Satellite-based passive microwave remote sensing data has wide applications in global environmental change research because of the characteristics of all-day observation,penetrability and uninfluenced by clouds.The main drawback of passive microwave remote sensing is the extreme low spatial resolution,which limits the precision of surface parameters retrieval.Due to the restriction of size and weight from satellite payload,satellite passive microwave radiometer can not assemble a large antenna,which causes low spatial resolution.The spatial resolution improvement methods of passive microwave remote sensing data are introduced in the paper.The super spatial resolution method and passive microwave unmixing method based on image processing technology are mainly discussed.The research prospects of the unmixing methods are proposed after the introduction and evaluation of present methods.The effective improvement of spatial resolution of passive microwave data can be used to service for more research and application fields well.The research results provide important information for passive microwave unmixing method with higher precision and more application field.
The high resolution satellite technique has an ability to detect the smaller targets on the ground.Vehicle detection using high resolution satellite imagery has become a hotspot in remote sensing research field.This paper firstly analyzed the study situation of vehicle detection from remote sensing imagery,and then discussed the imagery features of vehicles,the process and the primary methods for vehicle detection,including the basic detection methods using spectrum/structure features of vehicles and the intelligent methods combining various of vehicle features.The crucial technologies to detect vehicles were introduced in detail as well.Finally,this paper concluded that methods such as machine learning or object-oriented method using more object features can be more adaptable to complex road environment.
In the current agricultural production and management,there is a urgent need of the direct and immediate informaiton guide.With the improvement of technology,it is possible to realize the non-destructive prediction of crop qualities by real\|time monitoring the main impact factors of crop growth through different means of remote sensing technology.This article describes the quality properties of a variety of crops and the formation factors.Also the commonly used spectral parameters are introduced in the induction of monitoring crop qualities.Based on the respective ground and aerospace platform,an overview of remote sensing research about monitoring the main crops quality and the main method of remote sensing to monitor the quality is given.The problems are analysed that about remote sensing technology in crop quality,and solutions also be concreted.
Accurate registration and orthorectification of remote sensing image is the important fundament for image fusion,change detection,image mosaic,quantitative remote sensing modeling,and synergy application of multitemporal and multisensor images.This paper takes the program package named LEDAPSAROP (Landsat Ecosystem Disturbance Adaptive Processing SystemAutomated Registration and Orthorectification Package) of NASA as a case,to illustrate its principle and program design,and analyze its accuracy of registration and orthorectification.The experiment shows that AROP can search enough control points in reference image,and these points distribute evenly.The error of registration and orthorectification is no more than 0.5 pixels,which can meet the requirement of remote sensing quantitative application in mountainous regions.The characteristic of errors presents that the lateral error is obviously bigger than the course error,and the errors seem correlate to the DEM and slope.At present,AROP can be used to process the registration of CBERS image,but still need study further for deal with HJCCD image.
According to the feature of line frequency changed with aircraft altitude and speed for the airborne hyperspectral imager based on line array CCD,the output DN would be different under the same conditions of incident radiation.Therefore there would not have comparability between each line ground objects of images,especially there would be color difference between adjacent strips.A relative radiometric correction model based on line frequency difference was developed.By introducing the effect of line frequency difference to the traditional relative radiometric correction model,a three-dimensional surface reflecting the effect of line frequency difference for relative radiometric correction of airborne hyperspectral data was obtained. The verification field images obtained from a UAV hyperspectral imager was processed for comparing the color difference of adjacent strips and analyzing the corrected uniform targets,the results show that the non-uniformity of images has been corrected well and the deviation by line frequency difference has been conquered.
It is known that the same or even similar spectrum features and quantized data exist in the same type of the surface features in an image by analyzing the spectrum and surface features.And there are many correlations between signals in homogeneous regions.But the noises are random,independent and de-correlation.It means that the spectrum reflecting values in homogeneous regions of a remote sensing Image are correlating.The noises can be separated from the values,if the real signals can be estimated based on the correlation.The paper presents a method,called extracting edge and spatial dimension correlation (EESDD),to estimate the standard deviation (SD) of signals or signal-to-noise ratio (SNR).The method is full use of the surface features to extract the edges,and estimate the SD or SNR by the correlation of signals in homogeneous regions.EESDD can be applied in estimating noise not only for a single wavelength image,but also for a hyperspectral image.And this method is used for the homogeneous and inhomogeneous regions,too.Comparing to the local mean standard deviation (LMSD) and the edge-extracted local mean standard deviation (EE-LMSD) in Hyperion data with different scene contents,the result of EESDD is far better than those of LMSD and EE-LMSD.
ICESatGLAS waveform data have being used widely in estimation of ecological parameters of forest in recent years.In order to judge the potential of large footprint LiDAR data of ICESatGLAS on estimating the forest volume in complex terrain,a case study of Shangrila,Yunnan Province,comparison between the average canopy height retrieved from the GLA01 data and canopy height measured in the field and slope data,and the impact of slope on the canopy height estimation by GLAS data had been also explored.Meanwhile,the relationship among the slope,the average canopy height and the field measured volume in large footprint had been discussed.The results show that the slope can reduce the accuracy of the estimation of the canopy height by using large footprint LiDAR data.However,there is a good relationship among the average canopy height derived from the GLAS data,canopy height and volume measured in the field and slope.It indicates that the potential of forest volume estimation by using GLAS data is great.
In recent years the significant changes have been occurred in the land surface environment of the megacities,and it has also brought the great impact to the ecosystem in regions.Therefore,it is urgent to quantity various kinds of land surface characteristic parameters in urban and research their relationships,so as to analysis their effects towards the urban heat environment.For this reason,we take the municipality of Beijing as an example to extract four typical surface parameters with Landsat-5 TM remote sensing image on June 2,2009,including the percentage of impervious layer,land surface temperature,land use/land cover and vegetation index,and their quantitative relationships have been analyzed.Finally,it can be found that the high impervious surface region has been expanded to the six-ring in Beijing with the rapid development of the urbanization,and the LST maintained at 40 ℃ or more within the six-ring,where it was the high temperature region.Especially in the business districts the LST was as high as 45 ℃,and the LST fluctuant magnitude was small within the six-ring.Whats more,the forest and agricultural land played the role of cooling effect at the maximum cooling scale of 6 ℃,and it was clear that the LST of the bared soil region nearly arrived at that of high-density resident in the summer.
Evapotranspiration(ET)as the main form of exchanges between the water and heat to the soil—atmosphere of the wetland ecosystem greatly affects the water thermal equilibrium.Estimation of the evaporation quantity reasonably and accurately is one of great significance to study hydrologic recycle,energy balance and scientific management of the wetland ecosystem.Wetland in Yellow River Delta(YRD),which is the most integrated,broadest,youngest wetland ecological system in warm temperate zone of the earth,is not only sensitive to climate change,but also fragile to the environment change.Remote sensing technique was used to estimate ET in YRD in our case study.Coupling the model of Surface Energy Balance Algorithm for Land(SEBAL)and Two=Source Energy Balance(TSEB),SEBAL was used to calculate land surface characteristic parameters and flux,and TSEB was used to calculate vegetation transpiration,soil evaporation and total ET separately of YRD by Considering the inner structure of the surface layer which could be divided into two sub layers representing soil and vegetation.Actual evapotranspiration of the study area was meatured from stations and calculated with Penman=Monteith equation to compare with results from remote sensing,which showed that the result is reasonable.Then the spatial distribution of the daily ET was analyzed and the characteristics of ET with different features,especially with the typical vegetation of reed marshes and meadows,were also analyzed to help realize the wetland conditions.The results illuminates that daily ET estimated is reasonable.The method can be used for the accurate estimation of ET in YRD.
Developing effective methods for urban vegetation stress detection should be an important part of urban forestry resources management,and also for the favorable urban ecology and environment construction.In this paper we use Hyperion hyperspectral images in the eastern Guangzhou city,China,which is captured by the EO-1 satellite.Based on satellite-borne hyperspectral data,we want to conduct the vegetation stress identification,through these two methods:on one hand we selecting the appropriate vegetation indices for image classification;on the other hand we through “Spectral Mixture Analyst” to obtain vegetation abundance,and conduct several experiment steps as well.These two methods are both based on atmospheric correction preprocessing.Comparing these experimental results,it shows that in the feature extraction of vegetation,the vegetation abundance analysis methods is higher precision than the vegetation index method,which non vegetation feature spectral is excluded;and we through field spectrum measurement for validation,the result supports that the abundance analysis which is vegetation spectral theory base is better to characterize the vegetation stress feature,image spectral and field measurement have certain similarity.This research above will be for the application in urban forestry management by qualitatively and quantitatively.
Based on the AVHRR GIMMS NDVI and MODIS NDVI,this paper constructed yearly maximum NDVI sequence covering Northeast China from 1982 to 2009 using per-pixel unary Linear Regression Model.The constructed NDVI through the consistency check and well be used in the temporal analysis and correlation analysis with climatic factors.Results showed that there were three change stages in the past 28 years:A modest increase trend during 1982～1992,slow decrease trend during 1982～2006 and a recovery during 2006～2009.Result from variation analysis on spatial pattern showed that obvious changes occurred mainly in Inner Mongolia Automomous Region (IMAR) arid and semi-arid zone.Mean NDVI value in 2000s and 1990s were bigger than 1980s.Change value ranged from -0.3 to 0.3 and increase trend were mainly found in Chifeng and west of Songnen plain which have a proportion about 6.45% in total area.A high spatial heterogeneity was observed in the correlation between yearly maximum NDVI and climate (temperature and precipitation).Pixels which had a bigger coefficient of partial correlation between NDVI and temperature than between NDVI and precipitation account for 54% of all pixels.At whole scale,temperature was the dominate influence factor compared with precipitation.For different vegetation types,the yearly maximum NDVI was affected by annual average temperature as follows:forests>grasslands>marshes>shrubs>crops; and by precipitation:grasslands>crops>shrubs>marshes>forests.
Polarization information is a unique advantage of SAR data and opens up a new way for radar remote sensing application research.Polarization decomposition is a new type of polarization processing approach,analyzing the scattering mechanisms from the point of mathematics and physics in order to identify the target.In this study,we use Quad-polarization RadarSat-2 data to study polarization response of rice and its variation in time domain,analyze scattering mechanisms of rice,typical features and their differences,and then extract rice based on the characteristics of its scattering mechanism in hilly areas of Guizhou plateau.
The road is a very important geographic dataset and it plays an important role in national economy and national defense. Research on the automatic or semi-automatic road extraction has always been regarded as a hot and difficult topic. Although lots of theories and methods have been proposed,there is no one that is widely used in production environment. Since road extraction from high resolution remote sensing images is difficult,in this paper,we propose a new method for road extraction. Firstly,a RGB color space image is converted to HSV color space image and an improved regional segmentation algorithm is used to separate the road regions. Then,a combined mathematical morphology operation (such as erode,dilate etc.) with regional shape indices (such as area,compactness etc.) is applied to separate the road objects and non-road objects. The experiment results over several complex road images show that this method can be used to extract roads under more complex conditions.
It has the features of time-sensitive,short cycle and quick operation etc that making thematic mapping with road information extracted from remote sensing image.In this paper,having introduced the method and process of road extraction,treatment and production of thematic map.Having the Google high resolution image as the data source,and extracting the road nets to produce the thematic map of the study areas by combining the spectral and geometric feature information of the road and the method of computer classification and morphological processing.The result of test shows that the method can meet the demand of thematic mapping,and the road nets extracted from the Google Image are very clear,and has good treatment result on the cavity and edge erosion of the road caused by vehicles,people and shadow.So it is an economical,efficient and practicable method.
Cloud shelter in the optical remote sensing image may cause low data utilization rate and affect the subsequent process of remote sensing image such as target identification,so the research of real time and efficient cloud detection method is very important.We proposed a high speed and high accuracy classification algorithm for the cloud classification based on the combination of Support Vector Machine(SVM) and unsupervised clustering algorithm.This method uses the ISODATA clustering results to select the training set for SVM in order to reduce the training time of SVM.It takes advantage of the high accuracy capability of SVM and the fast clustering speed of ISODATA.The experiment shows that the SVM training time in the proposed method is greatly lower than it in the method using SVM alone,and the proposed method can improve the detection rate.
To explore the oil and gas of the most potential but harsh natural conditions domestic area Tibet,remote sensing interpretation technology was used to study in Biru basin.Firstly by the observation of different lithology,strata,linear structure on the field reconnalssance route,the feature of lithology,strata,linear structure in the remote sensing images were reduced.On the basis of it,the interpretation key of lithology,strata,linear structure was presented.According to interpretation key,the lithology,strata,linear structure and circular structure of Biru basin was interpreted.And then oil and gas information of remote sensing image over the known oil & gas micro-leakage and abnormal information area were superimposed by different methods such as band ratio,tone abnormity on the basis of fusion of band and Principal Component Analysis (Hydroxyl/Iron),finding that hydrocarbon micro-leakage abnormal area and the known gas and oil shows are positive correlation.The abnormal information area and Lunpola basin oil reservoir were superimposed,finding that the abnormal information area and underground reservoir are negatively related.So the hydrocarbon micro-leakage non-anomalies area is possibly the petroliferous area.The hydrocarbon miscro-leakage abnormal area was determined by those methods,and then delineated one first-classic micro-leakage non- anomalies area,combining basin source rock,reservoir,cap-rock and trap conditions,pointed out that the area enclosed by Biru—Boliechan—Bianba was considered to be the first class oil & gas beneficial exploration areas,is the next objective of oil and gas exploration.
Normalized Difference Vegetation Index (NDVI) is an important parameter that can well reflect the growth characteristics of vegetation.The retrieval of vegetation indices through satellite data are usually influenced by the sensor itself,the atmosphere,aerosols and others.Thus,the derivation of objective and accurate vegetation index is essential to reveal the vegetation growth changes and other quantitative remote sensing analysis.Based on the different levels of radiation correction (radiometric calibration and atmospheric correction),we used the digital numbers (DN) values originally from Landsat ETM+,TOA reflectance,and surface reflectance for the calculation of NDVI,respectively.According to field survey on paddy rice cropping systems in the Poyang Lake Region in China in mid-to-late September 2010,from the paddy field and the landscape scales,we monitored the temporal dynamics of Landsat-derived NDVI.The results showed that,the NDVI profile calculated from DN values,TOA reflectance and surface reflectance basically reflect the information on the paddy rice growing systems per year,and the corresponding ranges of NDVI became bigger.NDVI dropped to its low level during the transplanting stage and reached a plateau level during the panicle initiation to heading.In the development of paddy rice growth period,the NDVI curves derived from TOA reflectance and Surface reflectance could better demonstrate the growth system characteristics of paddy rice.Therefore,we concluded that the Surface reflectance retrieved from radiometric calibration and atmospheric correction should be a relatively objective and accurate parameter for NDVI calculation in the quantitative remote sensing of vegetation growth condition.
The vegetation cover is an important factor to evaluate the regional ecological environment in the terrestrial ecosystem.In this study,we detected the vegetation coverage, using Chinese Environment and Disaster Monitoring and Forecasting Satellite (HJ-1A CCD) and Landsat TM data,applying the models of Gradient Difference by Three-band modol and Fraction Dimidiate Pixel Model with parameters of NDVI and RDVI.Through comparing the accuracy of the classification results respectively by different models and different paraments,we discovered that the Fraction Dimidiate Pixel Model with RDVI got the best effect.In view of this,we analyzed the dynamic changes of the vegetation coverage between 2000~2009 in the study area,then pointed out that the vegetation recover is well.
There are many accumulation river islands in Ma-wu-tong Section of the Yangtze river,and the area of river islands change rapidly.We chose eight river islands and studied their changes on TM (Thematic Map),and the dates of MSS/TM/ETM are 1979,1989,2001,2006 and 2009.We determined the spectrum of typical objects in river islands by ISI921VF-256T Field Spectroradiometer,combined the right bands,and interpreted the images by geometric correction,subset images,binarization processing,convolution and clump.The results show that:① TM remote sensing data can be used to achieve change information of river islands quickly and accurately;② The eight river islands extended continually during the past 30 years,and the extended speed was not averagely,which might be resulted from reducing water or decreasing the sediment-carrying capacity;③ Evolution of river islands are controlled by natural factors and human factors.A direct result of natural factors is ablation or siltation of river islands,and human factors change would result in the change of external substances,which is composition of river islands.
Based on the division of regions on Helan mountains,the vegetation changes of four seasons from 1998 to 2010 in different regions has been analyzed using the SPOT-NDVI data.The results shown that average vegetation cover was highest in high elevations and second highest in middle elevations,and then the low elevation areas.The vegetation cover on east slope of the study area was higher than the west slope,north part and south part of Helan mountains.The growth period of vegetation was from May to October,with the highest vegetation cover in August.The vegetation was generally getting better in the past 13 years,in which the vegetation grew better in autumn for most areas,especially in the middle and high elevations areas.However,considering the negative effects of the climatic arid trend on vegetation growth in the study area,the trend of vegetation grow would probably reverse in the future.
After Yunnan,Guizhou and Guangxi provinces were designated the restricted development zone,studies on karst rocky desertification had been becoming more and more enthusiastic.Information extraction is an important content of the studies.Current methods for karst rocky desertification information extraction are based primarily on static index about Karst Rocky Desertfication phenomena,as has obvious shortcomings.Given the shortcomings of current methods and integrity of static and dynamic information of karst rocky desertification,this paper discusses automatic method of karst rocky desertification information extraction using change detection based on the mechanism of the karst rocky desertification.Thereby this paper takes the border region of Guizhou and Yunnan provinces as a study case to test the application of change detection method using two-phase TM/ETM+ images in 1992 and 2001.Differential change detection is made in geometric and radiometric corrected images in 1992 and 2001,the result of change detection is classified as unchanged region and changed region which is divided into two parts of karst rocky desertification and retrograde desertification based on mechanism of the karst rocky desertification.Recognition accuracy for the kappa coefficient reaches 0.854 in the test region.Finally,the finding shows that the method using change detection has better efficiency and accuracy in the recognition of desertification information,as has important significance to the countermeasures of karst rocky desertification.
Interferometric synthetic aperture radiometer is an attractive conception that uses Fourier synthesis to achieve high spatial resolution avoiding the disadvantages of manufacturing and carrying large antennas.183.31 GHz Interferometric aperture synthesis radiometer will play a significant role in detecting the vertical distribution of atmospheric water vapor,heavy rainfall and cloud liquid water.It makes great sense to provide important parameters for the weather prediction as well.This paper has discussed the analysis of interferometry model and design of the 183.31 GHz interferometer system with two elements.And then two main important tests like simulated point target interferometry and point target interferometry at various angles to get functional verification of the full-scale system are presented.Simulated point target interferometry experimental result which illustrates that the measured amplitude of visibility function is the same,and the results distribute on a round show good stability,which is crucial for a full-scale system.Point target interferometry at various angles shows that real and image parts of the complex visibilities are very close to the expected ideal orthogonal sinusoids.The relationship between baseline length and the real and image parts of the complex visibilities curve frequency is linear which verifies that visibility function and modified brightness temperature is Fourier transform pair alike.
