원문정보
초록
영어
Faced with the poor ability of traditional transform domain tools to capture the image information and the high requirements on the precision and real-time of medical imaging, a novel fusion technique for medical images based on shearlet transform (ST) and compressive sensing (CS) model is proposed in this paper. Due to the better competence of image information capturing, ST is utilized to conduct the multi-scale and multi-directional decompositions of source images. In addition, the measurement matrix is adopted to realize the sparse representation of the high-frequency coefficients obtained from ST. The fusion data of high-frequency sub-images can be attained via the largest-value method. Finally, the final fused image can be obtained by using inverse ST. Compared with current typical techniques especially the non-negative matrix factorization based ones; simulation experimental demonstrates that the proposed one has remarked superiorities in terms of both subjective and objective evaluations.
목차
1. Introduction
2. Basic Theories
2.1. ST Basic Model [8-10]
2.2. Basic Compressive Sensing Model [11, 12]
3. Medical image Fusion based on ST and CS Model
3.1. Image Fusion Basic Model
3.2. Low-frequency Microcosmic Subband Image Fusion
3.3. High-frequency microcosmic subband image fusion
4. Experimental Results and Analysis
4.1. Experiments Introduction
4.2. Experimental Results and Analysis
5. Conclusion
Acknowledgments
References