PROJECT TITLE :

Performance Evaluation Of Different Inpainting Algorithms For Remotely Sensed Images - 2017

ABSTRACT:

Image inpainting refers to a method in which missing areas of a picture are stuffed is such a approach that it appearance plausible to the human eye by retrieving data from the encircling pixels. Degradation in remote sensing images is typically caused by dead pixels, noise, clouds, sensor downside or communication system downside. The aim of this paper is to guage the performance of various inpainting algorithms for remote sensing pictures. Satellite Images are tested on totally different inpainting algorithms and efficiency of every algorithm is evaluated on the premise of processing time, Root mean square error and peak signal to noise ratio.