Coarse Classification of Terrain Image Information by Using Sobel Edge Detection Technique

摘要

A new approach for the classification of man-made features and natural features in a terrain using a single image is presented. Prominent features that distinguish man-made objects from natural features are identified. Techniques are developed to detect image features and based on the features extracted and use decision based classification to identify regions in an image possessing man-made features. These techniques include feature extraction, edge detection, and classification. Some of these methods are novel and others present unique properties and advantages compared to previous related works. Experimental results are presented using real images, the result clearly identify the regions with man-made features. This technique for terrain classification is developed for air vehicles navigating in a terrain this technique would assist the air vehicles in efficiently searchingavigating a terrain.