Computer Engineering and Applications ›› 2006, Vol. 42 ›› Issue (18): 8-12.

• 博士论坛 • Previous Articles     Next Articles

Absolute Value Activity, Regional Similarity and
Their Application in Multi-spectral Image Fusion


  • Received:1900-01-01 Revised:1900-01-01 Online:2006-06-21 Published:2006-06-21



Abstract: In this paper, the new measurements named Absolute Value Activity(AVA)and Regional Similarity(RS)for source images are proposed and their properties are discussed. Based on AVA and RS, a new multi-spectral image fusion algorithm is also proposed and applied by using real multi-spectral images. Subjective qualitative evaluation and objective quantitative analysis of the experiment results prove that the new algorithm can integrate the information and retain the features of source images more effectively compared with several classical algorithms in the field of multi-spectral image fusion.

Key words: Image fusion, Wavelet transform, Absolute Value Activity (AVA), Regional Similarity (RS)

摘要: 本文提出了图像绝对值活跃度和区域相似度的新测度准则,分析了其性质,并在此基础之上建立了基于源图像绝对值活跃度和区域相似度的新的多光谱图像融合算法。对真实多光谱图像融合实验结果的主观定性评价和客观定量分析说明,与目前融合领域中的几种同类经典算法相比,该算法能够更有效地融合源图像信息并更好地保持源图像特征。

关键词: 图像融合 小波变换 绝对值活跃度 区域相似度