Study On Moderate-High Spatial Resolution Remote Sensing Image Segmentation And Information Extraction

Remote sensing technology has obtained great achievements in resent years; one of the most outstanding examples is the emergence of high spatial resolution remote sensing. High resolution remote sensing image has widely used in many fields such as mapping, agriculture, forestry, urban planning, territorial resources management, geological survey, mineral exploration, military affairs etc. it pumped new life into the remote sensing application and research. The total quantity of data will increase exponentially with the increase of the spatial resolution of remote sensing image; the traditional visual interpretation method does not satisfy the customer. So intelligent interpretation based on computer is the only way to solve the problem.In recent years, numerous classification methods were described in the literature and they can be classified into two large classes: traditional pixel-based classification and object-oriented image analysis method. Traditional pixel -based classification techniques is applicable to medium and low spatial resolution remote sensing image interpretation and object-oriented image analysis method is applicable to for medium and high spatial resolution. Object oriented image analysis has been utilized many fields successfully such as forestry resource management, agriculture management, land cover, land use, mapping etc. There's no object oriented image analysis system for medium and high spatial resolution remote sensing image at present in China, it restricts popularization and application of remote sensing image. Thus it is important and urgent to study and develop new methods to extract thematic information from high resolution remote sensing images. In this paper, we explored the intelligent interpretation method of high resolution remote sensing images, and the main results are following:(1) In this paper, we present a new frame of object oriented remote sensing image intelligent interpretation (Very High Resolution Image Intelligent Interpretation, VHRIII).it synthesizes image processing system (pixel-based), geographical information system (vector-based) and data mining (intelligent computing) technology's strong points to interpret image from pixels to segments and to thematic information.(2) To summarize eight basic requirements of medium and high resolution remote...