Multifocus image fusion using Discrete Wavelet Transform and Sparse Representation

Multifocus image fusion increases the depth of field of a sensor by combining the images of same scene with different focus settings. So, extracting the useful image information of different source images plays a crucial role in image fusion process. In this paper, a novel fusion algorithm based on Discrete Wavelet Transform (DWT) and Sparse Representation (SR) is proposed. Initially, DWT is applied to extract the low frequency components and high frequency components of source images. High frequency components are merged using SR based fusion approach and low frequency components are combined using variance as activity level measurement. Finally, inverse DWT is performed on the fused coefficients to get the fused image. Experimental results demonstrate the effectiveness of proposed method in terms of visual perception and quantitative analysis.