Dual-tree biquaternion wavelet transform and its application to color image fusion

Abstract

The dual-tree framework of wavelet transforms has gained popularity since they provide low-redundancy directional analyses of images. However, the existing dual-tree wavelets are developed for gray-scale images, which can not process the color image in a holistic manner. In this paper, we proposed the definition of dual-tree biquaternion wavelet transform for color image. The color image decomposition and reconstruction based on the proposed transformation can be implemented through a quaternion-valued dual-tree filter bank. The application of proposed transform to the color image fusion is also discussed. The experimental results show that the proposed method preserves more detailed information of the source color images. Besides, the experiments conducted on public multifocus color image dataset demonstrate the superiority of our strategy over the state-of-the-art methods in both objective metrics and visual observations. The experimental results indicate that the proposed transformation can be a useful tool for the color image fusion tasks.