Multimodal Image Fusion using Discrete Wavelet Transform and Support Vector Machine

Abstract

Image fusion process effectively integrates the images taken by multiple sensors and produces a single image extracting all the relevant information from the source images. Image fusion provides an effective way of reducing volume of information while at the same time extracting all the useful information from the source images. For image fusion Discrete Wavelet Transform (DWT) is used for feature extraction and Support Vector Machines (SVM) is used for the classification of the coefficients of DWT. DWT reduces resolution and reduces the computation time. It also decreases the storage space required. The proposed fusion method incorporates feature extraction of the wavelet transform subbands such as energy, entropy, mean and standard deviation then classification of these features of can be done using classifiers such as Support Vector Machines.