CN100365435C - Analogue technology for imaging spectrograph remote-sensing image in satellite - Google Patents
Analogue technology for imaging spectrograph remote-sensing image in satellite Download PDFInfo
- Publication number
- CN100365435C CN100365435C CNB2005101277366A CN200510127736A CN100365435C CN 100365435 C CN100365435 C CN 100365435C CN B2005101277366 A CNB2005101277366 A CN B2005101277366A CN 200510127736 A CN200510127736 A CN 200510127736A CN 100365435 C CN100365435 C CN 100365435C
- Authority
- CN
- China
- Prior art keywords
- image
- spectral
- spectrum
- remote sensing
- remote
- 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
Landscapes
- Image Processing (AREA)
Abstract
The present invention relates to an analogue technology for the imaging spectrograph remote-sensing image in a satellite, which belongs to an innovation invention in the remote-sensing information scientific subject. The present invention discloses a method which can make use of low spectral resolution remote-sensing images to generate the high spectral resolution remote-sensing images in an analogue way. On the basis of the image texture information and the addressed information which are provided by the low spectral resolution remote-sensing data, the present invention obtains analogue remote-sensing images in the range of each spectral band of the high spectral resolution remote-sensing images and simultaneously avoids the condition that amount of the calculation is difficult to be accepted by the analogue calculation of the radioactive transfer in the process of the remote-sensing imaging operation according to the prior knowledge support, such as spectra databases, etc. A user can make use of the invention technology to see the analogue image before the satellite emission or the on-board examination, which is useful for the sensing understanding of a decision maker so as to select the proper operating mode and parameters for a remote-sensing device. Simultaneously, the present invention is also good for researching the mechanism of the electromagnetic wave transmission in the process of capturing the remote-sensing images.
Description
Technical Field
The invention relates to a practical technology in the field of earth observation, in particular to a novel optical remote sensing image spectrum subdivision method. The invention can provide the needed simulation image before satellite emission or airborne test, and guide the selection of the working mode and parameters of the remote sensing sensor; the perceptual knowledge of a decision maker before sensor development and satellite launching is increased; helping to compensate for the lack in the remote sensing database. In summary, the present invention is a practical information support technology.
Background
Remote sensing has global observation capability, and enables scientists to obtain earth observation images with multi-space resolution and multi-spectral resolution, thereby better developing the research work of earth science. To meet the needs of different studies, it is often desirable to be able to obtain remote sensing images reflecting different spatial, spectral and temporal resolutions under the influence of a variety of natural conditions. However, due to the limitation of atmospheric influence and observation conditions of the sensor, sufficient remote sensing images cannot be acquired to meet the requirements of scientific research, so that the problem is widely regarded by telesensists, and the image simulation technology proposed for the problem is rapidly developed.
The remote sensing image simulation technology is a remote sensing technology which adds the influence of specific limiting factors on the basis of a remote sensing theoretical model, remote sensing prior knowledge and the existing remote sensing image and obtains a simulated image under specific conditions through mathematical physical calculation. The remote sensing image simulation process comprises three parts, namely, determining a simulated data source, establishing a simulation conversion equation and generating a simulated image. The data source provides various parameters (such as vegetation coverage type, soil condition and the like) of the surface condition of the simulated area for the image to be simulated, and is used for comprehensively describing the terrain and the natural environment of the area to be simulated. The acquired remote sensing image can provide certain natural information of a simulation area and can be used as a simulation data source; when the remote sensing image of the simulation area is not easy to obtain, the natural condition of the simulation area can be described by using the assumed comprehensive parameters such as a vegetation model, a geographical model, an atmospheric model and the like, and the required natural information can be obtained. However, the data provided by various data sources are approximate descriptions of the natural earth surface, and have inevitable differences from the actual natural conditions, so how to select a proper data source has an important influence on the simulation accuracy. The conversion equation is the medium between the data source and the analog image, and establishes the analog relation between the data source parameter and the data received by the remote sensing sensor based on the remote sensing and mathematical knowledge. When the conversion equation is established, various factors influencing the generation of the simulated image are considered as comprehensively as possible, and the real condition of the sensor during receiving is simulated as truly as possible. The influence factors are mainly atmospheric radiation influence and sensor receiving condition influence, namely, the atmospheric loss and absorption of radiation energy, the influence of a sensor imaging system, the influence of observation angle observation height and the like. The image type of the analog image to be generated can be an optical image or an image of an SAR or other imaging microwave radiometer, and because the difference of different image imaging mechanisms is large, the difference of different types of analog images on the analog technology is also obvious.
Disclosure of Invention
The invention provides a new optical remote sensing image simulation technology, namely a new remote sensing image spectrum subdivision method. Different from the conventional technical method for forming the low-spectrum remote sensing image by utilizing the high-spectrum remote sensing data, the method obtains the simulated remote sensing image of each spectrum wave band of the high-spectrum resolution image on the basis of the image texture information and the land information provided by the low-spectrum resolution remote sensing data through the simulated operation of radiation transmission during remote sensing imaging according to the support of prior knowledge such as a spectrum database and the like, and simultaneously avoids unacceptable calculated amount.
The simulation method provided by the invention comprises the following steps: collecting simulation data; determining a low spectral resolution remote sensing image as a template; processing data of a spectrum knowledge base; obtaining a low-spectral-resolution remote sensing image classification map; obtaining the multiband reflectivity of each pixel point; and constructing a conversion model, and acquiring a reflectivity image of each spectral band of the high-spectral-resolution image.
The technical method provided by the invention can enhance the research on the electromagnetic wave transmission mechanism in the process of obtaining the remote sensing image and establish a forward model which is beneficial to remote sensing inversion; make up the deficiency in the remote sensing database, meet the needs of the remote sensing users; support is provided for the selection of the working mode and the parameters of the remote sensor; the simulation image can be seen before satellite launching or airborne experiment, the perceptual knowledge of a sensor development decision maker is increased, and the decision maker is helped to make a satellite launching plan.
Drawings
FIG. 1 is a flow chart of spectral matching classification according to the present invention;
fig. 2 is a technical route flow diagram of the present invention.
Detailed Description
As shown in fig. 2, the present invention provides a remote sensing image spectrum subdivision method, including: collecting simulation data; determining a low spectral resolution remote sensing image as a template; processing data of a spectrum knowledge base; obtaining a low-spectral-resolution remote sensing image classification map; acquiring multiband reflectivity of each pixel point; and constructing a conversion model, and acquiring a reflectivity image of each spectral band of the high-spectral-resolution image.
The complete technical process is as follows:
● And (4) simulating the collection of data.
The method comprises the following three parts: the spectrum knowledge base information of the reflectivity information of each spectrum wave band range of the high-spectrum resolution image, the data information of each spectrum wave band of the low-spectrum resolution image obtained by the existing remote sensing image and the range of each spectrum wave band of the high-spectrum resolution image obtained by satellite-borne spectrometer information can be provided in detail.
● And determining the low spectral resolution remote sensing image as the template.
The remote sensing images are divided into two types of optical images and microwave images according to an imaging mechanism, and the spectral resolution and the spatial resolution of the images of the same type are also different. Because the simulation process is to finely divide the low-spectral-resolution image on the spectral resolution, when the template image is selected, the remote sensing image which is closest to the simulation image on the image type and the spatial resolution needs to be selected, so that the error caused by the spatial resolution conversion is reduced, the precision of the simulation image is improved, and the credibility of the simulation image is increased.
● And (4) processing data of a spectrum knowledge base.
The spectrum knowledge base provides massive spectral reflectivity data of the ground features, and provides favorable help for spectral analysis of the ground features. When the data in the spectrum library and the remote sensing data are combined and applied, the data in the spectrum library are fitted to the waveband range of the remote sensing image by using the sensor waveband response function, so that the data in the spectrum library and the remote sensing image can be relatively applied on a uniform scale.
● And obtaining a low-spectral-resolution remote sensing image classification map.
By carrying out spectrum similarity analysis on each pixel point of the remote sensing image, the type of the ground object represented by each pixel point is identified by using a spectrum matching method. The accuracy of the ground object type discrimination has a decisive influence on the accuracy of the simulation.
● And obtaining the multiband reflectivity of each pixel point.
Only the reflectivity value of a specific spectral band can be obtained from the original remote sensing image, and the type of the ground object represented by each pixel can be definitely determined through the classification chart of the template image. The surface feature type is used as an attribute parameter, and mass waveband reflectivity values of the type are searched in a spectrum knowledge base, so that spectral resolution information is expanded.
● And constructing a conversion model, and acquiring a reflectivity image of each spectral band of the high-spectral-resolution image.
And constructing a conversion model by taking the reflectivity of each pixel point of the existing remote sensing image and the reflectivity value of each waveband range of the high-spectral resolution image searched from the spectrum knowledge base as input parameters to obtain a simulated image of each spectral waveband of the high-spectral resolution image. When the conversion model is established, objective differences (including atmospheric radiation influence, terrain influence and the like) between the measured data and remote sensing inversion data need to be considered, and images of spectral wave band ranges of the high-spectral-resolution images received by the sensor are simulated through simulation calculation of the model.
The spectral matching classification method of the present invention is described below.
Spectral reflectivity curves of different types of ground objects are different, and ground object identification and judgment can be completed through similarity analysis of the curves. The spectrum matching method is used for identifying the type of the ground object by comparing the shape of the spectrum reflectivity curve of the ground object to be detected with the shape of the standard spectrum reflectivity curve and finding out the most similar reflectivity curve, and the specific process is shown in figure 1. Two picture element vectors r = (r) 1 ,r 2 ,...,r L ) T And r '(r' 1 ,r′ 2 ,...,r′ L ) T Each of r j 、r′ j The dimensional components of different bands of each pixel are represented, s =(s) 1 ,s 2 ,...,s L ) T And s '= (s' 1 ,s′ 2 ,...,s′ L ) T The spectral values corresponding to each vector.
From knowledge of the information theory, the self-information of a particular band can be defined with the probability distribution function above:
I j (r)=-logp j (1)
I j (r′)=-logq j (2)
according to equations (1) and (2), the self-information difference of two image elements can be defined as:
D j (r‖r′)=I j (r)-I j (r′)=-log(p j /q j ) (3)
mean correlation entropy D of all bands j (r | < r') is:
spectral Information Divergence (SID):
SID(r,r′)=-(D(r‖r′)+D(r′‖r)) (5)
by finding out the value with the minimum spectrum information divergence, the feature type of the type can be automatically distinguished.
The transformation model construction method of the present invention is described below.
Suppose L is the signal energy (apparent reflectivity) in the low spectral resolution image band range, beta (lambda) is the weight value of the spectral response function corresponding to different bands, L' (lambda) is the spectral value of the ground object stored in the spectrum knowledge base, K is the low spectral resolution image band range, L i (K) The signal energy of the i-th type ground object in the low spectral resolution image K wave band range. Considering the influence of atmospheric radiation, terrain change and other factors, the data stored in the spectrum knowledge base and the data calculated by the remote sensing image have difference, so the parameter mu (lambda) is introduced to correct the difference between the two,when a certain pixel is classified as i ground feature, an analog conversion model is constructed as follows:
wherein N is the number of spectral band ranges including a high spectral resolution image within the spectral band range of the low spectral resolution image, and λ is the wavelength,(lambda) is used for reflecting the difference of the spectral characteristics of the same type of ground objects in the wave spectrum library, and M reflects the number of pixels of the ground objects identified as i type in the low spectral resolution image;reflecting the detail difference among the same type of pixels, wherein alpha is a difference adjustment parameter used for reflecting the detail difference among the same type of pixels caused by non-atmospheric influence. By the above formula, we establish the connection between the low-spectral resolution image signal information and the data in the spectrum knowledge base. Solving for the parameter μ (λ) using equation (6)And when alpha is calculated, the fact that solutions with practical significance cannot be solved by mathematical methods such as least square method with more condition variables is found, and mu (lambda) cannot be guaranteed to be positive numerical solutions. Considering that the wavelength ranges of the bands of the high spectral resolution images are very close, it is assumed that the sub-bands have the same μ (λ), and the μ (λ) is recordedAt this time, the least square method can be used to obtainAnd the corresponding alpha.Is the result of the comprehensive action of each subdivided wave band, and reflects the comprehensive action effect. While in fact the correction coefficients for each sub-bandμ (. Lamda.) is different, and therefore, it is also necessary toIs processed to obtainAnd μ (λ).
The difference of different sub-divided bands can be expressed from the difference of reflectivity values of different sub-divided bands, so that the following steps are introduced:
here, phi (lambda) is used to reflect the degree of deviation of different subdivided spectral bands, and can reflect the difference of each subdivided band, so thatRelation with mu (lambda)
The analog value of the analog subdivided spectral band range of a single pixel point can be obtained through the formula, and Sim (lambda) is the analog value of different bands obtained through an analog conversion model:
the analog values of the high spectral resolution image of each wave band and each spectral wave band of the high spectral resolution image of each pixel point can be obtained through the formula (10), the analog values of all pixel points of a certain wave band are combined to form a new two-dimensional matrix, and the matrix is the matrix for simulating the high spectral resolution image and the band spectral analog image, so that the work of simulating and obtaining the high spectral resolution image by the low spectral resolution image is completed.
The above description details the remote sensing image spectral subdivision method and simulation implementation techniques, but those skilled in the art will appreciate that various modifications, additions and substitutions are possible, within the scope and spirit of the invention, as defined in the accompanying claims.
Claims (7)
1. A remote sensing image spectrum subdivision method comprises the following steps:
collecting simulation data;
determining a low spectral resolution remote sensing image as a template;
processing data of a spectrum knowledge base;
obtaining a low-spectral-resolution remote sensing image classification map;
obtaining the multiband reflectivity of each pixel point;
and constructing a conversion model, and acquiring a reflectivity image of each spectral band of the high-spectral-resolution image.
2. The method of claim 1, wherein the collection of simulation material comprises three parts: providing detailed spectrum knowledge base data of reflectivity information of each spectrum wave band, acquiring data information of each spectrum wave band of a low-spectrum-resolution image by using the existing remote sensing image, and acquiring the range of each spectrum wave band of a high-spectrum-resolution image by using the parameters of the image to be simulated.
3. The method according to claim 1, wherein when determining the low spectral resolution remote sensing image as the template, the spectral resolution and the spatial resolution of the same type of image are different because the remote sensing image is divided into two types of optical image and microwave image according to the imaging mechanism.
4. The method of claim 1, wherein the data processing of the spectral knowledge base comprises: and fitting the data in the spectrum library to the wave band range of the remote sensing image by using the sensor wave band response function.
5. The method according to claim 1, wherein when the classification information of the surface features is obtained, the classification chart of the low-spectral-resolution image is obtained by performing spectrum similarity analysis on each pixel point of the remote-sensing image and using the support of a surface feature spectrum knowledge base by using a spectrum matching method to quickly and automatically identify the surface feature type represented by each pixel point.
6. The method of claim 1, wherein when obtaining the multiband reflectivity of each pixel point, the type of the surface feature is used as an attribute parameter, and the massive waveband reflectivity value of the type is found in the spectrum knowledge base.
7. The method as claimed in claim 1, wherein the transformation model is constructed by using the reflectance of each pixel of the remote sensing image and the reflectance of each band of the high-spectral resolution image found from the spectrum knowledge base as input parametersThe model is used for obtaining a simulation image of each spectral band of the high-spectral-resolution image; when a conversion model is established, objective difference between measured data and remote sensing inversion data needs to be considered, and images in each spectral band range of the high spectral resolution images received by the sensor are simulated through simulation calculation of the model; suppose L is the signal energy in the wave band range of the low spectral resolution image, λ is the wavelength, β (λ) is the weight value representing the spectral response function corresponding to different wave bands, L' (λ) is the spectral value of the ground object stored in the spectrum knowledge base, K is the wave band range of the low spectral resolution image, L ` i (K) The signal energy of the i-th ground object in the K wave band range of the low spectral resolution image, N is the spectral wave of the high spectral resolution image in the spectral wave band range of the low spectral resolution imageThe number of segment ranges, M, reflects the number of pixels identified as i-type ground objects in the low spectral resolution image; considering the influence of atmospheric radiation and terrain variation factors, the data stored in the spectrum knowledge base is different from the data calculated by the remote sensing image, so the parameters are introducedThe difference between the two is corrected,the method is used for reflecting the difference of the spectral characteristics of the same type of ground objects in a spectrum library, and phi (lambda) is used for reflecting the deviation degree of each spectral band of the high-spectral resolution image and can reflect the difference of each band; alpha is a difference adjustment parameter used for reflecting detail difference among similar pixels caused by non-atmospheric influence, and Sim (lambda) is a simulation value of different wave bands obtained through a simulation conversion model;
and finally constructing a conversion model through the formulas (1) and (2), wherein the model is described as a formula (3).
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNB2005101277366A CN100365435C (en) | 2005-12-05 | 2005-12-05 | Analogue technology for imaging spectrograph remote-sensing image in satellite |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNB2005101277366A CN100365435C (en) | 2005-12-05 | 2005-12-05 | Analogue technology for imaging spectrograph remote-sensing image in satellite |
Publications (2)
Publication Number | Publication Date |
---|---|
CN1776453A CN1776453A (en) | 2006-05-24 |
CN100365435C true CN100365435C (en) | 2008-01-30 |
Family
ID=36766073
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CNB2005101277366A Expired - Fee Related CN100365435C (en) | 2005-12-05 | 2005-12-05 | Analogue technology for imaging spectrograph remote-sensing image in satellite |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN100365435C (en) |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR100963797B1 (en) * | 2008-02-27 | 2010-06-17 | 아주대학교산학협력단 | Method for realtime target detection based on reduced complexity hyperspectral processing |
CN101320087B (en) * | 2008-07-23 | 2011-05-11 | 北京大学 | High optical spectrum reconstruction method and system based on TM image |
CN101598798B (en) * | 2008-12-31 | 2012-01-04 | 中国资源卫星应用中心 | System for rebuilding spectrum of high spectrum intervention data and method thereof |
CN101515038B (en) * | 2009-03-12 | 2011-09-14 | 北京航空航天大学 | Analogy method for remote sensing radiance data cube in flat terrain |
CN101799871B (en) * | 2010-01-18 | 2012-02-29 | 浙江林学院 | Regression parameter transformation based method for extracting thematic information of remote sensing images |
CN102507474B (en) * | 2011-10-28 | 2013-07-24 | 大连海事大学 | Method and system for identifying oil spilling target of ship |
US10656305B2 (en) * | 2016-07-01 | 2020-05-19 | The Boeing Company | Method and apparatus for simulating spectral information of geographic areas |
CN106652016B (en) * | 2016-09-08 | 2019-08-09 | 北京空间机电研究所 | A kind of remote sensing image emulation mode based on radiation geometry integrated design |
CN110909186B (en) * | 2018-09-14 | 2023-08-22 | 中国科学院上海高等研究院 | Hyperspectral remote sensing data storage and retrieval method and system, storage medium and terminal |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1472544A (en) * | 2003-06-05 | 2004-02-04 | 上海交通大学 | Remote sensing image picture element and characteristic combination optimizing mixing method |
CN1489111A (en) * | 2003-08-21 | 2004-04-14 | 上海交通大学 | Remote-sensing image mixing method based on local statistical property and colour space transformation |
WO2004044527A1 (en) * | 2002-11-14 | 2004-05-27 | Suomen Ympäristökeskus | Remote sensing method and system |
CN1544958A (en) * | 2003-11-20 | 2004-11-10 | 中国科学院上海技术物理研究所 | Airborne broom pushing type wide viewing field high spectrum remote sensing imaging system |
CN1581230A (en) * | 2004-05-20 | 2005-02-16 | 上海交通大学 | Remote-senstive image interfusion method based on image local spectrum characteristic |
-
2005
- 2005-12-05 CN CNB2005101277366A patent/CN100365435C/en not_active Expired - Fee Related
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2004044527A1 (en) * | 2002-11-14 | 2004-05-27 | Suomen Ympäristökeskus | Remote sensing method and system |
CN1472544A (en) * | 2003-06-05 | 2004-02-04 | 上海交通大学 | Remote sensing image picture element and characteristic combination optimizing mixing method |
CN1489111A (en) * | 2003-08-21 | 2004-04-14 | 上海交通大学 | Remote-sensing image mixing method based on local statistical property and colour space transformation |
CN1544958A (en) * | 2003-11-20 | 2004-11-10 | 中国科学院上海技术物理研究所 | Airborne broom pushing type wide viewing field high spectrum remote sensing imaging system |
CN1581230A (en) * | 2004-05-20 | 2005-02-16 | 上海交通大学 | Remote-senstive image interfusion method based on image local spectrum characteristic |
Non-Patent Citations (5)
Title |
---|
一种基于光谱知识库的TM影像地物识别方法. 陈方,牛铮,骆成凤,王长耀.遥感技术与应用,第20卷第8期. 2005 * |
典型地物波谱库的数据体系与波谱模拟. 苏理宏,李小文,梁顺林,王锦地.地球信息科学,第4期. 2002 * |
分布式遥感模型库的构建及其运行机制 唐世浩,苏理宏,帅艳民,王?,龙科峰. 唐世浩,苏理宏,帅艳民,王森,龙科峰.地球信息科学,第6卷第1期. 2004 * |
地物光谱数据库及其在遥感中的应用. 易维宁,陆亦怀,罗明.光电子技术与信息,第11卷第5期. 1998 * |
波谱模拟模型在遥感中的应用. 胡妮,刘思含,胡争光.地球科学与环境学报,第27卷第3期. 2005 * |
Also Published As
Publication number | Publication date |
---|---|
CN1776453A (en) | 2006-05-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN100365435C (en) | Analogue technology for imaging spectrograph remote-sensing image in satellite | |
Camps-Valls et al. | A survey on Gaussian processes for earth-observation data analysis: A comprehensive investigation | |
CN113128134B (en) | Mining area ecological environment evolution driving factor weight quantitative analysis method | |
Deng et al. | Algorithm for global leaf area index retrieval using satellite imagery | |
Zhang et al. | Super-resolution enhancement of Sentinel-2 image for retrieving LAI and chlorophyll content of summer corn | |
CN110046415A (en) | A kind of soil organic matter content remote sensing dynamic playback method of space-time fining | |
Wan et al. | A model for phenotyping crop fractional vegetation cover using imagery from unmanned aerial vehicles | |
Weyermann et al. | Correction of reflectance anisotropy effects of vegetation on airborne spectroscopy data and derived products | |
Banskota et al. | An LUT-based inversion of DART model to estimate forest LAI from hyperspectral data | |
CN114740180A (en) | Soil organic carbon estimation method and device based on multi-source remote sensing data | |
Asner et al. | Unmixing the directional reflectances of AVHRR sub-pixel landcovers | |
CN107688003B (en) | Blade reflectivity satellite remote sensing extraction method for eliminating vegetation canopy structure and earth surface background influence | |
Poorazimy et al. | Integrating airborne laser scanning data, space-borne radar data and digital aerial imagery to estimate aboveground carbon stock in Hyrcanian forests, Iran | |
Chen et al. | Linear downscaling from MODIS to Landsat: connecting landscape composition with ecosystem functions | |
Sun et al. | Leaf pigment retrieval using the PROSAIL model: Influence of uncertainty in prior canopy-structure information | |
Samain et al. | Spectral normalization and fusion of optical sensors for the retrieval of BRDF and albedo: Application to VEGETATION, MODIS, and MERIS data sets | |
CN117035066A (en) | Ground surface temperature downscaling method coupling geographic weighting and random forest | |
Wang et al. | Soil moisture retrieval from sentinel-1 and sentinel-2 data using ensemble learning over vegetated fields | |
Perkins et al. | High-speed atmospheric correction for spectral image processing | |
Li et al. | A multispectral remote sensing data spectral unmixing algorithm based on variational Bayesian ICA | |
CN112102218B (en) | Fusion method for generating high-spatial-resolution multispectral image | |
Hakkenberg et al. | Automated continuous fields prediction from Landsat time series: application to fractional impervious cover | |
Milstein et al. | Detail Enhancement of AIRS/AMSU Temperature and Moisture Profiles Using a 3D Deep Neural Network | |
Covey et al. | Intercomparison of climate data sets as a measure of observational uncertainty | |
Reyes-Muñoz et al. | Inferring global terrestrial carbon fluxes from the synergy of Sentinel 3 & 5P with Gaussian process hybrid models |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
C17 | Cessation of patent right | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20080130 Termination date: 20101205 |