Improvement in Stitched Image using Image Fusion

Abstract

It is often said that one picture is worth more than ten thousand words. Since the last few decades, image stitching in real time application has been a challenging field for image processing. In this digital era, images are an integral part of human life and play a big role. Image stitching is a process of merging two images in order to create a panoramic image. To perform the whole process, some steps should be followed such as image acquisition, feature extraction, image registration and blending. This paper leads to the study of the techniques of image stitching and how to improve the stitched image quality using image fusion techniques. The quality of image can be measure some parameters like PSNR, NAE, MI, MSE and SD etc. Image stitching is a traditional image process to create seamless panorama which shows new perspective of the world around us. Some applications of this technique are satellite imaging metrological and environment monitoring, 3D rebuilding objects, digital objects video stitching, and 3D image stitching [6].
Keywords: feature extraction, image registration, warping, blending and image fusion.