Open AccessMulti‐focus image fusion algorithm based on multilevel morphological component analysis and support vector machine
Author(s) -
Li Xiongfei,
Wang Lingling,
Wang Jing,
Zhang Xiaoli
Publication year - 2017
Publication title -
iet image processing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.401
H-Index - 45
eISSN - 1751-9667
pISSN - 1751-9659
DOI - 10.1049/iet-ipr.2016.0661
Subject(s) - support vector machine , computer science , pattern recognition (psychology) , focus (optics) , artificial intelligence , image fusion , consistency (knowledge bases) , component (thermodynamics) , feature (linguistics) , algorithm , feature vector , classifier (uml) , feature extraction , principal component analysis , image (mathematics) , linguistics , philosophy , physics , optics , thermodynamics
In this study, a novel algorithm is proposed for multi‐focus image fusion based on multilevel morphological decomposition and classifier. The attractive feature of the algorithm is that it decomposes images into several layers with different morphological components, which makes it preserve more detail information of source images. In the algorithm, source images are first decomposed by the multilevel morphological component analysis. Then, feature vectors are extracted from nature layers, and they are classified by a trained two‐class support vector machine. Then, consistency verification is employed to verify the decision matrix sets. Finally, coefficients are fused based on the decision matrix sets. Experimental results demonstrate the superiority of the proposed method in terms of subjective and objective evaluation.