원문정보
초록
영어
Medical data analysis from different medical image sensor has been increased in medical field. The image fusion techniques which are the efficient way of adding and improving medical imaging information have drawn increasing concentration from the medical society. In this paper, we proposed a new scheme of Neighbouring Pixel Selection (NPS) fusion rule for laplacian pyramidal multiscale decomposition-based fusion of medical images considering the consistency verification of both intrascale and interscale parameters. A combining set of coefficients from the Multiscale Representations (MSR) of the source images is computed by utilization of neighborhood information. An efficient fusion scheme is also proposed for different modalities. Experimental results show that proposed method produces better results than the existing methods.
목차
I. Introduction
II. Fusion Rules
1. Activity-Level Measurement
2. Coefficient Grouping
3. Coefficient Combination
4. Consistency Verification
III. Neighbouring Pixel Selection (NPS) Method
1. Medical Image Fusion
2. Flow Graph of NPS Fusion Rule
3. Problem Formulation
4. Neighbouring Coefficient Selection
5. LPT+NPS Based Fusion
6. NPS Fusion Algorithm
IV. Evaluation Criteria
1. Root Mean Square error (RMSE)
2. Peak Signal to Noise Ratio
3. Image Quality Index
4. Mutual Information
5. Objective Evaluation Metric
V. Result Analysis
VI. Conclusion
References