Color Multifocus Image Fusion Using Empirical Mode Decomposition

Abstract

In this paper, a recently proposed grayscale multifocus image fusion method based on the first level of Empirical Mode Decomposition (EMD) has been extended to color images. In addition, this paper deals with low contrast multifocus image fusion. The major advantages of the proposed methods are simplicity, absence of artifacts and control of contrast, while this isn’t the case with other pyramidal multifocus fusion methods. The efficiency of the proposed method is tested subjectively and with a vector gradient based objective measure, that is proposed in this paper for multifocus color image fusion. Subjective analysis performed on a multifocus image dataset has shown its superiority to the existing EMD and DWT based methods. The objective measures of grayscale and color image fusion show significantly better scores for this method than for the classic complex EMD fusion method.

Keywords

• Empirical Mode Decomposition (EMD)
• Discrete Wavelet Transform (DWT)
• image fusion
• multifocus
• low contrast