Analysis On The Varying Wetland In Binhai New District Based On The Remote Senseing Information

Wetland is a kind of special natural complex, including natural or artificial swamplands, whether perm anent or temporary, turf land or water areas possessing a depth of water less than 6 meters at tide ebb. Wetland is one of the eco-systems with most rich biological diverse on the earth, which has enormous function of ecological service. They have a considerable potential influence on the humanity life and social development.Taking Binhai new district in Tianjin city as the study area and using Landsat TM data of 1995 and 2007 as the data source, this paper analysis the change of wetland. In order to identify the type of the wetland, the remote sense images will be processed by combining the software of ERDAS and ArcGIS, the steps including pretreatment, classification, counting area, translating raster data to vector data, topology editor and graphic, and at last formed a graphic of wetland type distribute in Binhai new district of Tianjin city. In the paper detailed the traditional supervised classification methods and unsupervised classification methods of the theory and algorithms. Unsupervised classification algorithms cluster data according to several user-defined statistical parameters in an iterative fashion until either some percentage of pixels remain unchanged or a maximum number of iterations have been performed. This method of classification is most useful when no previous knowledge or ground truth data of an area is available. Supervised classification algorithms rely on user-defined training regions that represent pure samples of a particular class. With the various classifications are used in remote sensing images of Tianjin area. The article has analyzed the difference and relation of two methods from different aspects such as the principle, the course and ways of classification, The results showed supervised classification methods in classification accuracy higher than that of unsupervised classification, but supervised classification can be classified as an auxiliary. The results show that based on the wetland in Binhai new district, compared with the traditional methods of classified the type of wetland, combining the Remote sensing technology and GIS has many advantages, because it’s principle is simple and the graphic is completely and understandable. Meanwhile, the results of classification and data analysis showed that the wetland in Binhai new district is increasing through the past 12 years, the artificial wetland is increasing, the natural wetland is decreasing and breaking up, and as well analysis the reason of changing and the managing measures.