The Research On Multi-focus Image Fusion

Since the 1990s, the image fusion technology research is increasing, the advantage of image fusion is in medical imaging and diagnosis, intelligent traffic, human vision aid, intelligent manufacturing, industrial process and military field have been widely used. Because of in various fields of demand, drawing image fusion technology research for many researchers increasingly concerned. The main idea of image fusion is the process of combining two or more images captured from different kinds of sensors into an image. And this process can obtain resultant images with higher confidence, more accuracy and is suitable for human vision and computer processing.In the paper, we have researched the Multi-focus image fusion. firstly, image fusion algorithms based on wavelet transform and its processing schedule is introduced, from analyze the fusion results, we can know that the wavelet analysis method is not perfect, it can only present isotropy property of the object, but for anisotropy property the wavelet transform is not good. So in some edges, there is an obvious fuzzy phenomenon. Contourlet transform is introduced in this article because of it has better multi-resolution and directional advantage than the wavelet transform. It can provide more information for the fusion image because it can extract more features from the original images. So we have researched the principle and the implementation of algorithm of the Contourlet transform deeply, the simulation results show that the Contourlet transformation of image fusion is obviously superior to the wavelet transform.Secondly, a method of image fusion methods based on block segment is introduced. Focus measure is the key point, so several common focus measures has been compared in the performance of image fusion. The image fusion algorithm based on block segment is simple, but easy to produce a mass effect. A method for image fusion based on Contourlet transform and block segment is proposed, it can eliminate the image block effect and improve the fusion quality effectively.Finally, to the noise image fusion method has further researched and we have also given an effective noise image fusion scheme. The blocks segmentation algorithm and a new anti-noise evaluation operator applied to the noise image. The new measure is also compared with other multi-focus fusion methods and the experimental results show that the new measure can provide better performance than other focus measures and methods.