Development of an image fusion technique based on combined wavelets and DCT: Application to visual sensor networks

Abstract

In this thesis, we propose a novel image fusion technique for visual sensor networks (VSNs). The principal goal of this proposed technique is carried out on captured images from VSN to compose a single composite one with more accurate visual data and further information from the input images. It can be suitably used for visual perception and computer processing. This new technique is based on the Discrete Cosine Transform (DCT) and the Integer Lifting Wavelet Transform (ILWT). In order to achieve that, this work is divided into three main parts. The first part presents the preliminary concepts and applications of the image fusion. In the second part, a brief state of the art review of recent image fusion techniques is presented. In the last part, both qualitative and quantitative performance evaluation parameters for the proposed and the compared image fusion methods are presented to prove the outperforming of the proposed method on the compared methods.