CN105116464A - Polar sea ice melting pool extraction method based on neural network model - Google Patents
Polar sea ice melting pool extraction method based on neural network model Download PDFInfo
- Publication number
- CN105116464A CN105116464A CN201510493381.6A CN201510493381A CN105116464A CN 105116464 A CN105116464 A CN 105116464A CN 201510493381 A CN201510493381 A CN 201510493381A CN 105116464 A CN105116464 A CN 105116464A
- Authority
- CN
- China
- Prior art keywords
- melt pool
- ice
- neural network
- wave band
- snow
- 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
Abstract
The present invention relates to a polar sea ice melting pool extraction method based on a neural network model, belonging to the field of a remote sensing information automatic extraction method. With the consideration of the characteristic that MODIS surface reflectance data contains different types of surface spectral combinations, according to the difference of the spectral response curves of polar area ice, water and a melting pool, combined with a pixel unmixing principle, the polybasic linear equation group between each wave band reflectance and the proportions of the polar area ice, the water and the melting pool is established, and the constraints of an equation group solution is added. The neural network method is used to optimize and rapidly solve the linear equation group, and the proportions of the polar area ice, the water and the melting pool. The melting pool area is relative to a sea ice surface excluding water, thus a relative melting pool coverage ratio is extracted further, and the fully automatic extraction of a polar area sea ice melting pool range is realized.
Description
Technical field
The present invention relates to one
based on the polar region sea ice melt pool extracting method of neural network model, belong to sensor information method for full automatic extraction field.
Background technology
In sea ice melting process, melt water collects formation melt pool at sea ice surface pocket.Because the albedo of melt pool is between sea ice and seawater, extract melt pool coverage rate significant for the thermal procession of the accurately absorption of calculating solar radiation in ocean, polar region, understanding polar region gas-ice-Hai coupled system.
The importance outstanding behaviours of melt pool is in its physical characteristics.Melt pool is exactly pond on ice, so melt pool has all character of seawater, is especially its albedo, albedo determines that sea ice absorbs a how many standard of solar radiation energy, and albedo is high, then the solar radiation energy absorbed is just few, melt just slow, otherwise melt just fast.And the surface albedo of seawater is 10%---15%, the albedo of sea ice is 90%, and difference is very large.If a certain block of sea ice there is a little melt pool, this melt pool is likely dynamics or the Thermodynamic effect formation of seawater, so on this block sea ice due to the significantly minimizing of albedo, more solar radiation energy can be absorbed, especially seldom occur again freezing situation in summer, the thawing of this block sea ice will be accelerated.
Summer often contains a large amount of melt pool regions in the remote sensing image of polar region, and relevant data also show, the albedo decrease to some degree of arctic summer marginal ice, has confirmed melt pool and has made sea ice melt the phenomenon of aggravation.Obtaining the related data of melt pool, having great scientific meaning to analyzing the response of sea ice to global warming better.
Coverage rate is an important parameter in melt pool simulation, is also the main direction of studying of space remote sensing in melt pool research, but remote sensing algorithm relevant is at present still based on an example experiment of local, is not applicable to wider melt pool Remotely sensed acquisition algorithm.Carried out the observation of numerous sea ice surface and ship base to walk boat observe in different regions, the arctic, to study the size of the reflectivity of melt pool, spectral signature and melt pool and distribution etc.Although have higher observation accuracy rate and precision, be confined to course line, the Time and place yardstick of observed result is all extremely limited, only can reflect that course line relates to the melt pool situation in region.Need a kind of remote sensing technique extracting polar region sea ice melt pool in long period sequence and spatial dimension.
Summary of the invention
The technical problem to be solved in the present invention is: overcome the above-mentioned defect of prior art, a kind of sea ice melt pool extracting method based on neural network model is provided, the coating ratio of ice and snow, water body and melt pool is determined in the remote sensing images of arctic regions, multivariate linear equations based on Pixel Unmixing Models is resolved by rapid Optimum, reduce calculation cost and ensure precision, completing the extraction of polar region sea ice melt pool scope.
In order to solve the problems of the technologies described above, the technical scheme that the present invention proposes is: a kind of
based on the polar region sea ice melt pool extracting method of neural network model, its step comprises:
The first step, MODIS surface reflectivity product MOD09A1 is converted to polar region azimuthal projection by sinusoidal projection;
Second step, in conjunction with the land mask in MOD09A1 product and cloud mask data, extract the image of target area, and choose λ
1, λ
2, λ
3the surface reflectivity value of wave band;
3rd step, spectral response curve according to ice and snow, water body and melt pool, obtain three respectively at λ
1, λ
2, λ
3the standard spectrum reflectance value of wave band:
At λ
1wave band: the spectral reflectivity r of ice and snow
i(λ
1)=0.95, the spectral reflectivity r of water body
w(λ
1)=0.08, the spectral reflectivity r of melt pool
m(λ
1)=0.16;
At λ
2wave band: the spectral reflectivity r of ice and snow
i(λ
2)=0.87, the spectral reflectivity r of water body
w(λ
2)=0.08, the spectral reflectivity r of melt pool
m(λ
2)=0.07;
At λ
3wave band: the spectral reflectivity r of ice and snow
i(λ
3)=0.95, the spectral reflectivity r of water body
w(λ
3)=0.08, the spectral reflectivity r of melt pool
m(λ
3)=0.22;
4th step, to each pixel in image, according to wherein ice and snow, water body and melt pool proportion S
i, S
w, S
mwith the relation of three's spectral reflectivity, set up multivariate linear equations;
S
i*r
i(λ
1)+S
w*r
w(λ
1)+S
m*r
m(λ
1)＝R(λ
1)
S
i*r
i(λ
2)+S
w*r
w(λ
2)+S
m*r
m(λ
2)＝R(λ
2)
S
i*r
i(λ
3)+S
w*r
w(λ
3)+S
m*r
m(λ
3)＝R(λ
3)
S
i+S
w+S
m＝1
In formula, R (λ
1), R (λ
2), R (λ
3) be that pixel is at λ respectively
1, λ
2, λ
3the reflectance value of wave band;
5th step, from image, choose the pixel of 5%-10% quantity, using the training sample as artificial neural network, utilize the system of equations of the pixel direct solution Problem with Some Constrained Conditions chosen, the solution obtained is ice and snow in each pixel, water body and melt pool proportion, as the training sample set of artificial neural network;
6th step, structure neural network model, input layer is the λ of MODIS reflectivity data
1, λ
2, λ
3wave band; Output layer is the image of ice and snow, water body and melt pool proportion;
7th step, operation neural network model, rapid Optimum solving equation group, acquisition result is ice and snow, water body and the melt pool ratio S shared by respectively in pixel
i, S
w, S
m;
8th step, pass through formula S
m* (1-S
w), the melt pool calculated in target area image covers relative scale, covers relative scale draw the melt pool scope distribution of acquisition target area according to melt pool
figure.
The data source that the inventive method uses is surface reflectivity 8 days generated datas (MOD09A1) of the Moderate Imaging Spectroradiomete (MODIS) of open download.MOD09A1 comprises 7 wave bands, wavelength 459-2155nm, spatial resolution 500m.Data area, between 0 to 1, illustrates the surface reflectivity value on this wave band.In sea ice region, polar region, in image, the surface reflectivity of each pixel is made up of jointly the reflectance value of the ice and snow in pixel, melt pool and water body.
The present invention is based on the sea ice melt pool extracting method of neural network model, also there is following further feature:
1, before the pre-service of first step remote sensing image, in remote sensing image processing software, first determine projection information and the geographical location information of MODIS surface reflectivity product MOD09A1 image, the sinusoidal projection of former image is converted to polar region azimuthal projection.
2, λ in second step
1the wavelength coverage of wave band is 620-670nm, λ
2the wavelength coverage of wave band is 841-876nm, λ
3the wavelength coverage of wave band is 459-479nm.
3, in the 5th step, quasi-Newton method (BFGS algorithm, i.e. Broyden-Fletcher-Goldfarb-Shanno algorithm) solving equation group is used for a small amount of pixel, obtain the ratio shared by ice and snow, water body and melt pool difference in each pixel.
4, in the 6th step, neural network model is multilayer feedforward perceptron, and be made up of an input layer, two-layer hidden layer and an output layer, ground floor hidden layer comprises 9 nodes, and second layer hidden layer comprises 27 nodes.
5, running the optimum configurations that neural network model uses in the 7th step is: the RMS error amount of training stop condition is 0.1; The iterations of training is 10000 times.
6, when the 8th step calculates melt pool covering relative scale, melt pool proportion in pixel is converted into the value of relatively non-coverage of water, draws the distribution of melt pool scope
figuretime, be divided into some grades according between melt pool coating ratio value location, and compose with the gray scale of different brackets formation image.
The invention has the beneficial effects as follows:
The spatial and temporal distributions information obtaining polar region sea ice melt pool is the work of a far reaching research and technology.Present invention achieves
based on the polar region sea ice melt pool extracting method of neural network model, utilize the difference of dissimilar earth's surface spectral signature, by the multivariate linear equations of neural network model rapid solving pixel analysis, determine ice and snow in image, water body and melt pool proportion, extract the distribution of melt pool scope.Compare with traditional supervised classification method with visual interpretation, automaticity of the present invention, classification speed and precision all greatly improve.Concrete beneficial effect is as follows:
First, the present invention is successfully extracted polar region sea ice melt pool scope distributed intelligence, can be applied to polar region sea ice variation monitoring further from trend prediction, the different Snow and Ice Albedo Changeement melted under state, sea ice to researchs such as the response mechanisms of Global climate change.
The second, the data acquisition that the present invention uses is convenient, downloads disclosed MODIS data, can carry out polar region sea ice melt pool research work by application, need not purchase data specially.
3rd, the present invention utilizes ice and snow, water body and melt pool in the difference of remote sensing images polishing wax feature, set up the quantitative relationship of respective proportion and reflectivity based on Decomposition of Mixed Pixels theory, automatically complete the remote sensing image classification under mixed pixel model, accurately extract melt pool coating ratio information.Overall process, without the need to manpower intervention, can directly be classified after determining neural network model and parameter thereof.
4th, the present invention constructs neural network model and calculates to complete Decomposition of Mixed Pixels, by original MODIS wave band data input neural network, and the multivariate linear equations of automatic calculation Problem with Some Constrained Conditions, the i.e. image of exportable ice and snow, water body and melt pool proportion.Process optimization, calculation process is quick, and the precision of result of calculation meets the requirements.The method also has a heavy meaning: owing to optimizing and accelerating calculating process, can complete the classification process of a large amount of remotely-sensed data, realizes on a large scale, the sea ice melt pool extraction of long-term sequence.
Accompanying drawing explanation
Below in conjunction with
accompanying drawingthe present invention is further illustrated:
fig. 1polar region sea ice melt pool based on neural network model extracts flow process
figure.
fig. 2experimental example region wave band 1 image (500 × 500 picture dots).
fig. 3the standard spectrum reflectance curve of sea ice region, polar region ice and snow, water body and melt pool.
fig. 4artificial Neural Network Structures
figure.
fig. 5melt pool covers relative scale
figure.
Embodiment
This experimental example image data is MODIS surface reflectivity product MOD09A1, download from NASA website (http://ladsweb.nascom.nasa.gov/data/search.html) and obtain, longitude and latitude scope is 66 ° of N-68 ° of N, 74 ° of W-78 ° of W, the time is on June 26th, 2014 (Julian date the 169th day)
as Fig. 1for the flow process of this experimental example
figure, comprise following content based on the concrete implementation step of the sea ice melt pool extracting method of neural network model:
The first step, MODIS surface reflectivity product MOD09A1 is converted to polar region azimuthal projection by sinusoidal projection.
The original MOD09A1 data downloaded are HDF form, and its filename and product IDs number contain the essential information of data, and example is as follows:
Example: MOD09A1.A2014169.h15v02.005.2014178104315.hdf, contains the data acquisition time (2014), obtains date (Julian date the 169th day), article reel number (h15v02), version number's (the 4th edition), date in production time and time (Julian date the 178th day 10: 43: 15 in 2014).
According to the time provided and geographical location information, choose sample data and in remote sensing image processing software, check its projection and geographic coordinate information.Raw data is Sinusoidal sinusoidal projection, and without level surface spheroid parameter, spatial resolution is 463.3127 meters.
Enter in remote sensing image processing software
row is thrownshadow is changed, optimum configurations is as follows: specify and be newly projected as PolarStereographic polar region azimuthal projection, and level surface selects WGS-84 spheroid, and parasang is rice, correcting method selects Triangulation, uses Nearest nearest neighbor method to be 500 meters by resolution resampling.It is Geotiff that Output rusults preserves form.
Second step, in conjunction with the land mask in MOD09A1 product and cloud mask data, extract comprise λ
1, λ
2, λ
3the image of the example area of three wave bands.
In conjunction with the land mask product in MOD09A1 product, avoid choosing land, polar region; In conjunction with cloud mask data, avoid choosing the many regions of cloud overlay capacity.Choose the region of 500 × 500 Pixel sizes as region of interest, and select first three wave band λ of data
1, λ
2, λ
3, use the image of remote sensing image processing software to cut function, export the example area image extracted in order to sea ice melt pool,
as Fig. 2be depicted as region of interest wave band 1 image.Result comprises λ
1, λ
2, λ
3the surface reflectivity data of three wave bands.
Wherein: λ
1the wavelength coverage of wave band is 620-670nm, λ
2the wavelength coverage of wave band is 841-876nm, λ
3the wavelength coverage of wave band is 459-479nm.
3rd step, determine that ice and snow, water body and melt pool three are respectively at λ
1, λ
2, λ
3the standard spectrum reflectance value of wave band.The reflectance curve of the typical earth surface type in sea ice region, polar region can be checked in standard spectrum storehouse,
as Fig. 3shown in, a is the curve of spectrum of ice and snow, and h is the curve of spectrum of water body, and g is the melt pool curve of spectrum.According to the curve of spectrum of three, select separately respectively at λ
1, λ
2, λ
3the reflectance value that wave band is corresponding.
At λ
1wave band: the spectral reflectivity r of ice and snow
i(λ
1)=0.95, the spectral reflectivity r of water body
w(λ
1)=0.08, the spectral reflectivity r of melt pool
m(λ
1)=0.16;
At λ
2wave band: the spectral reflectivity r of ice and snow
i(λ
2)=0.87, the spectral reflectivity r of water body
w(λ
2)=0.08, the spectral reflectivity r of melt pool
m(λ
2)=0.07;
At λ
3wave band: the spectral reflectivity r of ice and snow
i(λ
3)=0.95, the spectral reflectivity r of water body
w(λ
3)=0.08, the spectral reflectivity r of melt pool
m(λ
3)=0.22.
4th step, to each pixel in image, according to wherein ice and snow, water body and melt pool proportion S
i, S
w, S
mwith the relation of three's spectral reflectivity, set up multivariate linear equations.And increasing system of equations constraint condition: three's proportion is all nonnegative value and summation is 1.
S
i*r
i(λ
1)+S
w*r
w(λ
1)+S
m*r
m(λ
1)＝R(λ
1)
S
i*r
i(λ
2)+S
w*r
w(λ
2)+S
m*r
m(λ
2)＝R(λ
2)
S
i*r
i(λ
3)+S
w*r
w(λ
3)+S
m*r
m(λ
3)＝R(λ
3)
S
i+S
w+S
m＝1
In formula, R (λ
1), R (λ
2), R (λ
3) be that pixel is at λ respectively
1, λ
2, λ
3the reflectance value of wave band, this value obtains in second step;
5th step, from image, choose the pixel of 5%-10% quantity, using the training sample as artificial neural network, for the pixel chosen, use the system of equations of quasi-Newton method (BFGS algorithm, i.e. Broyden-Fletcher-Goldfarb-Shanno algorithm) direct solution Problem with Some Constrained Conditions.
The solution obtained is ice and snow in each pixel, water body and melt pool proportion.Using the training sample data collection of these results as artificial neural network, in order to next step neural network training model.
6th step, structure neural network model.Neural network structure
as Fig. 4shown in (illustrate: be the relation between clear expression hidden layer, employ respectively in 3 and 6 node on behalf ground floor 9 with 27 nodes in the second layer.)
The neural network model of this experiment is multilayer feedforward perceptron, by an input layer, and two hidden layers (ground floor comprises 9 nodes, and the second layer comprises 27 nodes), an output layer composition.Input layer is the λ of MODIS reflectivity data
1, λ
2, λ
3wave band; Output layer is the image of ice and snow, water body and melt pool proportion.
7th step, operation neural network model.It is as follows that the parameter that neural network model uses is set: the RMS error amount of training stop condition is 0.1; The iterations of training is 10000 times.
The pixel in all experimental example regions is solved to the system of equations set up in the 5th step.Acquisition result is ice and snow, water body and the melt pool ratio S shared by respectively in pixel
i, S
w, S
m.
Melt pool in 8th step, experiment with computing example area image covers relative scale.Utilize the 7th step acquired results, pass through formula S
m* (1-S
w) complete corrected Calculation.
Melt pool is covered between relative scale location and is divided into 5 grades: 0-0.2,0.2-0.4,0.4-0.6,0.6-0.8,0.8-1.0, and compose with the gray-scale value of different brackets, form melt pool scope distributed image,
as Fig. 5shown in.
In addition to the implementation, the present invention can also have other embodiments.All employings are equal to the technical scheme of replacement or equivalent transformation formation, all drop on the protection domain of application claims.
Claims (7)
1., based on a polar region sea ice melt pool extracting method for neural network model, its step comprises:
The first step, MODIS surface reflectivity product MOD09A1 is converted to polar region azimuthal projection by sinusoidal projection;
Second step, in conjunction with the land mask in MOD09A1 product and cloud mask data, extract the image of target area, and choose λ
1, λ
2, λ
3the surface reflectivity data of wave band;
3rd step, spectral response curve according to ice and snow, water body and melt pool, obtain three respectively at λ
1, λ
2, λ
3the standard spectrum reflectance value of wave band:
At λ
1wave band: the spectral reflectivity r of ice and snow
i(λ
1)=0.95, the spectral reflectivity r of water body
w(λ
1)=0.08, the spectral reflectivity r of melt pool
m(λ
1)=0.16;
At λ
2wave band: the spectral reflectivity r of ice and snow
i(λ
2)=0.87, the spectral reflectivity r of water body
w(λ
2)=0.08, the spectral reflectivity r of melt pool
m(λ
2)=0.07;
At λ
3wave band: the spectral reflectivity r of ice and snow
i(λ
3)=0.95, the spectral reflectivity r of water body
w(λ
3)=0.08, the spectral reflectivity r of melt pool
m(λ
3)=0.22;
4th step, to each pixel in image, according to wherein ice and snow, water body and melt pool proportion S
i, S
w, S
,with the relation of three's spectral reflectivity, set up multivariate linear equations;
S
i*r
i(λ
1)+S
w*r
w(λ
1)+S
m*r
m(λ
1)＝R(λ
1)
S
i*r
i(λ
2)+S
w*r
w(λ
2)+S
m*r
m(λ
2)＝R(λ
2)
S
i*r
i(λ
3)+S
w*r
w(λ
3)+S
m*r
m(λ
3)＝R(λ
3)
S
i+S
w+S
m＝1
In formula, R (λ
1), R (λ
2), R (λ
3) be that pixel is at λ respectively
1, λ
2, λ
3the reflectance value of wave band;
5th step, from image, choose the pixel of 5%-10% quantity, using the training sample as artificial neural network, utilize the system of equations of the pixel direct solution Problem with Some Constrained Conditions chosen, the solution obtained is ice and snow in each pixel, water body and melt pool proportion, as the training sample set of artificial neural network;
6th step, structure neural network model, input layer is the λ of MODIS reflectivity data
1, λ
2, λ
3wave band; Output layer is the image of ice and snow, water body and melt pool proportion;
7th step, operation neural network model, rapid Optimum solving equation group, acquisition result is ice and snow, water body and the melt pool ratio S shared by respectively in pixel
i, S
w, S
m;
8th step, pass through formula S
m* (1-S
w), the melt pool calculated in target area image covers relative scale, covers relative scale draw acquisition target area melt pool scope distribution plan according to melt pool.
2. the polar region sea ice melt pool extracting method based on neural network model according to claim 1, it is characterized in that: before the pre-service of described first step remote sensing image, in remote sensing image processing software, first determine projection information and the geographical location information of MODIS surface reflectivity product MOD09A1 image, the sinusoidal projection of former image is converted to polar region azimuthal projection.
3. according to claim 1, it is characterized in that: λ in described second step
1the wavelength coverage of wave band is 620-670nm, λ
2the wavelength coverage of wave band is 841-876nm, λ
3the wavelength coverage of wave band is 459-479nm.
4. according to claim 1, it is characterized in that: in the 5th step, quasi-Newton method solving equation group is used for a small amount of pixel, obtain the ratio shared by ice and snow, water body and melt pool difference in each pixel, as training sample.
5. according to claim 1, it is characterized in that: in the 6th step, neural network model is multilayer feedforward perceptron, be made up of an input layer, two-layer hidden layer and an output layer, ground floor hidden layer comprises 9 nodes, and second layer hidden layer comprises 27 nodes.
6. according to claim 1, it is characterized in that: running the optimum configurations that neural network model uses in the 7th step is: the RMS error amount of training stop condition is 0.1; The iterations of training is 10000 times.
7. according to claim 1, it is characterized in that: when the 8th step calculates melt pool covering relative scale, melt pool proportion in pixel is converted into the value of relatively non-coverage of water, when drawing melt pool scope distribution plan, be divided into some grades according between melt pool coating ratio value location, and compose with the gray scale of different brackets formation image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510493381.6A CN105116464A (en) | 2015-08-12 | 2015-08-12 | Polar sea ice melting pool extraction method based on neural network model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510493381.6A CN105116464A (en) | 2015-08-12 | 2015-08-12 | Polar sea ice melting pool extraction method based on neural network model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105116464A true CN105116464A (en) | 2015-12-02 |
Family
ID=54664494
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510493381.6A Pending CN105116464A (en) | 2015-08-12 | 2015-08-12 | Polar sea ice melting pool extraction method based on neural network model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105116464A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106290183A (en) * | 2016-07-19 | 2017-01-04 | 国网天津市电力公司 | snow state parameter measuring method based on spectral reflectance |
| CN107679476A (en) * | 2017-09-26 | 2018-02-09 | 南京大学 | A kind of Sea Ice Types Classification in Remote Sensing Image method |
| CN108519058A (en) * | 2018-03-21 | 2018-09-11 | 国家海洋环境监测中心 | A kind of method that remote sensing image estimates annual sea ice thickness |
| CN110866363A (en) * | 2019-11-12 | 2020-03-06 | 南京信息工程大学 | Method for inverting arctic fusion pool distribution by using artificial neural network |
| CN111476197A (en) * | 2020-04-24 | 2020-07-31 | 中科天盛卫星技术服务有限公司 | Oil palm identification and area extraction method and system based on multi-source satellite remote sensing image |
| CN113140000A (en) * | 2021-03-26 | 2021-07-20 | 中国科学院东北地理与农业生态研究所 | Water body information estimation method based on satellite spectrum |
| CN113723228A (en) * | 2021-08-16 | 2021-11-30 | 北京大学 | Method and device for determining earth surface type ratio, electronic equipment and storage medium |
| CN113723228B (en) * | 2021-08-16 | 2024-04-26 | 北京大学 | Method and device for determining earth surface type duty ratio, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100740347B1 (en) * | 2006-05-25 | 2007-07-18 | 재단법인서울대학교산학협력재단 | Method for retrieving cloud thermodynamic phase using 6.7 micro channel |
| CN101634711A (en) * | 2009-08-24 | 2010-01-27 | 中国农业科学院农业资源与农业区划研究所 | Method for estimating temperature of near-surface air from MODIS data |
| CN101655564A (en) * | 2009-09-15 | 2010-02-24 | 中国农业科学院农业资源与农业区划研究所 | Method for inversing surface temperature and emissivity from MODIS data |
| CN101936877A (en) * | 2010-03-05 | 2011-01-05 | 中国农业科学院农业资源与农业区划研究所 | Method for inverting atmospheric water vapor content from MODIS (Moderate Resolution Imaging Spectroradiometer) data |
| CN104007443A (en) * | 2014-04-24 | 2014-08-27 | 南京大学 | Method for extracting polar region sea ice melting tank by utilizing spectral feature |
-
2015
- 2015-08-12 CN CN201510493381.6A patent/CN105116464A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100740347B1 (en) * | 2006-05-25 | 2007-07-18 | 재단법인서울대학교산학협력재단 | Method for retrieving cloud thermodynamic phase using 6.7 micro channel |
| CN101634711A (en) * | 2009-08-24 | 2010-01-27 | 中国农业科学院农业资源与农业区划研究所 | Method for estimating temperature of near-surface air from MODIS data |
| CN101655564A (en) * | 2009-09-15 | 2010-02-24 | 中国农业科学院农业资源与农业区划研究所 | Method for inversing surface temperature and emissivity from MODIS data |
| CN101936877A (en) * | 2010-03-05 | 2011-01-05 | 中国农业科学院农业资源与农业区划研究所 | Method for inverting atmospheric water vapor content from MODIS (Moderate Resolution Imaging Spectroradiometer) data |
| CN104007443A (en) * | 2014-04-24 | 2014-08-27 | 南京大学 | Method for extracting polar region sea ice melting tank by utilizing spectral feature |
Non-Patent Citations (2)
| Title |
|---|
| 张干 等: "利用MODIS数据反演融池覆盖率及与航拍影像的比对", 《第28届中国气象学会年会——S6冰冻圈与极地气象学》 * |
| 张良培 编: "《高光谱遥感》", 30 June 2011 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106290183A (en) * | 2016-07-19 | 2017-01-04 | 国网天津市电力公司 | snow state parameter measuring method based on spectral reflectance |
| CN106290183B (en) * | 2016-07-19 | 2018-12-04 | 国网天津市电力公司 | Snow state parameter measurement method based on spectral reflectance |
| CN107679476A (en) * | 2017-09-26 | 2018-02-09 | 南京大学 | A kind of Sea Ice Types Classification in Remote Sensing Image method |
| CN107679476B (en) * | 2017-09-26 | 2020-10-09 | 南京大学 | Sea ice type remote sensing classification method |
| CN108519058A (en) * | 2018-03-21 | 2018-09-11 | 国家海洋环境监测中心 | A kind of method that remote sensing image estimates annual sea ice thickness |
| CN110866363A (en) * | 2019-11-12 | 2020-03-06 | 南京信息工程大学 | Method for inverting arctic fusion pool distribution by using artificial neural network |
| CN110866363B (en) * | 2019-11-12 | 2023-04-07 | 南京信息工程大学 | Method for inverting arctic fusion pool distribution by using artificial neural network |
| CN111476197A (en) * | 2020-04-24 | 2020-07-31 | 中科天盛卫星技术服务有限公司 | Oil palm identification and area extraction method and system based on multi-source satellite remote sensing image |
| CN113140000A (en) * | 2021-03-26 | 2021-07-20 | 中国科学院东北地理与农业生态研究所 | Water body information estimation method based on satellite spectrum |
| CN113723228A (en) * | 2021-08-16 | 2021-11-30 | 北京大学 | Method and device for determining earth surface type ratio, electronic equipment and storage medium |
| CN113723228B (en) * | 2021-08-16 | 2024-04-26 | 北京大学 | Method and device for determining earth surface type duty ratio, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Raynolds et al. | A raster version of the Circumpolar Arctic Vegetation Map (CAVM) | |
| Chirici et al. | Wall-to-wall spatial prediction of growing stock volume based on Italian National Forest Inventory plots and remotely sensed data | |
| Puliti et al. | Modelling above-ground biomass stock over Norway using national forest inventory data with ArcticDEM and Sentinel-2 data | |
| Szabó et al. | Automated registration of potential locations for solar energy production with Light Detection And Ranging (LiDAR) and small format photogrammetry | |
| CN105116464A (en) | Polar sea ice melting pool extraction method based on neural network model | |
| Bode et al. | Subcanopy Solar Radiation model: Predicting solar radiation across a heavily vegetated landscape using LiDAR and GIS solar radiation models | |
| Berterretche et al. | Comparison of regression and geostatistical methods for mapping Leaf Area Index (LAI) with Landsat ETM+ data over a boreal forest | |
| Feliciano et al. | Assessing mangrove above-ground biomass and structure using terrestrial laser scanning: A case study in the Everglades National Park | |
| US20210209803A1 (en) | Computer-based method and system for geo-spatial analysis | |
| CN103942420A (en) | Rapid estimation method for solar energy in construction size | |
| Pinton et al. | A new algorithm for estimating ground elevation and vegetation characteristics in coastal salt marshes from high‐resolution UAV‐based LiDAR point clouds | |
| Fleming et al. | Comparison of methods toward multi-scale forest carbon mapping and spatial uncertainty analysis: Combining national forest inventory plot data and Landsat TM images | |
| Nagarajan et al. | UAS based 3D shoreline change detection of Jupiter Inlet Lighthouse ONA after Hurricane Irma | |
| CN115077656B (en) | Reservoir water reserve retrieval method and device | |
| Boori et al. | Remote sensing and land use/land cover trajectories | |
| Deb et al. | Aboveground biomass estimation of an agro-pastoral ecology in semi-arid Bundelkhand region of India from Landsat data: A comparison of support vector machine and traditional regression models | |
| Novero et al. | The use of light detection and ranging (LiDAR) technology and GIS in the assessment and mapping of bioresources in Davao Region, Mindanao Island, Philippines | |
| Kaya et al. | Determination of long-term volume change in lakes by integration of UAV and satellite data: The case of Lake Burdur in Türkiye | |
| Gülçin et al. | Mapping Urban Green Spaces Based on an Object-Oriented Approach | |
| Samanta et al. | Preparation of digital data sets on land use/land cover, soil and digital elevation model for temperature modelling using Remote Sensing and GIS techniques | |
| Boori et al. | Assessing land cover change trajectories in Olomouc, Czech Republic | |
| Pouyan et al. | Spatial and seasonal modeling of the land surface temperature using random forest | |
| Morakinyo et al. | Mapping of land cover and estimation of their emissivity values for gas flaring sites in the Niger Delta | |
| Araújo et al. | Morphodynamic study of sandy beaches in a tropical tidal inlet using RPAS | |
| Zhao et al. | Performance comparison among typical open global DEM datasets in the Fenhe River Basin of China |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20151202 |