Efficient analysis of satellite image denoising and resolution enhancement for improving classification accuracy

Abstract

Geospatial information gathered through different sensors and geographic objects is generally indistinct vague and uncertain This ambiguity turns out to be obvious due to the multi granular formation from the multi sensory satellite images that cause error accumulation at each stage Remote sensing and related techniques such as geographic information systems have a profound impact in real time applications Satellite images are often corrupted by noise in their acquisition and transmission process Removal of noise from the image by attenuating the high frequency components removes some important details as well In order to improve the visual appearance and retain the useful information in images this research concentrates on image denoising as the first initiative to remove additive noise while retaining as many important signal features as possible For denoising many researchers exploit the directional correlation in either spatial or frequency domain However the orientation estimation for directional correlation becomes inefficient and error prone in noised circumstances This work proposes a new Hybrid Directional Lifting technique for image denoising that involves pixel classification and orientation estimation A small amount of noise is added to improve the performance of the technique