The Cloud Detection, Land Cover Patterns And Land Surface Temperature Retrieved From Remotely Sensed Data Of HJ-1B

With the successful ascension and imagery of HJ-1 satellite, remote sensing development of China is prompted. For China's own space satellites, application has become the focus in recent research. Own to the factors of shoot time, atmosphere and climatic conditions, most satellite images taken over study regions are covered by cloud pixels and the usage of images is reduced. In order to avoid the influence which cloud pixels give to the surface parameters retrieval, the application of HJ-1B satellite data is illuminated from cloud detection and real surface temperature reversion in this paper.Proper separation of cloud and clear sky is not only the pretreatment of atmosphere and surface parameters retrieval, but also the basis of numerous researches and other product development. The accuracy of Cloud identification products directly affects the results of other parameters retrieval. It thence appears that the cloud recognition is the indispensable product of HJ satellite's line of products. In terms of the visible-near infrared, visible light threshold form HJ-1B data is detected, the detection results is ideal. The final images which are wiped off cloud meet the demands of the following land surface parameter retrieval.Surface temperature, vegetation coverage, surface aledo and other surface biophysical parameters are not only important factors to material and energy cycle, climate change, energy balances in the world but also the hotspots of recent quantitative remote sensing study. With China's economy and the rapid development of urbanization, the influence which land-cover model of the surface give to directional distribution of the thermal environment in land surface is more and more significant. Because of the complexity of the underlying surface itself, regardless of the surface parameter retrieval or the relationship between land cover patterns and urban heat land effect is still difficult. Therefore, on the basis of previous studies, selecting a quantitative extraction which is suitable to HJ-1B satellite itself to extract land cover parameter and surface temperature, then from the same class at different levels, analyzing of influence which the various land cover parameters give to directional distribution of the thermal environment in land surface, which give evidence to research the rule of directional distribution of the thermal environment. It is important to retrieve surface parameters quantitatively by new data source.Fuzhou is selected as the study area. The HJ-1B satellite data in Feb.23,2010 is taken as data source .Jim?nez-Munozoz & and Sobrino algorithm are used to realize the true surface temperature retrieval. In terms of surface emissivity estimation method, IGBP classification is used in Land classification. According to different radiance based on different land types, radiance in study regions is estimated .Thus the actual temperature of the surface is retrieved. NDVI in the study area are extracted by spectral characteristics of red band and near infrared. Secondly, NDVI binary model is used to extract the vegetation cover (FVC) and urban impervious surface information in study areas.Taking advantage of surface parameters obtained above, the relationship between surface temperature and other surface parameters (vegetation coverage, impervious surface density, NDVI) are analyzed quantitatively from various angles. The results show that it is not appropriate to represent the relationship between surface temperature and other surface parameters though a simple linear model. Comparatively speaking, for different levels of the same type of ground surface, the effect which the parameter give to analyze directional distribution of the surface temperature is positive. Using stable impervious surface information combined with the NDVI is not only proper to retrieve directional distribution of urban thermal environment, but also provide a method for its research.