Study Of Multifocus Image Fusion Algorithms In Wavelet Domain

Image fusion integrates the information obtained by different imaging systems about the same scene and objects or by the same sensor in different time according to some rules and then creates a new image with higher quality. With the availability of multisensor data in many fields such as remote sensing, machine vision, medical imaging and military applications, image fusion has emerged as a new and promising research area.This paper mainly discusses an important application of image fusion in digital camera: multifocus image fusion. Firstly, we briefly review the basic theories of image fusion and multifocus image fusion, some existing spatial-domain and frequency- domain methods for multifocus image fusion are discussed, and the results of those methods are evaluated and analyzed by experiments. Secondly, we carefully analyze the disadvantages of the existing wavelet-based image fusion algorithms. The analysis shows that their underlying fusion criteria are unsuitable for preserving high-frequency detail information of source images. Finally, a new wavelet-based multifocus image fusion algorithm is proposed, which is based on the energy of cross band in wavelet domain. Experimental results show that the proposed algorithm improves the image fusion performance effectively. Particularly, it preserves the detail structures in images.