Morphology-based visible-infrared image fusion framework for smart city

Abstract

Sparse representation-based approaches are often applied to image fusion. Owing to the difficulties of obtaining a complete and non-redundant dictionary, this paper proposes a hierarchical image fusion framework that applies layer-by-layer deep learning techniques to explore the detailed information of images and extract key information of images for dictionary learning. According to morphological similarities, this paper clusters source image patches into smooth, stochastic, and dominant orientation patch group. High-frequency and low-frequency components of three clustered image-patch groups are fused by max-L1 and L2-norm based weighted average fusion rule respectively. The fused low-frequency and high-frequency components are combined to obtain the final fusion results. The comparison experimentations confirm the feasibility and effectiveness of the proposed image fusion solution.