Discrete Wavelet Transform and Cross Bilateral Filter based Image Fusion

Abstract

Abstract-The main object ive of image fusion is to obtain an enhanced image with more relevant informat ion by integrating complimentary information fro m two source images. In this paper, a novel image fusion algorith m based on discrete wavelet transform (DWT) and cross bilateral filter (CBF) is proposed. In the proposed framewo rk, source images are decomposed into low and high frequency subbands using DWT. The low frequency subbands of the transformed images are combined using pixel averag ing method. Meanwhile, the high frequency subbands of the transformed images are fused with weighted average fusion ru le where, the weights are co mputed using CBF on both the images . Finally, to reconstruct the fused image inverse DWT is performed over the fused coefficients. The proposed method has been extensively tested on several pairs of mu lti-focus and multisensor images. To compare the results of proposed method with d ifferent existing methods, a variety of image fusion quality metrics are emp loyed for the qualitative measurement. The analysis of co mparison results demonstrates that the proposed method exhibits better results than many other fusion methods, qualitatively as well as quantitatively.

A Novel Method for Infrared and Visible Image Fusion using Filters

“…The bilateral filter [1], [2], [3] is a non-linear filter introduced by Tomasi and Manduchi used to smooth or blur an image. Several qualities that explain its success are, 1.…”