A Novel DWT Based Multi-focus Image Fusion Method

Abstract

This paper presents a novel discrete wavelet transform (DWT) based approach for multi-focus image fusion. The method is developed by setting different fusion rules for combining the coefficients of low frequency and high frequency sub-bands separately followed by a consistency verification process. In our method, the coefficients in the low frequency domain are selected based on a maximum sharpness focus measure scheme, while choosing the high frequency sub-bands coefficients, a maximum neighboring energy based fusion scheme is proposed. The performance assessment of the proposed method was conducted and compared with several conventional existing fusion methods. Experimental results can clearly demonstrate that the proposed method is an effective and feasible multi-focus image fusion algorithm.