Image fusion algorithm based on improved K-singular value decomposition and Hadamard measurement matrix. (August 2018)

• Record Type: Journal Article
• Authors: Yang, Qiang; Wang, Huajun
• Is Part Of: Journal of algorithms & computational technology. Volume 13(2019)
• Publication Date: 2018-08
• Subjects: Sparse representation -- measurement matrix -- image reconstruction -- image fusion; Computer algorithms -- Periodicals; Numerical calculations -- Periodicals; Computer algorithms; Numerical calculations; Periodicals; 518.1
• DOI: 10.1177/1748301818791511
• Languages: English
• ISSNs: 1748-3018

Abstract

To solve the problem of high time and space complexity of traditional image fusion algorithms, this paper elaborates the framework of image fusion algorithm based on compressive sensing theory. A new image fusion algorithm based on improved K-singular value decomposition and Hadamard measurement matrix is proposed. This proposed algorithm only acts on a small amount of measurement data after compressive sensing sampling, which greatly reduces the number of pixels involved in the fusion and improves the time and space complexity of fusion. In the fusion experiments of full-color image with multispectral image, infrared image with visible light image, as well as multispectral image with full-color image, this proposed algorithm achieved good experimental results in the evaluation parameters of information entropy, standard deviation, average gradient, and mutual information.