which eliminates alpine terrain, to be influenced, and the negative effect of shade caused by reduce landform facilitates subsequent shadow removal, another party
Face protrudes mountain area Water-Body Information by the higher wave band of water body reflectivity and the lower band combination of reflectivity, using ratio method,
To be conducive to the identification and extraction of Water-Body Information, complete Water-Body Information can be effectively extracted；It solves the prior art to be directed to
Mountain stream information extraction is using softwares such as ENVI and ArcGIS, using single band threshold method, multiband spectrum-photometric method, water body
Index method, the methods of novel water body index carry out mountain stream information to remote sensing image and extract existing not arrange systematically
Except the interference of mountainous region shade and building, the technologies such as the tiny river in mountain area is asked a question in the presence of leakage, and extraction accuracy is to be improved are asked
Topic.
Specific implementation mode
The present invention provides a kind of mountain stream information extracting methods, include the following steps：
First, data prediction.To Landsat Landsat 8OLI (geographical spatial data cloud (http://
Www.gscloud.cn/ it)) is pre-processed, including radiation calibration, atmospheric correction, inlays and cut.Radiation calibration process is
Convert the gray value of the remote sensing image data of acquisition to the reflectivity of actual physical meaning.Radiation calibration is by establishing digital quantity
Quantitative relationship between change value and radiance value in corresponding visual field reaches the target for eliminating the error that sensor itself generates.
Scaled values are reduced to earth's surface real information by atmospheric correction again, and ensure true recovery ground-object spectrum information.By atmospheric correction it
Image afterwards is inlayed and is cut to determine the range in research area.
Second, structure complex spectrum threshold value water body index (CSTWI).To studying the typical feature in area after data prediction
(forest land, meadow, arable land, construction land, lake (reservoir), river) establishes 100 training samples, and draws typical feature wave spectrum
Curve.Count different each wave band brightness values (DN) of atural object simultaneously.By being calculated with drag：
Wherein b2 is blue wave band, b3 is green light band, b5 is near infrared band, b6 is shortwave Thermal infrared bands 1, b7 is
Shortwave Thermal infrared bands 2；αi(i=1,2,3,4,5) ratio of the mean value of each wave band brightness value (DN) in river is respectively counted.
The range of CSTWI, by choosing 50 river samples, the basic model of river information is gone out in conjunction with statistics with histogram between 0-1
It encloses, the river information less than 1% is averaged as threshold value, multiband combination is carried out, enhance high reflectance using ratio method
Difference between wave band and antiradar reflectivity wave band, to extract mountain stream.
Third, dem data extraction research area's massif shade based on download.Dem data derives from geographical spatial data cloud
(http://www.gscloud.cn/), select 30 meters of resolution ratio dem data.Using ArcGIS softwares, it is based on imaginary photograph
Mingguang City source calculates illumination value, shadow region when analog image is shot, in conjunction with shade to each grid cell of elevation grid map
Histogram threshold value, to extract practical shadow region.Research on utilization area vector boundary file acquisition research area's image is clapped
Practical shadow region when taking the photograph, and the massif shadow regions extracted are removed from the mountain stream tentatively extracted more.
4th, based on research area's dem data extraction research area network of waterways, establish river buffering area.First according to research area DEM
Data extract the network of waterways to determine that the approximate location in river, binding area remote sensing image are established research area river buffering area, ensured
Buffering area all covers the mountain stream for studying area's image.It is further excluded from the river information after removal massif shade
Other terrestrial object informations that no network of waterways distributed area is accidentally extracted.
5th, calculate research area's land surface temperature (LST).Research area's land surface temperature passes through atmospheric correction method meter
It obtains.The radiance for calculating the Land surface emissivity and its image wave band 10 (Band10) of downloading image first, passes through
The official websites NASA (http://atmcorr.gsfc.nasa.gov/) inquiry acquisition atmospheric profile parameter, calculate synthermal lower black matrix
The practical surface temperature of image is downloaded in radiance, then inverting.Vector boundary file cutting based on research area is studied
The surface temperature figure in area.
It wherein downloads image Land surface emissivity (ε) to be calculated by normalized differential vegetation index (NDVI) threshold method, institute
It is with formula：
ε=0.004Pv+0.986
In formula, ε is Land surface emissivity, and Pv is research area's vegetation coverage.Pv is calculated with following formula：
Pv=[(NDVI-NDVISoil)/(NDVIVeg-NDVISoil)]
In formula, NDVI is normalized differential vegetation index, NDVIVegIt is the region of completely vegetative coverage, NDVISoilIt is complete
Exposed soil or NDVI values without vegetative coverage region, Pv is vegetation coverage, based on experience value NDVIVeg=0.70 and NDVISoil=
0.05, i.e., when the NDVI values of some pixel are more than 0.70, Pv values are 1；When NDVI values are less than 0.05, Pv values are 0.
Band10 radiances are to obtain radiance image by carrying out radiation calibration to the Band10 for downloading image.
Synthermal lower black body radiation brightness passes through：
B (Ts)=[Lλ- L ↑-τ (1- ε) L ↓]/τ ε
Transmitance τ, air uplink radiation brightness L in formula ↑, Downward atmospheric long-wave radiation brightness L ↓ tri- parameter can pass through NASA
Official website input is downloaded the relevant informations such as imaging time, center longitude, the relative region air pressure of image and is generated, and the radiation of black matrix is bright
Degree B (Ts) unit is Wm2·sr1·μm1.Thermal infrared radiation brightness L λ are to be using TIRS1 in Landsat8TIRS
(10.6~11.19 μm) of Band10 is by radiation calibration as a result, ε is Land surface emissivity.
After the radiance B (Ts) for estimating black matrix, according to Planck law inverse function, real surface temperature is obtained.
Ts=K2/ln (K1/B (Ts)+1)
In formula, K1, K2 are constant, for Landsat8TIRS Band10 data, (m2 μm of K1=774.89W/
Sr), K2=1321.08K, B (Ts) are the radiance of black matrix；Unit conversion can be to take the photograph after subtracting 273.15 by Ts result of calculations
Family name's degree (DEG C).
6th, calculate research area's surface albedo (Albedo).Surface albedo is to pass through
Albedo=0.356 × b2+0.130 × b4+0.373 × b5+0.085 × b6+0.072 × b7-0.0018
Empirical equation is calculated, and wherein b2 is research area's image blue wave band, b4 is research area's image red spectral band, b5
For research area's image near infrared band, b6 be research area's image shortwave Thermal infrared bands 1, b7 is research area's image shortwave thermal infrared
Wave band 2.It can carry out that research area's albedo is calculated using the wave band calculator (Band math) of ENVI softwares.
7th, calculate research area's normalization building index (NDBI).Normalization building index is to pass through
It is calculated, wherein b5 is research area's image near infrared band, b6 is research area's image shortwave Thermal infrared bands 1.
It can carry out that research area normalization building index is calculated using the wave band calculator (Band math) of ENVI softwares.
8th, study LST, Albedo and NDBI the figure layer resampling in area, standardization, the building of debug extraction
Object information.Area LST, Albedo and NDBI figure layer resamplings, standardization, respectively according to the histogram of different figure layers will be studied
And the 50 training sample threshold values chosen, potential construction zone is isolated, by the figure of three potential construction zones
Layer is intersected, and determines practical building distributed areas, and from the mountain stream after the terrestrial object information for excluding to be distributed without the network of waterways
Further remove the architecture information of error extraction.Different remotely-sensed data guiding principle amounts differ greatly, in order to ensure between different data
Comparativity, need to after calculating study area LST, Albedo and NDBI image be standardized respectively, make its range
Control is between 0-1：
Z=[(A-Amin)÷(Amax-Amin)] × 100%
In formula：A indicates the value before image standardization, AminIndicate the minimum value being worth before image standardization, AmaxIt indicates
The maximum value being worth before is standardized, Z indicates the result after graphics standard.
9th, morphologic filtering processing.Using the closed operation filtering operation in mathematical morphology filter, to extracting research
The tiny river aperture in area is eliminated and gap filling.
Technical solution of the present invention is further illustrated with the data instance of Guizhou Province's Yin Jiangxian Che Jiahe small watersheds in 2016.
First, data prediction.To Landsat8OLI (geographical spatial data cloud (http://www.gscloud.cn/))
It is pre-processed, including radiation calibration, atmospheric correction, inlays and cut.Radiation calibration process is the remote sensing image that will be obtained
The gray value of data is converted into the reflectivity of actual physical meaning.Radiation calibration by establish digital quantization value in corresponding visual field
Quantitative relationship between radiance value reaches the target for eliminating the error that sensor itself generates.Atmospheric correction again will calibration
Value is reduced to earth's surface real information, and ensures true recovery ground-object spectrum information.Image after atmospheric correction is inlayed
With cutting to determine the range in research area.
First, data prediction.To Landsat Landsat 8OLI (geographical spatial data cloud (http://
Www.gscloud.cn/ it)) is pre-processed, including radiation calibration, atmospheric correction, inlays and cut.Radiation calibration process is
Convert the gray value of the remote sensing image data of acquisition to the reflectivity of actual physical meaning.Radiation calibration is by establishing digital quantity
Quantitative relationship between change value and radiance value in corresponding visual field reaches the target for eliminating the error that sensor itself generates.
Scaled values are reduced to earth's surface real information by atmospheric correction again, and ensure true recovery ground-object spectrum information.By atmospheric correction it
Image afterwards is inlayed and is cut to determine the range in research area.
Second, structure complex spectrum threshold value water body index (CSTWI) tentatively extracts mountain stream.After data prediction
Typical feature (forest land, meadow, arable land, construction land, lake (reservoir), river) in research area establishes 100 training samples,
And draw typical feature spectral profile.Count different each wave band brightness values (DN) of atural object simultaneously.In terms of being carried out by drag
It calculates：
Wherein b2 is blue wave band, b3 is green light band, b5 is near infrared band, b6 is shortwave Thermal infrared bands 1, b7 is
Shortwave Thermal infrared bands 2；αi(i=1,2,3,4,5) ratio of the mean value of each wave band brightness value (DN) in river is respectively counted.
The range of CSTWI, by choosing 50 river samples, the basic model of river information is gone out in conjunction with statistics with histogram between 0-1
It encloses, the river information less than 1% is averaged as threshold value, the corresponding each wave band brightness value average value in river of this research is 490,
789,306,173 and 127, then corresponding ratio determine that weight is respectively 0.34,0.66,3.8,2 and 1.Multiband combination is carried out,
Enhance the difference between high reflectance wave band and antiradar reflectivity wave band using ratio method, to tentatively extract mountain stream.
Third, dem data extraction research area's massif shade based on download.Dem data derives from geographical spatial data cloud
(http://www.gscloud.cn/), select 30 meters of resolution ratio dem data.Using ArcGIS softwares, it is based on imaginary photograph
Mingguang City source calculates illumination value, shadow region when analog image is shot, in conjunction with shade to each grid cell of elevation grid map
Histogram threshold value (100), to extract practical shadow region.Research on utilization area vector boundary file acquisition studies area's shadow
Practical shadow region when as shooting, and the massif shadow regions extracted are removed from the mountain stream tentatively extracted more.
4th, based on research area's dem data extraction research area network of waterways, establish river buffering area.First according to research area DEM
Data extract the network of waterways to determine that the approximate location in river, binding area remote sensing image are established research area river buffering area, ensured
Buffering area all covers the mountain stream for studying area's image.Ranging from 400 meters of the river buffering area that this research is established.From going
Other terrestrial object informations extracted are missed except no network of waterways distributed area is further excluded in the river information after massif shade.
5th, calculate research area's land surface temperature (LST).Research area's land surface temperature passes through atmospheric correction method meter
It obtains.The radiance for calculating the Land surface emissivity and its image wave band 10 (Band10) of downloading image first, passes through
The official websites NASA (http://atmcorr.gsfc.nasa.gov/) inquiry acquisition atmospheric profile parameter, calculate synthermal lower black matrix
The practical surface temperature of image is downloaded in radiance, then inverting.Vector boundary file cutting based on research area is studied
The surface temperature figure in area.
It wherein downloads image Land surface emissivity (ε) to be calculated by normalized differential vegetation index (NDVI) threshold method, institute
It is with formula：
ε=0.004Pv+0.986
In formula, ε is than table emissivity, and Pv is research area's vegetation coverage, is calculated with following formula：
Pv=[(NDVI-NDVISoil)/(NDVIVeg-NDVISoil)]
In formula, NDVI is normalized differential vegetation index, NDVIVegIt is the region of completely vegetative coverage, NDVISoil is complete
Exposed soil or NDVI values without vegetative coverage region, Pv is vegetation coverage, based on experience value NDVIVeg=0.70 and NDVISoil=
0.05, i.e., when the NDV I values of some pixel are more than 0.70, Pv values are 1；When NDVI values are less than 0.05, Pv values are 0.
Band10 radiances are to obtain radiance image by carrying out radiation calibration to the Band10 for downloading image.
Synthermal lower black body radiation brightness passes through：
B (Ts)=[Lλ- L ↑-τ (1- ε) L ↓]/τ ε
Transmitance τ, air uplink radiation brightness L in formula ↑, Downward atmospheric long-wave radiation brightness L ↓ tri- parameter can pass through
The input of the official websites NASA is downloaded the relevant informations such as imaging time, center longitude, the relative region air pressure of image and is generated, the spoke of black matrix
It is Wm to penetrate brightness B (Ts) unit2·sr1·μm1.Thermal infrared radiation brightness L λ are to utilize TIRS1 in Landsat8TIRS
That is (10.6~11.19 μm) of Band10 is by radiation calibration as a result, ε is than table emissivity.
After the radiance B (Ts) for estimating black matrix, according to Planck law inverse function, real surface temperature is obtained.
Ts=K2/ln (K1/B (Ts)+1)
In formula, K1, K2 are constant, for Landsat8TIRS Band10 data, (m2 μm of K1=774.89W/
Sr), K2=1321.08K, B (Ts) are the radiance of black matrix；Unit conversion can be to take the photograph after subtracting 273.15 by Ts result of calculations
Family name's degree (DEG C).
6th, calculate research area's surface albedo (Albedo).Surface albedo is to pass through
Albedo=0.356 × b2+0.130 × b4+0.373 × b5+0.085 × b6+0.072 × b7-0.0018 experiences
Formula is calculated, and wherein b2 is research area's image blue wave band, b4 is research area's image red spectral band, b5 is research area's image
Near infrared band, b6 are research area's image shortwave Thermal infrared bands 1, b7 is research area's image shortwave Thermal infrared bands 2.It utilizes
The wave band calculator (Band math) of ENVI softwares can carry out that research area's albedo is calculated.
7th, calculate research area's normalization building index (NDBI).Normalization building index is to pass through
It is calculated, wherein b5 is research area's image near infrared band, b6 is research area's image shortwave Thermal infrared bands 1.
Index is built in the normalization that research area can be carried out being calculated using the wave band calculator (Band math) of ENVI softwares.
8th, LST, Albedo and NDBI the figure layer resampling in area are studied, standardization excludes the buildings extracted more
Information.To study area LST, Albedo and NDBI figure layer resamplings, standardization, respectively according to the histogram of different figure layers and
The 50 training sample threshold values chosen, isolate potential construction zone, by the figure layer of three potential construction zones
Intersected, determine practical building distributed areas, and from the mountain stream after the terrestrial object information that exclusion is distributed without the network of waterways into
The one step removal architecture informations extracted more.Due to differing greatly for different remotely-sensed data guiding principle amounts, in order to ensure between different data
Comparativity, need to after calculating study area LST, Albedo and NDBI image be standardized respectively, make its range
Control is between 0-1：
Z=[(A-Amin)÷(Amax-Amin)] × 100%
In formula：A indicates the value before image standardization, AminIndicate the minimum value of the value before image standardization, AmaxTable
Show that the maximum value of the value before standardization, Z indicate the result after graphics standard.
9th, morphologic filtering processing.Using the closed operation filtering operation in mathematical morphology filter, ground to what is extracted
Study carefully the tiny river in area and carries out aperture elimination and gap filling processing.
It can be seen that of the invention：
(1) Landsat Landsat 8OLI/TIRS and digital elevation model (DEM) remotely-sensed data are as intermediate resolution
The data acquisition of (30m) is relatively easy, and cost is relatively low, and revisiting period is short, can be widely applied to mountain stream information extraction.
(2) size of the width of mountain stream extraction, the accuracy of small water-body extraction and extraction area is for mountain area water
The protection etc. of resource investigation and ecological environment has a major impact, this research establishes a set of six step of mountain stream information extraction
Method solves river present in traditional river information extraction method and accidentally carries, leaks and carry and ask a question more, with traditional Clean water withdraw
Method is opposite, and precision improves 3.5%.Key step is as follows：1. data preparation and pretreatment：Landsat 8OLI/TIRS remote sensing
Image and digital elevation model (DEM)；2. building complex spectrum threshold value water body index (CSTWI)；3. cloudy based on DEM extraction massifs
Shadow；4. extracting the network of waterways based on DEM, water system buffering area is established to determine river home position range；5. calculating surface temperature, the reflection of light
Rate and normalization building standard of indexization processing are simultaneously removed it by separation；6. morphologic filtering is handled.
(3) Water-Body Information is extracted using structure complex spectrum threshold value water body index (CSTWI), with single band threshold method, more
The tradition Clean water withdraw method such as wave band spectrum-photometric method, water body index is compared, and using multiwave information, carries out band combination,
Water-Body Information is protruded by ratio operation, is conducive to the influence for eliminating landform, the result of extraction is more precisely reliable；
(4) have that index is simple, is easily obtained using LST, Albedo and NDBI characterization building information, method is accurate,
Quick feature；Reduce the possibility artificially judged by accident, while substantially reducing interpretation time and labor cost, the row of also improving
Except the precision and efficiency of river nearby buildings；
(5) during being based on research area DEM massif shadow extractions, using ArcGIS softwares, it is based on imaginary illumination light
Source calculates illumination value to each grid cell of elevation grid map, and shade when simulation remote sensing image is shot is true in conjunction with histogram
Determine threshold value, operation is easy, and efficiently, can effectively exclude the influence of massif shade.
Claims (9)
1. a kind of mountain stream information extracting method, it includes：
Step 1 carries out including radiation calibration, atmospheric correction, edge to 8 OLI remote sensing image datas information of Landsat Landsat
It is embedding to be handled with cutting；
Step 2 calculates complex spectrum threshold value water body index；
Step 3 studies area's massif shade according to research area's dem data extraction；
Step 4 studies the area network of waterways according to research area's dem data extraction, establishes river buffering area；
Step 5 calculates research area's land surface temperature；
Step 6 calculates research area's surface albedo；
Step 7 calculates research area normalization building index；
Step 8, the land surface temperature to studying area, surface albedo and normalization building index figure layer resampling carry out standard
Change is handled, the building information of debug extraction；
Step 9, using the closed operation filtering operation in mathematical morphology filter, the tiny river aperture to extracting research area disappears
It removes and gap filling.
2. a kind of mountain stream information extracting method according to claim 1, it is characterised in that：To remote sensing described in step 1
Image data information carries out including that the method for radiation calibration and atmospheric correction is：The method of radiation calibration is：By the remote sensing of acquisition
The gray value of image data is converted into the reflectivity of actual physical meaning；The method of atmospheric correction is：Scaled values are reduced to ground
Table real information, and ensure true recovery ground-object spectrum information；Image after atmospheric correction is inlayed and is cut with true
Surely the range in area is studied.
3. a kind of mountain stream information extracting method according to claim 1, it is characterised in that：It is calculated described in step 2 multiple
Close spectrum threshold water body index formula be：
In formula：B2 is blue wave band, and b3 is green light band, and b5 is near infrared band, and b6 is shortwave Thermal infrared bands 1, and b7 is short
Wave Thermal infrared bands 2；αi(i=1,2,3,4,5) ratio of the mean value of each wave band brightness value (DN) in river is respectively counted.
4. a kind of mountain stream information extracting method according to claim 1, it is characterised in that：Basis described in step 3 is ground
Studying carefully the method that area's massif shade is studied in area's dem data extraction is：30 meters of resolution ratio dem data of selection, using ArcGIS softwares,
Illumination value, shadow region when analog image is shot are calculated to each grid cell of elevation grid map based on imaginary lighting source
Domain, in conjunction with Shaded histograms threshold value, to extract practical shadow region, research on utilization area vector boundary file acquisition is ground
Study carefully practical shadow region when area's filming image.
5. a kind of mountain stream information extracting method according to claim 1, it is characterised in that：Basis described in step 4 is ground
Study carefully the area dem data extraction research area network of waterways, the method for establishing river buffering area is：According to research area's dem data extraction the network of waterways with
It determines that research area river buffering area is established in the position in river, binding area remote sensing image, ensures that river buffering area will study area
The mountain stream of image all covers.
6. a kind of mountain stream information extracting method according to claim 1, it is characterised in that：It calculates and grinds described in step 5
The method for studying carefully area's land surface temperature is：Research area's land surface temperature is calculated by atmospheric correction method, is calculated first
The Land surface emissivity of image and its radiance of image wave band 10 (Band10) are downloaded, is obtained by the inquiry of the official websites NASA big
Gas sectional parameter calculates synthermal lower black body radiation brightness, and the practical surface temperature of image is downloaded in then inverting.
7. a kind of mountain stream information extracting method according to claim 1, it is characterised in that：It calculates and grinds described in step 6
The formula for studying carefully area's surface albedo is
Albedo=0.356 × b2+0.130 × b4+0.373 × b5+0.085 × b6+0.072 × b7-0.0018
In formula：B2 is research area's image blue wave band, and b4 is research area's image red spectral band, and b5 is research area's image near-infrared wave
Section, b6 are research area's image shortwave Thermal infrared bands 1, and b7 is research area's image shortwave Thermal infrared bands 2.
8. a kind of mountain stream information extracting method according to claim 1, it is characterised in that：It calculates and grinds described in step 7
Study carefully area normalization building index formula be：
In formula：B5 is research area's image near infrared band, b6 is research area's image shortwave Thermal infrared bands 1.
9. a kind of mountain stream information extracting method according to claim 1, it is characterised in that：Pair research described in step 8
The land surface temperature in area, surface albedo and normalization building index figure layer resampling are standardized, debug
The method of the building information of extraction is：Area's land surface temperature, surface albedo and normalization building index figure layer will be studied
Resampling, standardization are isolated respectively according to the histogram of different figure layers and 50 training sample threshold values of selection
Potential construction zone intersects the figure layer of three potential construction zones, determines practical building distributed areas, and
The architecture information of error extraction is removed from the mountain stream after the terrestrial object information for excluding to be distributed without the network of waterways.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810366283.XA CN108537795A (en) | 2018-04-23 | 2018-04-23 | A kind of mountain stream information extracting method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810366283.XA CN108537795A (en) | 2018-04-23 | 2018-04-23 | A kind of mountain stream information extracting method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108537795A true CN108537795A (en) | 2018-09-14 |
Family
ID=63477501
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810366283.XA Pending CN108537795A (en) | 2018-04-23 | 2018-04-23 | A kind of mountain stream information extracting method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108537795A (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109801304A (en) * | 2019-01-11 | 2019-05-24 | 重庆邮电大学 | A kind of building index construction and construction area rapid extracting method inhibiting bare area |
| CN110147746A (en) * | 2019-05-09 | 2019-08-20 | 武汉大学 | Method and system based on the minimum and maximum possible surface water body range of Sentinel-2 image rapidly extracting |
| CN111046613A (en) * | 2019-02-21 | 2020-04-21 | 北京大学 | Optimal river channel calculation method based on path tracking and river network extraction method based on multi-temporal remote sensing image |
| CN111339909A (en) * | 2020-02-24 | 2020-06-26 | 南京大学 | Extraction algorithm of surface linear water body based on multi-scale region growth |
| CN111693463A (en) * | 2020-06-10 | 2020-09-22 | 中国科学院空天信息创新研究院 | Antarctic peninsula optimized lichen coverage index extraction method |
| CN111753680A (en) * | 2020-06-09 | 2020-10-09 | 生态环境部卫星环境应用中心 | High-resolution satellite data-based river ecological flow guarantee degree remote sensing rapid discrimination method |
| CN111858808A (en) * | 2020-07-17 | 2020-10-30 | 重庆市勘测院 | Automatic evaluation method for topographic map characteristic ground object precision based on massive real measuring points |
| CN111862134A (en) * | 2020-07-13 | 2020-10-30 | 中国科学院南京地理与湖泊研究所 | Method for extracting offshore aquaculture pond based on Sentinel-1 and GEE |
| CN112016391A (en) * | 2020-07-16 | 2020-12-01 | 珠海欧比特宇航科技股份有限公司 | Fishpond identification method, fishpond identification system and fishpond identification medium based on high-resolution satellite remote sensing image |
| CN112270225A (en) * | 2020-10-15 | 2021-01-26 | 天津大学 | Canyon type high-sand-content meandering river remote sensing image extraction method |
| CN112729145A (en) * | 2020-12-21 | 2021-04-30 | 兰州大学 | Method for determining relation between landslide displacement and reservoir area water level fluctuation |
| CN112990262A (en) * | 2021-02-08 | 2021-06-18 | 内蒙古大学 | Integrated solution system for monitoring and intelligent decision of grassland ecological data |
| CN113469127A (en) * | 2021-07-23 | 2021-10-01 | 平安国际智慧城市科技股份有限公司 | River sensitivity grading method, device, equipment and storage medium |
| CN113610714A (en) * | 2021-05-12 | 2021-11-05 | 珠江水利委员会珠江水利科学研究院 | Estuary and coast hydrodynamic force information enhancement display method based on remote sensing |
| CN115841628A (en) * | 2022-12-12 | 2023-03-24 | 中国水利水电科学研究院 | High-precision irrigation district water consumption information automatic interpretation system |
| CN116385883A (en) * | 2023-04-13 | 2023-07-04 | 珠江水利委员会珠江水利科学研究院 | Unmanned plane mountain shadow area vegetation coverage correction method, equipment and medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140119639A1 (en) * | 2012-10-31 | 2014-05-01 | Microsoft Corporation | Water-body classification |
| CN105046087A (en) * | 2015-08-04 | 2015-11-11 | 中国资源卫星应用中心 | Water body information automatic extraction method for multi-spectral image of remote sensing satellite |
-
2018
- 2018-04-23 CN CN201810366283.XA patent/CN108537795A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140119639A1 (en) * | 2012-10-31 | 2014-05-01 | Microsoft Corporation | Water-body classification |
| CN105046087A (en) * | 2015-08-04 | 2015-11-11 | 中国资源卫星应用中心 | Water body information automatic extraction method for multi-spectral image of remote sensing satellite |
Non-Patent Citations (4)
| Title |
|---|
| 万建鹏: "基于Landsat_TM_ETM_影像鄱阳湖地区水体提取研究", 《中国优秀硕士学位论文全文数据库》 * |
| 孙亚勇等: "基于星载SAR 数据的山区水体提取方法研究", 《中国水利水电科学研究院学报》 * |
| 李艳华等: "基于国产GF-1遥感影像的山区细小水体提取方法研究", 《资源科学》 * |
| 毛莹莹等: "基于面向对象法的Landsat 8 影像山区细小水体提取方法", 《亚热带资源与环境学报》 * |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109801304A (en) * | 2019-01-11 | 2019-05-24 | 重庆邮电大学 | A kind of building index construction and construction area rapid extracting method inhibiting bare area |
| CN109801304B (en) * | 2019-01-11 | 2022-12-27 | 重庆邮电大学 | Building index construction and building area rapid extraction method for inhibiting bare land |
| CN111046613A (en) * | 2019-02-21 | 2020-04-21 | 北京大学 | Optimal river channel calculation method based on path tracking and river network extraction method based on multi-temporal remote sensing image |
| CN110147746A (en) * | 2019-05-09 | 2019-08-20 | 武汉大学 | Method and system based on the minimum and maximum possible surface water body range of Sentinel-2 image rapidly extracting |
| CN110147746B (en) * | 2019-05-09 | 2020-11-17 | 武汉大学 | Method and system for rapidly extracting maximum and minimum possible surface water body ranges based on Sentinel-2 image |
| CN111339909A (en) * | 2020-02-24 | 2020-06-26 | 南京大学 | Extraction algorithm of surface linear water body based on multi-scale region growth |
| CN111753680A (en) * | 2020-06-09 | 2020-10-09 | 生态环境部卫星环境应用中心 | High-resolution satellite data-based river ecological flow guarantee degree remote sensing rapid discrimination method |
| CN111753680B (en) * | 2020-06-09 | 2021-03-30 | 生态环境部卫星环境应用中心 | High-resolution satellite data-based river ecological flow guarantee degree remote sensing rapid discrimination method |
| CN111693463A (en) * | 2020-06-10 | 2020-09-22 | 中国科学院空天信息创新研究院 | Antarctic peninsula optimized lichen coverage index extraction method |
| CN111693463B (en) * | 2020-06-10 | 2022-10-11 | 中国科学院空天信息创新研究院 | Antarctic peninsula optimized lichen coverage index extraction method |
| CN111862134A (en) * | 2020-07-13 | 2020-10-30 | 中国科学院南京地理与湖泊研究所 | Method for extracting offshore aquaculture pond based on Sentinel-1 and GEE |
| CN111862134B (en) * | 2020-07-13 | 2023-11-07 | 中国科学院南京地理与湖泊研究所 | Offshore aquaculture pond extraction method based on Sentinel-1 and GEE |
| CN112016391A (en) * | 2020-07-16 | 2020-12-01 | 珠海欧比特宇航科技股份有限公司 | Fishpond identification method, fishpond identification system and fishpond identification medium based on high-resolution satellite remote sensing image |
| CN112016391B (en) * | 2020-07-16 | 2023-12-08 | 珠海欧比特卫星大数据有限公司 | Fishpond identification method, system and medium based on high-resolution satellite remote sensing image |
| CN111858808A (en) * | 2020-07-17 | 2020-10-30 | 重庆市勘测院 | Automatic evaluation method for topographic map characteristic ground object precision based on massive real measuring points |
| CN111858808B (en) * | 2020-07-17 | 2023-10-31 | 重庆市勘测院 | Automatic evaluation method for feature and ground object precision of topographic map based on massive actual measurement points |
| CN112270225A (en) * | 2020-10-15 | 2021-01-26 | 天津大学 | Canyon type high-sand-content meandering river remote sensing image extraction method |
| CN112729145B (en) * | 2020-12-21 | 2022-03-11 | 兰州大学 | Method for determining relation between landslide displacement and reservoir area water level fluctuation |
| CN112729145A (en) * | 2020-12-21 | 2021-04-30 | 兰州大学 | Method for determining relation between landslide displacement and reservoir area water level fluctuation |
| CN112990262A (en) * | 2021-02-08 | 2021-06-18 | 内蒙古大学 | Integrated solution system for monitoring and intelligent decision of grassland ecological data |
| CN113610714A (en) * | 2021-05-12 | 2021-11-05 | 珠江水利委员会珠江水利科学研究院 | Estuary and coast hydrodynamic force information enhancement display method based on remote sensing |
| CN113469127A (en) * | 2021-07-23 | 2021-10-01 | 平安国际智慧城市科技股份有限公司 | River sensitivity grading method, device, equipment and storage medium |
| CN113469127B (en) * | 2021-07-23 | 2024-06-07 | 平安国际智慧城市科技股份有限公司 | River sensibility grading method, device, equipment and storage medium |
| CN115841628A (en) * | 2022-12-12 | 2023-03-24 | 中国水利水电科学研究院 | High-precision irrigation district water consumption information automatic interpretation system |
| CN115841628B (en) * | 2022-12-12 | 2023-09-19 | 中国水利水电科学研究院 | Automatic interpretation system for high-precision irrigation area water consumption information |
| CN116385883A (en) * | 2023-04-13 | 2023-07-04 | 珠江水利委员会珠江水利科学研究院 | Unmanned plane mountain shadow area vegetation coverage correction method, equipment and medium |
| CN116385883B (en) * | 2023-04-13 | 2024-01-05 | 珠江水利委员会珠江水利科学研究院 | Unmanned plane mountain shadow area vegetation coverage correction method, equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108537795A (en) | A kind of mountain stream information extracting method | |
| CN107730527B (en) | Remote sensing satellite image-based plateau region ice lake extraction method | |
| CN104361338B (en) | A kind of peat bog information extracting method based on ENVISAT ASAR, Landsat TM and dem data | |
| CN101114023A (en) | Lake and marshland flooding remote sense monitoring methods based on model | |
| CN110136194A (en) | Snow Cover measuring method based on satellite-borne multispectral remotely-sensed data | |
| CN103824077A (en) | Urban impervious layer rate information extraction method based on multi-source remote sensing data | |
| Li et al. | Estimation of chlorophyll-a concentrations in a highly turbid eutrophic lake using a classification-based MODIS land-band algorithm | |
| Jiang et al. | A novel index to detect green-tide using UAV-based RGB imagery | |
| CN103914692A (en) | Remote sensing recognition method for surface water systems around coal mine | |
| Wang et al. | Water body extraction from LANDSAT ETM+ image using MNDWI and KT transformation | |
| CN109784251A (en) | Small water remote sensing recognition method along high-speed rail | |
| CN108592888A (en) | A kind of Residential area extraction method | |
| CN114821343A (en) | Mangrove remote sensing rapid and accurate extraction method based on cloud platform | |
| CN104778668A (en) | Optical remote sensing image thin cloud removing method based on visible light band spectrum statistical characteristics | |
| CN113420497A (en) | Remote sensing estimation method for total phosphorus concentration of turbid lake | |
| CN109211798A (en) | A kind of annual sea ice distributed intelligence extracting method based on remote sensing image spectral signature | |
| Yuan et al. | Two drifting paths of Sargassum bloom in the Yellow Sea and East China Sea during 2019–2020 | |
| CN112613545B (en) | Method for extracting water body based on multispectral remote sensing data | |
| CN112270225A (en) | Canyon type high-sand-content meandering river remote sensing image extraction method | |
| CN111667432B (en) | Remote sensing image shadow removing method based on physical model | |
| Meng et al. | Monitoring human-induced surface water disturbance around Taihu Lake since 1984 by time series Landsat images | |
| CN110378290A (en) | A kind of cloudy optical image data flood Water-Body Information rapid extracting method and system | |
| CN114399514A (en) | Mountain water extraction method based on Sentinel-2 image | |
| Zhang et al. | An integrated approach to reconstructing snow cover under clouds and cloud shadows on Sentinel-2 Time-Series images in a mountainous area | |
| Hogda et al. | Mapping of air pollution effects on the vegetation cover in the Kirkenes-Nikel area using remote sensing |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180914 |