A new method for multi-focus image fusion using Countourlet transform

Image fusion based on Contourlet transform will produce a pseudo-Gibbs effect for the lack of translation invariance. We proposed a method which combined Contourlet transform with image blocking fusion to solve this problem. Firstly a new fusion rule based on Contourlet transform was proposed to get the initial fused image. Secondly, source images and the initial fused image were partitioned into equal size image blocks. The source image blocks, which were more similar to the corresponding initial fused image blocks, were chosen as the final fused image blocks. For the border part between the clear part and blurring one, the initial fusion results were kept. Experimental results show that this method can effectively eliminate the image distortion arising from Contourlet transform lack of translation invariance. The fusion effect is better than the effect of image blocking and Contourlet transform fusion methods.