摘要:Supported by NSFC,we have carried out the research on adjacency effect since 2004 and conducted 2 satellite and 1 aerial synchronous experiments.The experiments on May 11st and July 25th, 2005 and the results analysis were introduced in this paper.With improved measurement design and measurement device,data are analyzed by comparing the spectral curves from ground measurement,flat platform measurement and framed platform measurement,the results show that the improved measurement can prevent more scatter lights to come in the adjacent measurement system.Two adjacent effect analysis results show that if the two pixels have the same brightness there are mutual adjacent influence source and the influence strength is around 12.5%,if the two pixels do not have the same brightness,the influence strength toward weaker pixel,in other word the less brighter pixels get more adjacent effect.
关键词:Adjacency effect;synchronous experiments;PSF model
摘要:The study established an integrated model combining radiative transfer and geometric optical properties in order to simulate multi-spectral reflectance of semi-arid natural grassland.The model introduced a geometrical similarity parameter,which describes the influences of plant crown shape on leaf/branch angle distribution frequencies,fraction of shadowed or illuminated canopy and background,canopy reflectance,and finally scene reflectance.The simulated results agreed with the measured values in the three Leymus chinense grasslands with different degradation degree,and the lower was degradation degree,the better were simulated results.For the moderately degraded grasslands,if plant crown height-to-width ratio(chw) was invariant,when 45°≤solar zenith angle(sza)≤75°,no significant differences occurred in scene reflectance(R) for different plant crown shapes;and when 0°≤sza<45° or 75 °<sza≤90°,there existed significant differences in R between conic plant crown and spherical and cylindrical plant crown,but no significant differences between spherical and cylindrical plant crown in all cases.If chw was changed,the above conclusions were still agreed.
摘要:Remote sensing images contain a lot of mixed image pixels,but it is difficult to classify these pixels.If the number of pixel’s endmember is regarded as unchangeable,the traditional pixel unmixing algorithm cannot get a good result.In this paper we develop a new method of selective endmembers for pixel unmixing based on the fuzzy ARTMAP neural network,which firstly compares the pixel’s spectral to the conference one and then gets the number of endmember.When it is taken into account,we use an ARTMAP neural network to extract subpixel information.Finally,the experimental results show that the selective endmember algorithm has been improved over conventional ANN algorithms and conventional linear algorithms.
摘要:A road network change detection algorithm based on linear feature is proposed in this paper.Firstly,changed linear features are detected based on edge gradient information from multi-temporal remote sensing images.Secondly,the changed roads that fit well with the constraints of the model of road are extracted according to the local property of changed linear features.Finally,the changed roads are connected to changed road network based on the global connection constraints.This road network change detection approach uses the gradient information of edges as the change detection element instead of matching the line information,so it decreases the complexity of the change detection.The experimental results in multi-temporal remote sensing show the effectiveness of the proposed approach.
摘要:A method called simulated annealing is adopted for the optimization of a two-dimensional circle array composed of 20—30 antennas.Based on the optimized array and principle of synthetic aperture imaging,some problems about millimeter wave imaging algorithm using Gridding treatment and linear interpolation in UV domain are studied.And then the imaging of the optimized array is simulated and analyzed.The influences on imaging quality by number of antenna and the linear interpolation are also discussed.
关键词:synthetic aperture;optimization of array;u-v coverage;visibility function;mm-wave;Imaging
摘要:A new method to extract internal wave parameters from SAR imagery has been developed based on the Hilbert-Huang Transform(HHT).Internal wave parameters have been retrieved from the ERS-1 SAR image in the Luzou strait.HHT is used to decompose the internal wave profile data.The internal wave component is extracted according to the maximum normalized deflection.The results show that the new method can not only extract the wavelength of internal waves,but also extract wave shapes modulated by the internal waves.The mean wavelengths of the internal waves are calculated in order to solve the 180° ambiguity of the internal wave direction.The half width of each soliton is calculated.The internal wave amplitudes retrieved by this new method are about 104m which is in agreement with Liu’s report.
摘要:A new classification algorithm for high spatial resolution remotely sensed imagery is proposed,which integrates neighborhood information of multiscale such as 2×2,4×4,8×8 and 16×16 window sizes around the central pixel.In order to compress the information of the multiscale spatial features,a wavelet coefficients fusion algorithm is employed to reduce the dimension but retain the spatial information at the same time.After the stage of multiscale neighborhood feature extraction,a good tool of pattern recognition: SVM is employed to process the multiscale features,in this algorithm,four groups of spatial features based on four scales produce four classification maps.And then,these maps,which represent multiscale classification results,are fused by a scale selection parameter.The final fusion map is the result of multiscale features classification and shows an obvious adaptability to objects of different scales.Experiments of QuickBird and Ikonos show that the proposed classification algorithm of multiscale features fusion can achieve better results and better accuracies than the conventional per-pixel multispectral method.
摘要:Image fusion on high resolution image is one of the most important contents in remote sensing community.In this paper,image fusion algorithm based on Empirical Mode decomposition(EMD) is put forward for the first time.Firstly,intensity image is obtained by IHS transform on multi-spectral image.Secondly,high frequence component and low frequence component of high resolution image and intensity image are separated with 2D EMD realized by means of row and coloum extension of 1D EMD,which are applied to perform image fusion experiment of high resolution image.At last,fused intensity image is obtained by reconstruction with high frequence of high-resolution image and low frequence of intensity image and IHS inverse transform result in fused image.After presenting EMD principle,multi-scale decomposition and reconstruction algorithm of 2D EMD is defined and fusion technique scheme is advanced based on EMD.Panchromatic band and multi-spectral band3,2,1 of QUICKBIRD are used to assess the quality of the fusion algorithm.After selecting appropriate Intrinsic Mode Function(IMF) for the merger on the basis of EMD analysis on specific row(colum) pixel gray value series,the fusion scheme gives fused image,which is compared with generally used fusion algorithms(Wavelet,IHS,Brovey).The objectives of image fusion include enhancing the visiblity of the image and improving the spatial resolution and the spectral information of the original images.For assessing quality of an image after fusion,information entropy and standard deviation are applied to assess spatial details of the fused images and correlation coefficient,bias index and warping degree for measuring distortion between the original image and fused image in terms of spectral information.For all the proposed fusion algorithm,better results are obtained when EMD algorithm is used to perform the fusion experience.
关键词:image fusion;Experimental Model Decomposition(EMD);quantitatively evaluation
摘要:In the course of remote sensing image processing,in order to meet the application needs,now,many algorithms have been provided for the fusion between high-resolution panchromatic image and low-resolution multi-spectral image.This paper infers Mallat algorithm and a′Trous(porous algorithm) algorithm theoretically,then provides specific calculation procedure.On the basis of them,multi-resolution analysis grey modulation(MRAGM algorithm) is put forward,at the same time,taking the fusion between SPOT image and TM image as an example.At last,three kinds of method are used for image fusion between SPOT high-resolution panchromatic image and TM low-resolution multi-spectral image respectively,and the fused image has been evaluated from qualitative and quantitative aspect.The conclusion is drawn that the new method proposed in this paper is better than other kinds of fusion algorithm.
摘要:In this article,OMIS hyperspctral data was used to study feature selection ability of DT(Decision Tree) algorithm and the impacts of feature selection on DT.The DT was compared to three designed feature selection methods(SEP,MDLM and RELIEF) based on feature selection results and classification accuracy in which three different methods(ML、 BPNN and 1-NN) were applied.Moreover,the impacts of the three designed feature selection methods on DT classification results at different training sample sizes were analyzed.Results indicated that DT was a good feature selection method.After feature selection,DT algorithm outputted to those classification trees that used fewer features(average decrease was 43.36%),had fewer tree nodes(average increase was 18.61%),and had higher classification accuracy(average increase was 0.35%).When the training sample size was small,accuracy improvement was the most significant and meanwhile the tree size scarcely changed.
关键词:decision tree;feature selection;hyperspectral data
摘要:This paper took the western part of Beishan area in Xinjiang as an example to study the methods of extracting alteration information from multispectral remote sensing imagery.After studying advantages and disadvantages of spectral angle mapper(SAM),a method has been developed which improves SAM classification with irradiance of objects.This method has been successfully employed to the extration of alteration information in the test areas.Comparing with spectral angle mapper method,this method is able to remove majority false anomly information.An effective method to remove false anomly information in multispectral remote sensing anomly information extraction has been developed.
关键词:multispectral remote sensing;ultramafic rock;spectral angle mapper;irradiance;remote sensing anomaly information optimization
摘要:In the process of obtaining the snow cover area from MSS image with NDSI algorithm,there is a shortcoming that MSS does not have short wave infrared band which can help to distinguish high reflectance of cumulus.To improve the accuracy of obtaining the snow cover area from MSS,this paper explores a method of snow cover mapping based on NDVI background field.Firstly,the land surface reflectance image is derived through 6S model while making radiometric calibration.Then,according to the difference of spectral characteristic and the difference of NDVI characteristic of land objects,we build a Lookup Table on threshold of snow spectrum in the SPECTRAL extension of ENVI software.Through the three demonstrations of ETM+ and TM,the process of algorithm and the way to build LUT are illuminated in detail.And the accuracy of snow cover mapping of this method is tested by NDSI.The results show that the new algorithm has higher Overall accuracy and Kappa Coefficient compared with generally classified method(such as Maximum Likelihood Method).
摘要:The sensitivity analysis and validation of a SVAT(Soil-vegetation-atmosphere transfer) model(CUPID) are carried out to identify the parameters of CUPID with the largest influence on canopy component average temperatures and temperature profiles.The simulation process and field experimental designs are presented in detail.The results show that the overall uncertainty of the soil temperatures is unstable during growing season,while the uncertainty of canopy temperature stays at about 2%;air temperature and soil water content are the two biggest error-contributors.It is found by evaluation that the trend between simulated and measured temperatures is consistent.
关键词:sensitivity analysis;uncertainty analysis;temperature profiles;SVAT;component temperature simulation
摘要:Since 1980s of the 20th Century,outbreak of oriental migratory locust has rampantly emerged again in some regions of China.It is particularly important to monitor timely and accurately the intensity of damage from oriental migratory locust(Locusta migratoria manilensis Meyen) for realizing efficient control and prevention of this kind of insect pest.In this study Huanghua County of Hebei province was chosen as the study area and based on the hyper-spectra data measured at ground surface,the difference in canopy reflectance spectra and the characteristic parameters of hyper-spectra data were analyzed and compared for the reeds in normal growing condition and for those under damaging from locust.In addition,five kinds of model were developed to reflect the relations between the characteristic parameters of hyper-spectra and Leaf Area Index(LAI) of reeds.The results show that among those indices the locust damage spectra index(DSI) is one which is mostly suitable to use in indicating the intensity of locust damage in the study area.Following this step,a classification scheme for the locust damage intensity was worked out based on DSI data for use in the study area,i.e.no damage if DSI is over 62.856,slight damage if DSI is between 41.254 and 59.496,and serious damage if DSI is less than 41.254.
关键词:hyper-spectra;Leaf Area Index(LAI);locust damage spectra index(DSI);oriental migratory locust;characteristic parameters of hyper-spectra
摘要:Remote Sensing data collection planning of satellite is made according to(customer’s) request,satellite instruments and ground station available.The varity requests of(customer’s) and the instrument constrains of satellite.Lead to an optimization problem,that is the utility of the satellite resources and ground station resources.This paper analyses the factors affecting satellite mission planning,then presents algorithms rules for remote sensing data collection planning.According to these algorithms rules,we established the executive flow for optimization algorithms which can solve the conflicts between(customer’s) requests and utility constrains of satellite and ground station.The(customer’s) requests could be satisfied more effectively.
摘要:In the paper,the change of vegetations’ primary productivity was used as a central evaluating factor,the Defense Meteorological Satellite Program /Operational Linescan System(DMSP/OLS) data and multisource satellite and geospatial data were combined in a geographic information system,and the Normalized Difference Vegetation Index(NDVI) was used to quasi-quantificationally study the effect of urbanization on the photosynthetic productivity and its seasonal changing law in the urban agglomerations of Bohai Rim,from the view of land use change.This work is an important preparation to further study the ecological effects of urbanization in all country scope.The conclusions are as followings:(1) The yearly mean total productivity of the urban portion is less than that of non-urban part.(2)The yearly total average productivity commonly in August rises to its tiptops,and in January touches down its nadirs.Moreover,in one growing season,the tendency of the productivity of the urbanized portion is less than that of non-urban part from April to November,but from December to March of next year that is precise antithesis.In the meantime,the tendency also has some differences to every land cover class.(3) In a global scale view,the yearly total average productivity for urban and non-urban areas of all the study area including all the land cover types is 110.23d/km2 and 123.94d/km2, respectmely which has a dispersion of 13.71d/km2.It indicates that the continuing process of urbanization has been weakening the primary productivity of the Bohai Rim.
摘要:The estimation of soil organic carbon content(SOC) is one of the important issues in the researches of global carbon cycle.However,the estimated magnitude of SOC exist great differences among different scientists.There are two commonly used methods for the estimation of SOC,each method having both advantage and disadvantage.One of the methods is so called direct method,which is based on the samples of measured SOC and maps of soil or vegetation types.The other is so called indirect method,which is based on ecosystem process model of carbon cycle.The disadvantage of direct method is that it mainly discloses the difference of SOC among different soil or vegetation types,so it can hardly distinguish the difference of SOC in the same type of soil or vegetation.The indirect method,process-based method,bases on the mechanics of carbon transfer in the ecosystem and then could potentially improve the spatial resolution of the SOC estimation if the input variables have high spatial resolution.However,due to the complexity of the process-based model,the model usually simplifies some key model parameters that have spatial heterogeneity with constants.So this simplification will produce a great deal of uncertainties in the estimation of SOC,especially on the spatial precision.In this paper,the authors combined the process-based model(CASA model) with the measured SOC,in which the remote sensing data(AVHRR NDIV) was incorporated into the model to enhance the spatial resolution.To model the soil basal respiration,the Van’t Hoff model was used to combine with CASA model.The results showed that this method could significantly improve the spatial precision(8km spatial resolution).The results also showed that the relationship between soil basal respiration and SOC is very well,as the influence of environmental factors,i.e.temperature and moisture,have been removed from soil respiration,which makes the SOC become the most important factor of soil basal respiration.The statistic model of soil basal respiration and SOC showed that the determinant coefficient(R2) is 0.78.As the method in this paper contains advantages from both direct and indirect method,so it could significantly improve the spatial resolution and,at the same time,keep the estimation of SOC well matched with the measured SOC.
关键词:soil organic carbon;soil base respiration;remote sensing;process-based model;China
摘要:Besides empirical algorithm with the Blue-Green Ratio,fluorescence remote sensing of Chlorophylls is also an important and valid method for retrieving chlorophyll-a concentration in the ocean,especially for Case Ⅱ waters and the sea with algal blooming.This study reviews the history of initial cognitions,investigations and detailed approaches towards chlorophyll fluorescence,introduces the biological mechanism of fluorescence remote sensing and main spectral characteristics such as the positive correlation between fluorescence and chlorophyll concentration,the Red-Shift phenomena.In addition,there are many influential factors which increase the complexity of fluorescence remote sensing,such as fluorescence quantum yield,physiological status of various algae,substance with related optical property in the ocean,atmospheric absorption etc.Based on these cognitions,scientists have found two ways to calculate the amount of fluorescence detected by ocean color sensors: fluorescence line height and normalized fluorescence height.These two ways are currently the foundation to retrieve chlorophyll-a concentration in the ocean.As the in-situ measurements and synchronous satellite data are accumulated,the fluorescence remote sensing of chlorophyll-a concentration in Case Ⅱ waters should be understood more deeply and new algorithm could be expected.
关键词:chlorophyll-a;fluorescence remote sensing;fluorescence line height;normalized fluorescence height;MODIS;MERIS