15 October 2012 Nonlocal structure tensor for pixel level image fusion
Author Affiliations +
In this paper, we present a nonlocal structure tensor for feature description and apply it to the pixel level image fusion within the wavelet framework. Local geometric shape information in wavelet coefficients can be extracted and recognized by structure tensor. The structure tensor element is processed by use of the nonlocal means filter before calculating its eigen-values. With the eigen-values of two source data, an adaptive weight function is employed to reconstruct new detail coefficients of the fused image. Experimental results show the performance of the proposed scheme.
© (2012) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Fen Luo, Fen Luo, Bibo Lu, Bibo Lu, "Nonlocal structure tensor for pixel level image fusion", Proc. SPIE 8419, 6th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optoelectronic Materials and Devices for Sensing, Imaging, and Solar Energy, 84192N (15 October 2012); doi: 10.1117/12.978215; https://doi.org/10.1117/12.978215


Multiple wavelet coherence analysis
Proceedings of SPIE (February 15 2006)
Multifocus image fusion with trace-based structure tensor
Proceedings of SPIE (November 28 2011)
Multispectral image fusion using an improved wavelet transform
Proceedings of SPIE (September 17 2001)
Medical image processing using wavelets
Proceedings of SPIE (October 26 2006)

Back to Top