摘要
分块融合易产生块效应,而Contourlet变换融合方法不具备平移不变性。结合以上方法的各自优势,提出1种新的基于Contourlet变换与块分割相结合的多聚焦图像融合方法。首先对图像进行分块并评价子块清晰度,然后通过建立标记矩阵,确定图像清晰区域、模糊区域及两者的交界区域。对非交界区域,直接选取清晰子块的像素值作为融合后的像素值;对介于清晰与模糊之间的交界区域,采用1种新的基于Contourlet变换的融合算法完成融合。实验结果表明,该方法有效抑制了图像块效应,且具备平移不变性,融合效果明显优于分块和Contourlet变换融合方法。
Image block fusion is easy to produce block effect, while Contourlet transform does not have translation invariance. Combining these two methods according to their respective advantages, a method which combines Contourlet transform with block fusion method is proposed. Firstly, source images segmentation and evaluate sub-blocks definition are finished. Secondly, the clear part, blurting part and the border part between the clear part and blurring one is determined by establishing tag matrix. For non-border parts, clear sub-blocks are selected as the corresponding blocks of the fused image. The part between clear part and blurring one is fused based on Contourlet transform using a new fusion rule. Experimental results show that the method can suppress image block effect effectively, and the fusion effect is better than the effect of image block and Contourlet transform fusion methods.
出处
《电视技术》
北大核心
2011年第23期22-24,43,共4页
Video Engineering
基金
吉林省教育厅"十二五"科学技术研究项目(吉教科合字[2011]第108号)
吉林省科技发展计划项目(20100368)
