Sathiyavani,VeltechMultitech Engineering College |
Pavithra S.,VeltechMultitech Engineering College |
Krishnalakshmi A.,Jaya Sakthi Engineering College
ARPN Journal of Engineering and Applied Sciences | Year: 2015
A set of high-resolution remote sensing images covering multiple spatial features, we propose a classification based on unsupervised technique including pixel-wise and sub-pixel-wise methods to detect possible built-up areas from remote sensing images. The motivation behind is that the frequently recurring appearance patterns or repeated textures corresponding to common objects of interest in the input image data set can help us distinguish built-up areas from other features. In our proposed method have two main steps first; extract a large set of corners from each input image by an improved Harris corner detector. In the second step we incorporate the extracted corners into a likelihood function to locate candidate regions in input image. Experimental results demonstrated that the proposed approach have got accurate estimation compare to the existing algorithms in terms of detection accuracy. © 2006-2015 Asian Research Publishing Network (ARPN).