Theoretical analysis of correlation-based quality measures for weighted averaging image fusion

Recently introduced correlation-based quality measures have received lots of attention due to the fact that they do not need ground-truth reference images to evaluate the performance of image fusion algorithms. In this paper we focus on theoretical analysis of these correlation-based quality measures when they are used to judge the performance of weighted averaging image fusion algorithms. The purpose of this paper is to rigorously prove that the correlation-based quality measures have some undesired behavior under certain conditions. We employ a statistical model for the observed sensor images and study the properties of these correlation-based quality measures. Our analysis shows that when we change the power of the desired signal or the noise in the input images, these correlation-based quality measures exhibit bad behaviors in some cases, indicating higher quality when lower quality is evident. The sufficient conditions for when the undesired behavior occur and the intuitive explanation for our observation are given in this paper. Investigations with real images also demonstrate the utility of the theoretical analysis, by illustrating its predictive capabilities.

[1]  Alexander Toet,et al.  Performance comparison of different gray-level image fusion schemes through a universal image quality index , 2003, SPIE Defense + Commercial Sensing.

[2]  Vladimir Petrovic,et al.  Optimizing multiresolution pixel-level image fusion , 2001, SPIE Defense + Commercial Sensing.

[3]  Chuanbo Wen,et al.  Multiresolution Image Fusion Algorithm Based on Probabilistic Model , 2006, 2006 6th World Congress on Intelligent Control and Automation.

[4]  Thomas M. Cover,et al.  Elements of Information Theory , 2005 .

[5]  Nishan Canagarajah,et al.  A Similarity Metric for Assessment of Image Fusion Algorithms , 2008 .

[6]  Rick S. Blum.  On multisensor image fusion performance limits from an estimation theory perspective , 2006, Inf. Fusion.

[7]  Rick S. Blum,et al.  Theoretical analysis of an information-based quality measure for image fusion , 2008, Inf. Fusion.

[8]  Vladimir Petrovic,et al.  Objective image fusion performance measure , 2000 .

[9]  Ravi K. Sharma,et al.  Probabilistic Image Sensor Fusion , 1998, NIPS.

[10]  Henk J. A. M. Heijmans,et al.  A new quality metric for image fusion , 2003, Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429).

[11]  Richard K. Moore,et al.  Analysis of Image Quality for Image Fusion via Monotonic Correlation , 2009, IEEE Journal of Selected Topics in Signal Processing.

[12]  Rick S. Blum,et al.  Multi-sensor image fusion and its applications , 2005 .

[13]  G. Qu,et al.  Information measure for performance of image fusion , 2002 .

[14]  Rick S. Blum,et al.  Concealed weapon detection using color image fusion , 2003, Sixth International Conference of Information Fusion, 2003. Proceedings of the.

[15]  R.S. Blum,et al.  Experimental tests of image fusion for night vision , 2005, 2005 7th International Conference on Information Fusion.

[16]  Rick S. Blum,et al.  Fusion of visual and IR images for concealed weapon detection , 2002, Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997).

[17]  A. Bovik,et al.  A universal image quality index , 2002, IEEE Signal Processing Letters.

[18]  Vladimir Petrovic.  Multilevel image fusion , 2003, SPIE Defense + Commercial Sensing.

[19]  Vladimir S. Petrovic,et al.  Sensor noise effects on signal-level image fusion performance , 2003, Inf. Fusion.

[20]  Rick S. Blum,et al.  A statistical signal processing approach to image fusion for concealed weapon detection , 2002, Proceedings. International Conference on Image Processing.