Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Computer Science》 2005-02
Add to Favorite Get Latest Update

Multi-Focus Image Fusion Based on Wavelet Transform and Local Energy

MIAO Qi-Guang;WANG Bao-Shu School of Computer Science, Xidian University, Xi'an 710071 Guilin University of Electronic Technology, Guilin 541004  
A new multi-focus image fusion algorithm is given in this paper, which is based on the difference of the re- gion energy of each image. The wavelet decomposition is used to decompose the original image to two different parts, namely, the low frequency part and the high frequency part. The high frequency part contains the horizontal high fre- quency, the vertical high frequency and the diagonal high frequency. For the low frequency part, an approprlate pa- rameter R is chosen to reduce the proportion of the energy of the low frequency part to that of the whole image. In this way, the proportion of high frequency part to the low frequency part is improved. The adjusted images are recon- structed with wavelet reverse transform. For the new reconstructed images, the method of comparing the region en- ergy to determine in which image the object is clear is used. The clear object is decided by the comparison of the dif- ference of energy of the two new adjusted different focus images pixel by pixel. Through this way, the clear region of each original image are decided automatically, and merged into the fused image. The approach can be applied not only to multi-focus image fusion, but also to image fusion of medical images. with a similar property of multi-focus im- ages. Experiments show that the proposed algorithm works better in extracting the clear object from the original im- ages than that of the other image fusion methods in multi-focus image fusion.
【Fund】: 国防科技预研基金(51406050301DZ0107)
【CateGory Index】: TP391.41
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved