Image fusion metric based on mutual information and Tsallis entropy

被引:163
作者
Cvejic, N. [1 ]
Canagarajah, C. N. [1 ]
Bull, D. R. [1 ]
机构
[1] Univ Bristol, Dept Elect & Elect Engn, Bristol, Avon, England
关键词
D O I
10.1049/el:20060693
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
A novel image fusion performance metric using mutual information is proposed. The metric is based on Tsallis entropy, which is a one-parameter generalisation of Shannon entropy. Experimental results have confirmed that the proposed metric outperforms the standard MI metric by correlating better with the subjective quality of fused images.
引用
收藏
页码:626 / 627
页数:2
相关论文
共 6 条
[1]   MULTISENSOR IMAGE FUSION USING THE WAVELET TRANSFORM [J].
LI, H ;
MANJUNATH, BS ;
MITRA, SK .
GRAPHICAL MODELS AND IMAGE PROCESSING, 1995, 57 (03) :235-245
[2]   Fast and accurate image registration using Tsallis entropy and simultaneous perturbation stochastic approximation [J].
Martin, S ;
Morison, G ;
Nailon, W ;
Durrani, T .
ELECTRONICS LETTERS, 2004, 40 (10) :595-597
[3]  
Piella G, 2003, 2003 INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, VOL 3, PROCEEDINGS, P173
[4]   Information measure for performance of image fusion [J].
Qu, GH ;
Zhang, DL ;
Yan, PF .
ELECTRONICS LETTERS, 2002, 38 (07) :313-315
[5]   POSSIBLE GENERALIZATION OF BOLTZMANN-GIBBS STATISTICS [J].
TSALLIS, C .
JOURNAL OF STATISTICAL PHYSICS, 1988, 52 (1-2) :479-487
[6]   Objective image fusion performance measure [J].
Xydeas, CS ;
Petrovic, V .
ELECTRONICS LETTERS, 2000, 36 (04) :308-309