In view of the nonexistence of geometric model for epipolar image, this paper, based on the RPC model, puts forward a practical way of building geometric model from epipolar image generated by the projection track method. With the support of across-track stereo image data of SPOT 5 HRG, CBERS 2-03 and along-track stereo image data of SPOT 5 HRS, P5, GeoEye, IKONOS, this paper indicates that the RMS of vertical parallax of the epipolar image is within 0.2 pixel,  that the planar RMS calculated from forward intersection by the epipolar model and the original image model is within 1m, and that the altitude RMS is within 0.3 m.

This paper deals with a suitable atmospheric correction method based on HJ-1A CCD Image. A method based on 6S radiation transfer code and isochronous meteorological information on sensing time was used in the atmosphere correction，which is an operational method of correcting visible and near-infrared data for atmospheric effects. Validation analysis was conducted by comparing the reflectance data corrected and uncorrected atmospherically. Significant differences occurred between them: the values of the corrected reflectance of CCD1~3 decreased obviously，whereas those of CCD4 increased. The retrieved surface reflectance was evaluated against the ground-based reflectance acquired during HJ-1A satellite overpass，and it is shown that the curves of the figures are similar， and the compared values are agree with each other well. Based on the statistical analysis，some reasons for the errors were mentioned in the end of the paper.

 ERS Wind Scatterometer provides capability of the multiple angles by three different look antennas. In this study, the authors have made an evaluation whether the multi-incidence angle observations can help improve surface soil moisture estimations or not. With the theoretical surface backscattering model, i.e., the Advanced Integral Equation Model (AIEM), the authors first simulated a surface backscattering database with a wide range of surface roughness and soil moisture properties at different incident angles. Then,a parameterized surface backscattering model was developed using the simulated database. The newly developed simple model has the roughness function that can be described by a single combined roughness parameter from the commonly used surface roughness descriptors (RMS height and correlation length). This makes it possible for the model to be used as an inversion model.  The development of this simple model, its accuracy, and the inversion test can be demonstrated by using the ground measurements from the Intensive Observation Period (IOP'98) field campaign in 1998 of the Global Energy and Water Experiment (GEWEX) Asian Monsoon Experiment Tibet (GAME/Tibet).

 The radiometric calibration parameter is very important in quantitative application of remote sense data. In this paper, the Landsat Thematic Mapper Data of Beijing acquired in April 26, 2007 were applied to evaluating the radiometric calibration parameters supplied by USGS and RSGS. First, the radiance data were computed respectively with the calibration coefficients converted from calibration parameters.Second, the FLAASH atmosphere correction model was used to compute meteorological range and surface reflectance.Third, according to the aerosol observation data from AERONET and high resolution remote sensing image, the accuracy evaluation results of the two radiometric calibration parameters were obtained on the basis of the accuracy of meteorological range and classification of surface reflectance. The results show that parameters supplied by USGS can reflect radiometric features of TM sensor more accurately.

 In order to overcome the deficiency of sampling methods for SRTM and ASTER, the authors constructed a flow chart for SRTM and ASTER efficient fusion, with the DEMs of Dongzhi tableland as an example. Firstly, ASTER and SRTM were respectively transformed from spatial domain to frequency domain in terms of Fast Fourier Transform (FFT). Secondly, the ideal high and low pass filters were respectively employed to remove their low and high frequency errors. Thirdly, the filtered frequency domains were added up. At last, the summed frequency domain was transformed to spatial domain in terms of inverse FFT. The results indicate that, compared with errors of ASTER and SRTM, both the minimum and maximum errors of the fused DEM show an obvious decrease, the RMSE has a weak decrease and the number of the errors bigger than 30 m is much less than that of SRTM and ASTER.

 The Modulation Transfer Function (MTF) is an important comprehensive evaluation index for the performance of the optical imaging system and also the most important part of image degradation model. According to the characteristics of TDI-CCD image data，this paper discussed the ways of calculating MTF of the image system based on the edge detection in remote sensing image，and implementing MTF compensation to the degradation image with winner filter and modified inverse filter (MIF) respectively. The results show that the adaptive MTF restoration processing method can effectively improve image quality of high resolution TDI-CCD image data. At present，this method has been successfully applied to the domestic ground satellite pre-processing system in China.

In this paper，the splits window algorithm developed by Qin was used to retrieve the snow surface temperature (TS) from MODIS 1B data by using Interactive Data Language (IDL) programming. An IDL program was written for the TS retrieval. The retrieved TS data were saved in the HDF file readable into other software. Details of image processing and programming were shown in this paper. Finally，the authors chose northern Xinjiang as an example to demonstrate the applicability of the approach. The retrieved TS data were compared with the observed snow temperatures from automatic meteorological stations. The result of the comparison indicates that the retrieved TS distribution is consistent with the observed one, and the average error is 1.73℃, which is acceptable. Therefore, the TS distribution in northern Xinjiang could be rapidly retrieved using the proposed IDL approach.

 Remote sensing image signals are random signals whose band data histograms and two-dimensional scatter plots are respectively the basic estimates of one-dimensional probability density function and two-dimensional probability density function of the random model. In this paper，the basic types and patterns of one-dimensional probability density and the geometric parameter characteristics of Gaussian distribution ellipses are discussed based on the random models of remote sensing data. Then，problems such as the spatial geometric structure characteristics of the two-dimensional scatter plots generated by different histograms and the positioning in abnormal information space are analyzed. At last，the importance and practicability of the spatial structure analysis of remote Sensing data in alteration information extraction are demonstrated by a case study.

High-resolution remote sensing images have rich texture information, and combined texture information and image spectral information can improve the recognition accuracy of surface feature. In this paper, a new extended Local Binary Patterns (LBP) texture was applied to the high-resolution images classification in comparison with classifications using spectral data only and using combined spectral data and LBP texture features. The results show that the extended LBP has a good anti-noise performance, and the classification of image including the extended LBP texture can achieve a higher accuracy than the classifications using spectral data alone and using combined spectral data and LBP texture features.

In this study, spectral characteristics of water bodies and other objects in ALOS AVNIR-2 imagery in areas near Huizhou of Guangdong Province were analyzed. The impact of the threshold on the sensitivity of extracting water and the maximum precision of the four different water index models were studied. It is found that the water area in the image can be extracted well using these index models. The result shows that the application effects decrease in order of NDWI，NIR，NDVI and RVI. If the proper threshold is chosen，the overall accuracy can reach the highest and the value is nearly 98 % by using NDWI，which, moreover, is not sensitive to the threshold.

 Based on fine texture expression of the edge of the detailed information in the digital aerial image，the authors first preprocessed the digital aerial image by edge detection，principal component analysis and the texture filter of second-order probability statistics，secondly obtained the gray image of the contrast texture through the sharpening window of 7×7，then taking the gray image as an independent band，processed a pseudo color composition with the band combination of contrast (R)，the digital aerial image band (G) and the digital aerial image band (B). Finally，multiple segmentation and building extraction in towns and villages based on the pseudo color images were processed. With the digital aerial image acquired in December 2008 from Kangzhuang Town of Yanqing County in Beijing as an example，the authors processed building extraction by using  the method mentioned above. Compared with the object-oriented classification method，it not only highlighted the edges of the buildings but also reduced the redundant segmented objects. Besides，it achieved an effective solution of the shadow of the building and its confusing area，optimized the feature space，and improved the classification accuracy.

The key point for gold exploration using remote sensing technology is the information extraction of ore-forming structures and spectral characteristics of ferrite oxides and altered hydrous minerals. On the basis of the remote sensing image processing software (ENVI 4.0) and by using a new quick, exact and effective method for extraction of gold mineralization and alteration information，namely "multivariate data analysis＋ratio＋PCA＋DS＋classification"，the spectral characteristics of altered minerals can be extracted，and the influence of the vegetation can be suppressed. It is demonstrated that the gold mineralization and alteration information is in accordance with the known mine spot，and the method is very feasible.

 With the population growth and the rapid development of national economy, urban construction scale has also been growing. Urban expansion has led to a sharp reduction of arable land, which in turn seriously restricted the social and economic development. In this paper, adopting the remote sensing data and topographic map data acquired in different periods and using image processing and interpretation of human-machine interactive information retrieval method, the authors obtained the scale of urban development and land use change information of 10 different periods in Changsha. According to statistical analysis and data distribution, a mathematical model for urban growth in Changsha was established, and the factors responsible for helping and restricting the economic development in Changsha are analyzed and evaluated.

 Waterside line information extraction is a foundational task in studying mudflat sedimentation/erosion variation. This paper puts forward a new identification method more suitable for the ring waterside line information detection than the traditional Hough-like method. The authors firstly detected edges from the image segmentation achieved by the waterside line gray gradient information, then dilated and filled the interior by mathematical morphology to connect edges, and finally constructed segment-shaped structural elements to achieve a continuous and  smooth waterside line. Experimental results show that not only the method is simple, but also the extracted waterside line is ideal.

 In this paper, the land surface temperature (LST) was retrieved from Landsat TM image using the Mono-Window algorithm. To study the relationship between LST and different land use classes (LUC), the authors used several indices, which included Normalized Difference Vegetation Index (NDVI), Modified Normalized Difference Water Index (MNDWI), Normalized Difference Build-up Index (NDBI), and Normalized Difference Barren Index (NDBaI). It is found that the correlation between NDVI and LST is negative when NDVI is limited in range, and that there exist positive correlations between NDBI, NDBaI, MNDWI and LST. The multiple linear regression equation was established between LST, DEM and the above indices. Both qualitative and quantitative analytical results show that LUC can influence urban temperature. Therefore, with appropriate land use planning, the urban heat island (UHI) could be mitigated.

 Land use and land cover change constitutes significant problems in the study of global environmental change. In this paper, the authors used the single decision tree method to classify two images of Landsat TM/ETM+ in 1989 and 2000 and obtained land use patterns of the coastal zone around Jiaozhou Bay. In this process, atmospheric correction and geo-knowledge study are recommended. Then through "overlay" analysis, the transfer matrix of land use can be obtained. The characteristics of land use/cover change are mainly demonstrated as follows: land for construction use such as cities and countries increases rapidly at the expense of occupying large amounts of farmland, land for marine culture and saltern expands by taking the place of some farmland and sea area; nevertheless, farmland and inland beach decrease greatly, and land for forest and grass changes very little. In the past ten years or more, human activities such as increasing the amount of waste gas, waste water and industrial residue, dumping a lot of land-sourced nutritive salt and pollutants into the Jiaozhou Bay, have exerted increasingly strong effect on ecological environment,  which finds expression in the following aspects: the water of Jiaozhou Bay has deteriorated, the amount of heavy metals in sediments has increased, the population and biodiversity of plankton have decreased and harmful algal blooms have happened frequently.

As an important part of Beijing’s suburbs, the mountainous area of Beijing has both the characteristics of metropolitan suburbs and the features of mountains, and so its land use pattern is different from other areas. Based on statistics, the authors separated land-use transformations of the mountainous area of Beijing to random and systematic transitions through an in-depth analysis of land use transition matrix. A land-use transition is random if a land-use category gains from other categories in proportion to the availability of those other losing categories, or if a category loses to other categories in proportion to the size of those other gaining categories. Any large deviation from those proportions is referred to as a systematic transition. From the concurrent incidence of systematic gains and losses, it is concluded that the most dominant signals of change find expression in the following aspects: the conversion of higher coverage grassland and other woodland to cropland, the conversion of cropland and sparse woodland to other woodland, the conversion of sparse woodland to higher coverage grassland, the conversion of cropland and higher coverage grassland to water areas and the conversion of cropland to built-up areas. The transition areas from cropland to other woodland and built-up areas are relatively large and have obvious advantages for systematic transitions. While the areas of transitions from other woodland to cropland and from sparse woodland to higher coverage grassland are relatively small, the transitions are still attributed to dominant signals of change. This method is simple and practical and enables scientists to focus their research on the most important land transitions, which will help in linking patterns to processes and in management of land resources.

The surface deformation is one of the available indicators for monitoring and analysis of the coalfield fire area. However, it is difficult to extract surface deformation in the coalfield area if serious decorrelation noise exists. Based on analyzing the surface deformation characteristics of the coalfield fire area and comparing a variety of SAR datasets, the authors chose L band ALOS PALSAR dataset to conduct a differential synthetic aperture radar interferometry processing. Interferometric fringe was calculated for estimating the accurate baseline. The adaptive filter method was used to eliminate decorrelation noise after generating interferogram. Then surface deformation in the coal mine fire zone was extracted after two processes, i.e., flattening and topography phase removal. It is found that surface deformation is correlated with the coal spontaneous combustion zone to some extent, as evidenced by past in situ investigation and data analysis. The surface deformation analysis is helpful to confirming the situation of the underground coal fire burning. The on-site verification proves that detecting surface deformation of the underground coal spontaneous combustion zone by D-InSAR is acceptable.

 In order to extract the open water from the Poyang Lake by applying SAR data in different periods, the authors firstly analyzed the scattering mechanism of water，vegetation and sand，then used the texture of SAR imagery and polarized ratio and polarized difference to enhance the description of targets, and employed principal component transformation to enhance the contrast of water and other objects. With the first component，the area of open water on SAR imagery could be effectively and accurately extracted by setting up a proper threshold. The SAR images used were Envisat-ASAR and ALOS-PALSAR alternating polarization mode data. Meanwhile，after the analysis of two kinds of sample data with Jeffries-Matusita distance，ASAR data with C band could provide more accurate extraction of open water than those of PALSAR with L band. Finally，the open water of the Poyang Lake was extracted from SAR imagery in spring，summer and winter，and the changes of water area indicated the regular change of the open water of the Poyang Lake.

 The Differential Interferometric Synthetic Aperture Radar (D-InSAR) based on the ascending and descending orbit data can better reveal the feature of coseismic deformation fields. With the ASAR data from European Space Agency (ESA) and two-pass D-InSAR measurement, this paper obtained both ascending orbit and descending orbit coseismic deformation fields of Yutian Ms7.3 Earthquake. The results show that the max LOS uplifting value is about +13.3 cm and the subsiding value is about -82.0 cm in the ascending-orbit coseismic deformation field, and the max LOS uplifting value is about +36.5 cm and the subsiding value is about -66.5 cm in the descending-orbit coseismic deformation field. The Yutian Earthquake is mainly rupturing along the NNE-trending normal fault with a little left-striking, the northwest plate is the hanging-wall (subsiding) of the normal fault, and the southeast plate is the footwall (uplifting) of the normal fault. The coseismic deformation fields of ascending and descending orbits are different from each other in some aspects, but their variable tendencies and characteristics are similar, and the difference is mainly due to the two different observation modes.

With Hefei City as the study area, the authors investigated the relationship between urban landscape pattern and hydrological effect. The landscape classification maps were obtained from Landsat TM/ETM+ images taken in 1995, 1999, 2003 and 2007 with the support vector machine classifier. The thermal radiance transfer equation was estimated by mono-window algorithm designed by Qin Zhihao. The relationship between urban landscape pattern and hydrological effect was analyzed using landscape indices derived from the land cover classification maps. The result shows that urban hydrology and landscape structure are highly correlated with each other in Hefei City. With the increase and centralization of the built-up land, the surface runoff and urban water supply capacity have also increased, but the per capita water consumption has relatively decreased and the water quality is relatively low. It is concluded that adjusting the pattern of the urban landscape is effective for the improvement of the urban hydrologic effect.

Through correcting the Landsat-5 TM image of atmospheric effects and employing the calculation method of mono-window algorithm，the LST in Beijing was calculated by inversion. On such a basis，five vegetation parameters were computed respectively, i.e., Normalized Difference Vegetation Index (NDVI)，Ratio Vegetation Index (RVI)，Greenness Vegetation Index (GVI)，Modified Soil-Adjusted Vegetation Index (MSAVI) and vegetation fraction (fg). Combined with the spatial distribution of LST in Beijing，this paper compared and analyzed the relevance between LST and the five vegetation parameters. Quantitative analysis of vegetation effect on the Urban Heat Island (UHI) was also carried out. The results show that, of the five parameters of vegetation，fg has the strongest negative correlation with LST. The average urban LST is 1.6 K and 5.3 K higher than that of the suburban and outer suburban respectively.

 In this paper, MODIS data from 2002 to 2006 were processed, the terrestrial NPP in China were calculated through statistic analysis, and then the results were analyzed. The results show that, from 2002 to 2006, the average terrestrial NPP was 1.5 PgC. The five-year national terrestrial NPP composition analysis indicates that most of the national NPP concentrated in 0~300 gC/m2 , the percentage of NPP was about 70%~85%, and the annual NPP in most parts of China showed a downward trend, which was significantly affected by urban expansion, new industrial land and other human activities.

 This paper has explored the method and scheme for classifying different kinds of agricultural land from the high resolution remote sensing images by using multi-scale segmentation technology and rule-based image analysis approaches. Firstly, optimal segmentation scale was examined to construct a multi-scale segmentation level network according to the size of objects. Secondly, on the basis of spectrum, shape, texture and topology characteristics of images, several features of NDVI, shape indices, brightness, mean spectral value of red band, and ratio of near-infrared band, the standard deviation of near-infrared band and homogeneity were selected to classify objects into four agricultural land categories. The results show that these characteristics are effective in identifying agricultural land type and that the precision is higher than that of the traditional maximum likelihood classification.

With the remote sensing technology, geographic information systems and virtual reality technology, the authors established a virtual three-dimensional electronic sand scene. This design presents the features of mine exploitation and environment, topography, rivers, mountains and other surface information in the audio-visual way. the Electronic Sand Table can fully realize visualization of mine elements of information, which is likely to provide a platform and technical support for decision-maker.

With ASTER, ETM and SPOT remote-sensing images as the data sources, the authors used remote-sensing software to enhance the images so as to highlight the embankment information of the Ebinur Lake. By superimposition of the practical sampling points, satellite remote-sensing images and DEM data, the relic sites of lakeshore and terrace were determined. Through setting up the flooding model, the authors obtained the three-dimensional information of the Ebinur Lake terrace easily and accurately, formulated high-precision DEM data, created simulated three-dimensional landform model, and realized three-dimensional demonstration and three-dimensional flooding. The overall and partial characteristics of the landform can be observed from different angles, different points of view and different attitudes. With the help of the GIS system tool, the lake surface areas, volumes and flooding ranges of different periods can be accurately calculated, and the surface situation of the Ebinur Lake can be understood through inversion. All these results are of great significance in the study of the morphologic shape and evolution of the Ebinur Lake.

Firstly，an artificial light source image was simulated by digital elevation model DEM； then，an apparent 3D geological map was prepared by the data fusion of the geological map and the artificial light source image. This new type of geological map has not only the complete geological information of the geological map but also the intuitive features of the terrain， with some three-dimensional effect. At the same time, the digital apparent 3D geological map with geo-coding information can be used to determine the geographic coordinates of any position on the map. This map is potentially useful in such aspects of the field work as geological investigation，positioning and mapping，thus having a very high practical value.