Image Fusion Based on Transform Domain Techniques

Abstract

Nowadays, Images have become the main source of information transfer. Every single image conveys more or less information regarding an object. Images of an object taken in various orientations convey distinct information. Here, we choose image fusion as method for conveying the exact information in a single image rather using multiple images. Image fusion is emerging as a vital technology in many applications. Image fusion is a method to use image as redundant or complementary source to extract information from them with higher accuracy or reliability. Image fusion methods can be broadly classified into two – Spatial domain fusion and Transform domain fusion. The disadvantage of spatial domain approaches is that they produce spatial distortion in the fused image. Spectral distortion becomes a negative factor while we go for further processing, such as classification problem. This problem of spectral distortion in image fusion will be well handled by transform domain approach. Fusing the image for extracting the complete information, initially the object is sensed in different orientations and imaged. These images contain redundant or complementary information. In this paper we develop an algorithm which implements image fusion technique that provides information without redundancy. This algorithm is developed using MATLAB. We use image acquisition and image processing toolbox in MATLAB. In order to implement image fusion we make use of Wavelet Transform technique. Some of the applications of image fusion are Aerial and Satellite imaging, Medical imaging, Robot Vision, Concealed Weapon Detection, Multi Focus image fusion, Digital Camera application, Battlefield Monitoring. Key Words: Image, Fusion, Spatial domain, Transform domain