Manmade structure segmentation based on morphological operation from high resolution PAN image

Abstract

Manmade structures are important objects or targets for both civilian and defence applications. Isolation of manmade structures from background are very critical task due to noise, non-uniform intensity distribution and class overlapping of remote sensing satellite image. Image segmentation is a classic subject in the field of image processing and pattern recognition and also is a hotspot and focus of image processing techniques. The paper presents only the technique of manmade structures isolation from the background by morphological enhancement technique and local threshold method. The results demonstrate that the proposed algorithm is segmenting the manmade structures accurately and efficiently.