Image fusion based on complex-shearlet domain with guided filtering

• Published in: Multidimensional systems and signal processing Vol. 28; no. 1; pp. 207 - 224
• Main Authors: Liu, Shuaiqi, Shi, Mingzhu, Zhu, Zhihui, Zhao, Jie
• Format: Journal Article
• Language: English
• Published: New York Springer US 2017; Springer Nature B.V
• Subjects: Artificial Intelligence; Circuits and Systems; Computer vision; Defense industry; Electrical Engineering; Engineering; Filtration; Image processing; Signal,Image and Speech Processing; SML; CST; Guided filtering; Image fusion
• ISSN: 0923-6082; 1573-0824
• DOI: 10.1007/s11045-015-0343-6

Combined the advantages of time-frequency separation of complex shearlet (CST) with the feature of guided filtering, a new image fusion algorithm based on CST domain and guided filtering is proposed. Firstly, CST is utilized for decomposition of the source images. Secondly, two scale guided filtering fusion rule is applied to the low frequency coefficients. Thirdly, larger sum-modified-Laplacian with guided filtering fusion rule is applied to the high frequency coefficients. Finally, the fused image is gained by the inverse CST. The algorithm can not only preserve the information of the source images well, but also improve the spatial continuity of fusion image. Experimental results show that the proposed method is superior to other current popular ones both in subjective visual and objective performance.