This paper describes the influence of aerosols on global climate change and air pollution situation, points out the importance of using high-resolution satellite images for urban aerosols inversion, elaborates the research status of the satellite remote sensing retrieval of aerosol, and briefly introduces the principle of satellite remote sensing retrieval of aerosol, and the single channel and multi -channel method, the contrast method based on differences in the spatial structure, the inversion method based on multi-angle data, the inversion method based on polarization data and the inversion method based on laser radar satellite data, which constitute five kinds of satellite remote sensing inversion method widely used nowadays. According to the inversion method and the difficulties of the urban aerosol inversion, the inversion of aerosol in urban areas is analyzed and summarized, with a detailed discussion on the shortcomings of current inversion methods based on high-resolution images of urban areas and a forecast of breakthrough points and the solutions of the existing problems.

Extracting road edge from remote sensing image can extremely simplify the land survey workload for planning traffic networks. Because of the direction and size limitations of the template, the edge detection result derived by traditional algorithms shows high rate of false positive points and discontinuity, which is the reason why traditional edge detection algorithms can seldom achieve an ideal result in detecting continuous and integral road edge. To deal with this problem, this study proposes an improved Sobel operator which is based on an 8 directional 5×5 template. The optimal settings of each direction in the template are derived by the Pascal's triangle theory. The improved operator not only achieves a better performance of edge detection in different directions but reduces the false positive point effectively as well. In comparison with other operators, the improved Sobel operator proposed by this study has better integrity and continuity in road edge extraction results, especially in road curve detection.

Support vector machine (SVM) algorithm has been widely used for remote sensing image classification. For high spatial resolution image classification, traditional SVM algorithm usually leads to low efficiency due to large quantities of high dimensional sample data. This paper presents a simple improved SVM algorithm with the purpose of improving both efficiency and accuracy of classification models. The algorithm first uses PCA to reduce the dimension of sample features. The grid-based method is used to search for optimal parameters for SVM classification of PCA-based samples. Then new range around the PCA-optimal parameters is set up and used for optimal parameter search based on the original sample data. Finally, SVM with the optimal parameters is used to train the original sample data and classify the image. The new algorithm was evaluated by two classification experiments based on WorldView2 images including urban land cover land use classification and urban tree classification. Compared with the traditional SVM and SVM merely based on PCA data, the results show that the improved SVM algorithm could quickly and efficiently find the optimum parameters of the SVM classifier and achieves higher classification accuracy.

In order to analyze spatial distributions of socioeconomic activities at pixel scale, the authors used Landsat8 and NPP-VIIRS night-time light images as data sources and produced spatialization maps of primary industry GDP and the secondary, tertiary industry GDP in Beijing. The land use map of Beijing for the spatialization was produced from Landsat8 image with CART decision-tree algorithm. According to the correlation results between the primary industry GDP and areas of land use, a linear regression model was built based on the primary industry GDP and areas of plough. By analyzing the correlation relationships between five light indexes and the secondary, tertiary industry GDP, compounded night light index (CNLI) and the secondary, tertiary industry GDP presented apparent power function's correlation relationship. Using linear corrections and summation of two types of pixel level's GDP density maps produced both modes listed above, and a total GDP density map was generated with the resolution of five hundred meters in Beijing. The results of GDP relative errors show that the primary industry GDP and the secondary, tertiary industry GDP were 0.86%, 0.61% and 1.37% respectively. This suggests that this approach of pixel level's GDP spatialization can be applied to estimate Beijing's GDP and reflect characteristics of its economic distribution.

The inversion of geometry and motion parameters of the seismogenic fault based on elastic dislocation model and constrained by coseismic deformation data obtained by DInSAR has great significance for understanding the activity characteristics of the fault and assessing the risk of potential disasters. The existing inversion methods of fault slip parameters are mostly based on the uniform discretization with rectangular dislocation units or artificially setting discretizaiton, which will lead to "pseudo-slip" or the phenomenon that the slip distribution is too smooth to reflect the slip details on the fault plane. Therefore, the automated fault model discetization method is introduced in this paper, which takes into account the principle of the minimized residuals and the optimal smoothing scales of the model, so the reliable solution can be obtained under the effective constraints. The inversion of geometry and slip parameters of fault in Bam earthquake is taken as an example. The experimental results show that using the automated fault model discretization method to invert the motion parameters of the single fault can generate reliable results.

It is difficult to obtain labels of samples for hyperspectral data. Few labeled samples usually lead to low classification accuracy. In view of this situation, an improved spatial and spectral constraint graph-based semi-supervised classification algorithm (SS-GSSC) is proposed. First of all, Euclidean distance combined with radial basis function (RBF) is used to construct the spatial similarity edge weight; Spectral correlation angle (SCA) is used to calculate spectral similarity weights; Then, the two kinds of weights are combined to the form of product to restrict the similarity measurement; Finally, the label propagation algorithm is used to predict the test data labels so as to obtain the classification results. Classification experiments on Indian Pines image and DC Sub image show that, compared with the previous classification algorithm, the algorithm designed by the authors can better eliminate the phenomenon of the existence of the same category map spot included in other categories of scattered points, and can achieve higher classification accuracy under the condition of less label points (25 per class).

In order to complete the task of real-time object tracking, this paper proposes a new local texture description operator, which is called center symmetric-local similarity number(CS-LSN). The operator analyzed the center pixel and its eight neighborhood pixels, on the basis of the local similarity number (LSN); as the texture operator fails to distinguish the same local saliency of different texture structures, the authors added the local gradient information using the center pixel as the symmetric point, extracted the key pixels which correspond only to the five major patterns of the CS-LSN in the target candidate region, effectively restrained the influence of the background pixels and reduced the computation in the target representation model, and then represented target by CS-LSN texture feature and the chromaticity in the true target pixel, which was embedded into the mean shift(MS) tracking framework. The experimental results show that the proposed method can continuously track the target when the background and the color are similar under the condition of changing illumination and in the occlusion cases. The processing speed can reach 25 frames or so per second when the object is 29 pixel×25 pixel, so it can satisfy the demand of real-time application.

Generally, standard scenes with rational function model (RFM)are provided to users as the common form of high resolution satellite images. Since RFM parameters lack clear physical significance, the geometric constraint relationship of standard scenes in the same stripe could hardly be established even though both attitude and ephemeris errors are stable in the same stripe. On the basis of the difference between stripe scenes and standard scenes, as well as in consideration of the offset effect and integral time differences of standard scenes, a stripe constraint method based on the image-space polynomial model is proposed in this paper, which could be used to constrain the discontinuous standard scenes in the same stripe and realize the stripe-constrained block adjustment of high-resolution images. Triplet stereos covering seven standard scenes of ZY-3 and two adjacent stripes of IRS-P5 were used for the experiment. The results show that, with the proposed method in this paper, less ground control points are required for block adjustment with the same stripe constraint, and the higher accuracy is promised as well.

Corn field remote sensing images have a mass of end member spectral variability among marginal land area. When traditional method is used for corn block segmentation, it will produce a number of small corn plot areas at the edge and result in statistical errors of the planting area. According to the distribution characteristics of large corn planting area, an near-rectangle guided segmentation method for remote sensing images in corn field areas is proposed. First, the SUSAN(smallest univale segment assimilating nucleus)operator is used for edge detection from GF-1 fusion images. Then, according to the relationship between closed area and external near rectangular, the near rectangle-guided correlation function is built. At last, the near-rectangle guided threshold function is introduced into the graph-based segmentation algorithm to implement the field parcel segmentation of a specific shape. The results were compared with the graph-based segmentation algorithm, the watershed algorithm and the artificial interpretation sample. It is shown that the method proposed in this paper is effective in distinguishing different features, and the negative impact resulting from the endmember spectral variability can be reduced. The segmentation results are more in line with the actual characteristics of corn distribution, conforming with the actual statistics of the corn field area.

With the object-oriented classification method, the spatial distribution of Phragmites australis was obtained based on multi-temporal Landsat8 OLI data covering Shuangtai Estuary National Nature Reserve. By analyzing the sensitivity between different vegetation indexes and aboveground biomass (AGB) of Phragmites australis, the retrieval model of AGB for Phragmites australis was developed. Furthermore, the spatial pattern of AGB for Phragmites australis was observed. The results showed that the area of Phragmites australis was 4.39×10⁴ hm², accounting for 32.96% of the study area of the Shuangtai Estuary National Nature Reserve. Selecting NDVI as the variable for the power function, the authors formulated the optimal model for estimating AGB of Phragmites australis with an estimation accuracy of 79%. Average AGB of Phragmites australis was 4785.5 g/m² and total AGB was 2.06×10⁶ t. High values of AGB were observed in the northeast part, while low AGB values in the southwest. The results obtained in this study would provide data to support wetland ecosystem management and scientific guidance for Shuangtai Estuary National Nature Reserve.

How to retrieve the image data quickly and accurately from large remote sensing image database is a critical problem. Using integrated region matching (IRM) algorithm as image similarity measurement standard, this paper proposes a retrieval approach to retrieve urban area images from remote sensing image database according to Average High Frequency Signal Strength (AHFSS) values of the stored images, which are used to sort the retrieved images in descending order. The proposed approach firstly utilizes IRM algorithm to measure the similarity measurement of the stored images. Then, the proposed approach resorts the retrieved images in descending order according to AHFSS values of the stored images to obtain the final retrieval result containing high level semantic feature "urban areas". Experimental results show that the proposed approach increases the retrieval precision by 27% and has reasonable retrieval efficiency to meet users' requirements.

Clouds and their shadows in the remote sensing images are the key factors that influence the application of the data in many fields. Several methods, such as constructing cloud mask, replacement of the pixels, linear mixture spectral analysis, and principal component analysis, have been proposed in the past decades to solve this problem. In this research, based on the analysis of spectral curve, the authors utilized Decision Tree(DT)classifier and Random Forest (RF)classifier to obtain the cloud-shadow-water mask. There was little difference between the spectral curve of shadow and water due to the mixture of shadow and other surface materials such as vegetation and impervious surface. In this case, the DT classifier could not effectively distinguish shadow and water because the decision rule and threshold were determined by analyzing the spectral curves of different samples. RF classifier was based on the ensemble of the results derived from multiple decision tree classifiers, which was more robust than one decision tree classifier. In this study, when there were only a few training samples, results that were more accurate were derived from RF classifier compared with the results from DT classifier. When the size of training samples lay in the range of 250 and 400, no significant difference was found between the results derived from these two algorithms. This indicates that RF classifier could be used to deduce the cloud-cloud shadow-water mask successfully.

In order to solve the problem of remote sensing image segmentation under vector map constraints, this paper proposes a marker-based watershed method for remote sensing image segmentation under vector map constraints. Firstly, the constrained vector map is mapped to the edge of the image. The markers are extracted in the image with edge. Secondly, the pseudo marked areas are eliminated by area constraint. The flood for unmarked pixels is implemented by a priority queue arrow. Lastly, the edge pixels in the image are labeled by a similarity measurement. The label image is used for vectorization to get the segmentation result. The experimental result on the QuickBird image shows that the proposed method can realize image segmentation under vector map constraints. Compared with multiresolution segmentation method in eCognition software, the proposed method is more efficient when the remote sensing image is large.

Certain spectral characteristics have a direct impact on accuracy and efficiency of identifying the wetland vegetation. In this paper, the authors mapped wetland vegetation with 3 m spatial resolution for HyMap image data from Sherman Island of California's Sacramento-San Joaquin delta. The first-derivative spectral features and spectral absorption features of different species were analyzed by the method of stepwise discriminate analysis, and the spectral characteristic parameters with better classification accuracy were screened to identify species of wetland vegetation in C4.5 decision tree classifier. The results showed that the absorption features of four plants have larger differences than first-derivative spectral features. The results also showed that C4.5 decision tree classifier in combination with the first-derivative spectral characteristics and spectral absorption characteristics could be effective in distinguishing wetland vegetation and allowing for species-level detection.

The precipitation infiltration recharge condition has theoretical and practical significance in the study of groundwater resources. In this study, with the application of remote sensing technology and through constructing evaluation system and model based on expert experience and knowledge, the spatial differentiation of Beijing precipitation infiltration recharge conditions are evaluated at three scales, including grid scale (30 m×30 m), Karst water system bedrock region and its carbonate rock region. The results suggest that carbonate rock areas are conducive to the precipitation infiltration recharge in the whole study area. The precipitation infiltration recharge conditions are in worsening order of Fangshan, Mentougou, Yanqing, Pinggu and Huairou in the five Karst water systems bedrock areas. The precipitation infiltration recharge conditions are in worsening order of Huairou, Mentougou, Fangshan, Yanqing and Pinggu in the five carbonate rock areas, and the lithology and structure seem to be the key factors affecting the conditions of rainfall infiltration recharge.

In order to improve the automatic level of information extraction, this paper proposes an automated rural residential area detecting procedure from UAV remote sensing imagery based on object-oriented segmentation. Firstly, the bilateral filtering is performed on UAV remote sensing image to eliminate the image noise. Secondly, the filtered image is segmented based on mean shift clustering algorithm. Thirdly, the rural residential area is detected from the segmented image according to the features of rural residential objects. This method was tested using UAV remote sensing images, and the result showed that the rural residential area could be extracted effectively, with an average accuracy about 92%.

The Dong Ujimqin Banner area is located at the junction between the metallogenic belt from Erlian to Dong Ujimqin Banner and Da Khingan Mountains metallogenic belt, with complicated geological structures and mineralization conditions. The results of structural interpretation by integrated regional geological, geophysical, geochemical and remote sensing data show that there are many NW-trending faults in the area, apart from NE-NEE-trending structures. The most noticeable one is the NW-trending deep fault line along the section from Dong Ujimqin Banner to Wongtu. Discontinuous linear features are interpreted by remote sensing along this major fault line, which are the gradient variation boundaries of aeromagnetic field, gravity field and geochemical field gradient in this fault zone. This major fault obviously played a significant role in the process of Mesozoic sedimentation and mineralization in this area:to the east of fault line, there existed strong activity of Mesozoic magmatism, favorable conditions for mineralization, and potential of lead-zinc-silver, tungsten and iron; to the west of the fault line, Mesozoic magmatic activity was obviously weakened, and prospecting should be mainly targeted for copper, molybdenum and chromite. In prospecting, attention should be paid to the NE and NW trending structural intersection which is closely related to geochemical anomaly areas.

Under the impact of formation conditions and natural environment, moraine-dammed lake outbursts take place frequently and then give rise to floods and mudslides, which menaces people's production, life, survival and development and becomes one of the great geohazard hidden dangers in the Himalayas. Using QuickBird and ETM data of remote sensing satellite and combining the investigations in field work, the authors studied the glacier lake outburst potential. The research results show that the areal distribution of glacier lakes presents a trend of decrease gradually from SE to NW in the Himalayas. Most of the glacier lakes are distributed in Shannan, Shigatse, Nyingchi and Ngari. The region of maximal distribution density is Lho-brag area of Shannan. It can be found that there are nineteen glacier lakes, all of which are moraine-dammed lakes that have outburst hazard potential. Among the nineteen glacier lakes, there are 13 glacier lakes having great outburst hazard potential and 6 glacier lakes having secondary great outburst hazard potential. It is also indicated that there are several trigger factors impacting instability of moraine-dammed lakes, whose foremost reasons are surge triggered by avalanche, bedrock collapse, and rock avalanche around the moraine-dammed lakes.

In order to study the atrophy of the Dongting Lake in the past 100 years, the authors systematically monitored the atrophy of the Dongting Lake based on historical topographic map data, such as the Dongting Lake map in the late Qing Dynasty (AD 1896) which is treasured by the Library of Hunan Province nowadays, aviation photos since 1954, satellite images which were captured by the American KH-8 military satellite in 1967 and other aerospace remote sensing data. The results show that in the past 118 years from 1896 to 2014, the Dongting Lake area has shrunk by 2466.42 km², from 5146.71 km² to 2680.29 km². Now the Dongting Lake area is 47.92% of the lake basin area in 1896. A comparison with the area variation data of the Dongting Lake published by the Water Conservancy Department shows the data obtained by two sources are generally very close, with some differences in some specific areas. After the severe flooding disaster of the Yangtze River in 1998, the State Council explicitly put forward the lake-management policy of "Farmland Conversion to Lake Wetland" so as to stabilize the area of the Dongting Lake. The Dongting Lake is mainly enclosed by artificial levee whose location is irrelevant with sources of water and sands; thus, the operation of "the Three Gorges" since 2003 has had no influence on the area of the Dongting Lake.

After Wenchuan earthquake, the aftershocks happened frequently. Coupled with the rapid development of cities, they are likely to cause ground subsidence. So monitoring the surface of Chengdu Plain and obtaining the deformation information could provide scientific basis for the relevant decisions. In this paper, based on ENVISAT ASAR data, the authors monitored the ground subsidence of Chengdu Plain from 2008 to 2010 using SBAS-InSAR technology. The results show that the average surface deformations were between -8 to 14 mm in major cities in the Chengdu Plain during the monitoring period, the deformation is not prominent, and the western part of the plain showed a uplift trend caused by earthquake. The subsidence area in the north of the Chengdu City and south to the Deyang City was up to -22 mm with the expansion of the subsidence area. The monitoring results were validated by measured data and the accuracy is 2.9 mm. The Chengdu Plain has no regional tectonic setting of subsidence and has abundant groundwater resources, so the natural cause of subsidence is not obvious; the city building activity might be the major cause of subsidence.

The atmospheric correction of remote sensing image is one of the difficulties in quantitative remote sensing research. In this paper, aimed at the suspended sediment concentration (SSC) levels retrieval in the Caofeidian offshore area, the authors performed a comparative test on atmospheric correction of MODIS image of the study area by 6S and FLAASH models, and then evaluated the corrected image quality and the correction effects of target information(normalized water index, NDWI). The results show that these two models could reduce the atmospheric effect on remote sensing information of water body to some extent. By comparison, the corrected image quality by 6S is better than that by FLAASH and could more truly reflect the target information; therefore, 6S model can better perform atmospheric correction of remote sensing images with a high precision in coastal waters. Subsequently, the MODIS image after atmospheric correction by 6S was applied to invert the SSC in the study area, and the inversion results show that the average relative error(MRE) and the root-mean-square error(RMSE)are 24.79% and 4.32 mg/L, respectively. The results can provide a basis for the selection of atmospheric correction methods in caseⅡwaters, thereby laying a foundation for the study of sediment transport law as well as evaluation of water quality and water environment.

Landscape limited forms of land use and land use type influence the formation and evolution of the landscape. In view of that relationship, this paper proposes an object-oriented land use/land cover classification method. Taking Anshun City with diverse karst landform types as an example and using 2010 TM images as the basic data source, the authors divided the test data according to different scale response characteristics of different karst landforms combinations and with multi-scale segmentation technology. Using DEM data and the derivative features such as slope, relief amplitude, and surface incision and establishing membership functions determining the threshold, the authors extracted the karst landform information. Then, on the basis of landform types and according to the landuse characteristics of different landform types, the authors divided landuse types by using nearest neighbor analysis. Overall classification precision reached 87.35%. Finally, the distribution characteristics of landform and land cover were discussed on the basis of the classification results. The results show that the landform types in the study area are mainly peak cluster valley and peak cluster depression, with a few karst pimple plains and hoodoo basins. Land cover types are mainly shrubs and dry land, which account for 25.58% and 20.74% of total area respectively.

Vegetation is a natural "link" which links soil, air and water and an "indicator" in global climate change research. Using normalized difference vegetation index (NDVI) time-series analyses, we can provide better support for the relevant researches and decision-making. Using MODIS NDVI data binding with BFAST (breaks for additive seasonal and trend) method, the authors implemented monitoring vegetation dynamics in the Laohahe River Basin and the surrounding areas, and identified its NDVI time-series abrupt change points occurring in time. The meteorological data and the quality of the data itself were also used as an influence factor analysis of the main reason for the breakpoints. It is found that precipitation, relative humidity, temperature, sunshine and water evaporation are positively correlated with NDVI trends, while wind speed is less correlated with NDVI trends. What's more, the precipitation and sunshine hour impact on NDVI change has a certain lag.

The division of protected area is one of the most important tasks in planning of geoparks and also the foundation for future construction; nevertheless, some relevant problems exist such as vague standards of division. This study was based on the investigation of geosites and the framework of Visitor Experience and Resource Protection (VERP). Evaluation scale was composed of 7 indicators that were selected and weighted by AHP. The GIS was used as a tool to quantitatively evaluate the protected index of the geopark. The Shennongjia Global Geopark of Hubei Province was selected as the study area, and the results showed that the protected index was from 8.4 to 124.4, which indicated the aggregation and spatial difference. The final protected areas were divided on the basis of the preliminary map, with the consideration of the local economic demands and land using planning. The achievement shows that total area of protected areas in the Shennongjia Geopark is 174.749 km², which is the basis for the planning of division of functional zones and construction of geoparks in the future.

Cultivated land is an essential resource and condition for the survival of human beings. Evaluation of cultivated land suitability is the basis of the land reasonable utilization and can provide the scientific basis for utilization and management of cultivated land. Taking Tianjin Binhai New Area as the study object and based on the theory of cultivated land suitability, the authors confirmed the evaluation unit with combinations of GIS and RS. 5 evaluation factors were selected including soil geology and irrigation to build up the evaluation system with combinations of specific situations of Binhai Xin Area and the availability of data. Then the results of cultivated land suitability were divided into four levels based on GIS and weighted index summation model. The result showed that each level accounted for 5.65%, 27.25%, 44.42%, 22.68% of the total evaluation unit area respectively from level Ⅰ to level Ⅳ. The whole quality of cultivated land is not very good. The highly suitable regions are mainly distributed in Taiping Town, Xiaowangzhuang Town, Zhongtang Town, north of Airport Economic Zone, Hujiayuan Street, Xincheng Town, Chadian Town, Datian Town, Hexin Street and Yangjiabo Town. The authors put forward some advices for each level land of evaluation result based on the results of each index evaluation, which are very important for improving cultivated land quality of the whole Binhai New Area and using land reasonably and availably.

In this paper, an attempt has been made to estimate rainfall over the Xinjiang region by using the Tropical Rainfall Measuring Mission(TRMM)-3 B43 product for the period of 2003-2008 and rain gauge data from 105 weather stations at different time intervals. The results show that the TRMM rainfall product is generally consistent with the measured one on the yearly-scale; Goodness of fit of seasonal precipitation between TRMM estimates and observed data is high, with the goodness of fit being insignificantly different between the seasons and the correlation coefficients being above 0.7; For the month scale, TRMM product has statistically significant correlation(r=0.75)with rain gauge data; For individual stations, at most stations the correlation coefficients are reasonably high and the estimation biases are small. An analysis of the error obtained from each site showed that most of them (60%)are within 30%, while the overall fitting correlation coefficient reached 0.81, suggesting that there is good consistency between TRMM precipitation data and observed precipitation. Overall, TRMM precipitation products have certain deviation in time and space, but the deviations have certain regularity. Calibration is indispensable processing in application.

Spectral characteristics analysis is the basis of spectral feature classification and matching in hyperspectral image processing. In this paper, the authors selected five kinds of subalpine forest vegetation to measure their field spectra in the upper reaches of the Minjiang River, which include gramineae mottled bamboo, herbaceous fern,pilea notate, arbor china fir and shrubs palm. Through constructing the high spectral similarity measure index, five measuring methods, i.e., Euclidean distance(ED), spectral angle mapper(SAM), spectral information divergence(SID), spectral information divergence-spectral angle mapper(SID(TAN))and spectral distance based on Douglas-Peucker(SDDP), were used to analyze the relative capability for recognizing forest vegetation on the plateau. According to the results obtained, the spectral feature difference in the five kinds of forest vegetation mainly lies in peaks and troughs in the spectral curves; pilea notate has the highest relative spectral discriminatory probability in ED similarity measurement; mottled bamboo and fern have the highest relative spectral discriminatory probability in SID and SID(TAN); China fir has the highest relative spectral discriminatory probability in SDDP. SAM, SDDP, ED, SID(TAN)and SID of the relative spectral discriminatory entropy are 1.51, 1.59, 1.61, 2.16 and 2.18 respectively. The research results showed that the means reduced the amount of calculation for doing the similarity measurement which extracted the spectral feature vectors with the SFT, DPBSR and DABSR, DPSR. In order to ensure the condition of similar recognition capability, the means can greatly improve the retrieval efficiency of the program, and hence they are the fast and efficient hyperspectral feature matching and retrieval methods.

A lake water surface extraction method is proposed in this paper, which is combined with lake water surface boundary buffer analyzing and lake by lake segmentation threshold determining, with MOD09Q1 as the data source. The method was successfully used to extract the 8-day lake water surface dataset of the Tibetan Plateau, during the period of 2000-2012. The accuracy analysis results indicated that, compared with water surface data of the 133 sampling lakes that extracted from the 30 m Landsat TM images, the total precision of the proposed method is 93.98%; the extracted daily, monthly, and annual lake water surface areas of the Nam Co, the Qinghai Lake, and the Siling Co are well consistent with the remote sensing monitoring results generated by other researchers; and the differences of the lake quantities and acreage statistical results between the second national lake survey and this study are only 11 and 4.74%, respectively. This study provides a reference method for large scale and long time series single objects (such as water, vegetation, and urban) distribution area extraction. Furthermore, it provides reliable dataset for the lake change research of the Tibetan Plateau in the past 10 years.

Using object-oriented classification technology, the authors achieved the automatic extraction of the Yellow River Estuary coastline in 1973, 1977, 1984, 1991, 2000, 2010 and 2014, and completed the remote sensing dynamic monitoring and the change analysis for the Yellow River Estuary coastline from 1973 to 2014. The results show that the length of shoreline in the Yellow River Estuary was increased by about 298.8 km² in the past 42 years overall. The coastline change process had an obvious spatio-temporal characteristics. The change rate was different at different periods and different locations. According to the characteristics of the coastline expanion and retreat, it could be divided into 3 categories:fast silt coastline, strong erosion coastline and erosion regressive coastline. For the driving force, the location of Yellow River port and sea erosion seem to have been the main factors. The coastline was mainly influenced by the flow path of the Yellow River. And, the recent human activity seems to be the third driving force. Human activities, such as port building, saltpan construction and exploitation of coastal resources, have exacerbated the complexity of coastline change.

Based on an analysis of the mine remote sensing monitoring data processing, the authors designed and developed a resulting data compilation system for mine remote sensing monitoring. The system can achieve the data preparation, information extraction, production data compilation, production data quality inspection, production data storage and other functions. Using the Plug-In GIS framework technology to achieve a sustained development, integration, testing, and releasing of functional modules, the authors solved the problem of frequent functional requirements changes due to frequent changes of storing technical requirements and improved software maintainability and scalability. Using the metadata model design based on XSD, the authors made system updating and resulting data checking easy. The practical application shows that the system provides effective support for the production data compilation, improves the production data quality and work efficiency, strengthens the operability of the production quality inspection, and reduces the working intensity and workload.