计算机工程与应用 ›› 2015, Vol. 51 ›› Issue (23): 184-188.

• 图形图像处理 • 上一篇    下一篇

基于小波变换的多聚焦图像融合算法

赵立强,杨大志,周艳红,向  洁   

  1. 河北科技师范学院 数学与信息科技学院,河北 秦皇岛 066004
  • 出版日期:2015-12-01 发布日期:2015-12-14

Multi-focus image fusion algorithm based on wavelet transform

ZHAO Liqiang, YANG Dazhi, ZHOU Yanhong, XIANG Jie   

  1. School of Mathematics and Information Science & Technology, Hebei Normal University of Science & Technology, Qinhuangdao, Hebei 066004, China
  • Online:2015-12-01 Published:2015-12-14

摘要: 针对基于小波变换的多聚焦图像融合算法,改进融合规则和融合算子,低频分量采用以相关系数作为阈值的加权平均算法,高频分量采用基于区域特征的融合算法,并对最佳分解层数与最佳小波基的选取进行优化验证。通过对实验结果的分析,选用bior4.4小波,进行最佳分解层数小波分解,并应用改进的融合规则,在融合多聚焦图像的效果上,与其他多种融合算法相比,各项评价指标都比较理想。

关键词: 多聚焦图像融合, 小波变换, 小波基, 融合规则

Abstract: In view of multi-focus image fusion algorithm, fusion rules and fusion operators are improved. The low-frequency component uses the weighted average algorithm with correlation coefficient as threshold, the high-frequency component uses fusion algorithm based on regional characteristics. In addition, the optimal decomposition level and the selection of the best wavelet basis are verified. The evaluation indexes of multi-focus image fusion based on improved fusion rules and bior4.4 wavelet with optimal decomposition level, through the analysis of experimental results, are better than many other fusion algorithm.

Key words: multi-focus image fusion, wavelet transform, wavelet basis, fusion rules