A computational approach and comparative analysis for the satellite images based structural recognition using intelligent techniques of edge detection

Abstract

Over the past two decades the resolution level of satellite images increased drastically. A few years ago, satellite image with a spatial resolution of 10m was the ultimate in high resolution satellite imaging. Later on the resolution level reduced to 1m and even 40-60cm. Thus the High Resolution Satellite Imagery (HRSI) has relevant impact for cartographic applications. It proved to be a suitable alternative to aerial photogrammetric data to provide a new data source for structural feature extraction. Roads, buildings and bridges are the main man-made structures that are to be extracted from satellite images. The detection of shadows and clouds to get a shadow/cloud free image from high resolution satellite images greatly facilitates the newlineclassification and extraction of structural features However, the real possibility of using high resolution images for feature extraction depends on several factors: sensor characteristics (geometric and radiometric resolution), types of products commercialized by the companies managing the satellites, cost and time to obtain these products, cost of commercial software for processing such products. Also commercial vendors of satellite imagery provide images in both spatial and spectral domains (Panchromatic and Multispectral images). Utilizing images from both domains efficiently in different feature extraction problems is a challenging task.A number of methods have been developed for feature extraction from high resolution satellite imagery available in spatial and spectral domains. Most existing methods are based on knowledge base, which is not derived automatically. Therefore this work introduces Adaptive Fuzzy Edge newline newline