The Study Of Change Detection Technology With Synthetic Aperture Radar Images

The Study of Change Detection Technologywith Synthetic Aperture Radar ImagesZhao Xiaojie (signal and information processing)Directed by Prof.Wang Hongqi and associate Prof.Chong JinsongImage change detection techniques aim to detect the changes in image series over time. Change detection techniques are widely used in environment monitoring, land use, the study of changes in vegetation, assessment of deforestation, damage assessment and other environment changes. The major advantage of synthetic aperture radar (SAR) is its all-weather capability that allows the acquisition of time series of imagery with exact acquisition dates under all climatic conditions, One of the major applications of SAR is change detection. With the development of modernization, urban develops rapidly. It is important to detect land use in urban and urban expansion. This paper detects urban expansion using digital change detection techniques and SAR images.This paper discusses change detection techniques that are widely used in remote sensing community. In order to identify what change occurred in image series this paper mainly research post-classification comparison method. One of the major aspect of image classification is feature selection. Get lots of sample images by way of field survey. For the particular application using those sample images as training samples selects a set of features which are fit for this application from twenty-six features. In the process of researching post-classification comparison method this paper improve K-nearest neighbor classifier and gain better detection result.For all the researches in this paper, the Radarsat images in 1998 and 2000 of Beijing and field survey data are used to validate. The experimental result demonstrates the wide application of SAR images to urban change detection.