27 September 2017 Method for multispectral images denoising based on tensor-singular value decomposition
Author Affiliations +
Abstract
We propose a sparsity measure for third-order tensor, called all-dimension tensor nuclear norm (AD-TNN). In specific, we exploit the tensor-singular value decomposition and tensor nuclear norm (TNN), considering TNN along every dimension and then construct our AD-TNN measurement. Based on AD-TNN, we construct a model for multispectral images denoising. We also employ the alternating direction method of multipliers (ADMM) to solve our model. Experimental results show that our method outperforms all the compared methods under comprehensive quantitative performance measures.
© 2017 Society of Photo-Optical Instrumentation Engineers (SPIE)
Wenjia Zeng, Wenjia Zeng, Xianggan Zhang, Xianggan Zhang, Yechao Bai, Yechao Bai, } "Method for multispectral images denoising based on tensor-singular value decomposition," Journal of Applied Remote Sensing 11(3), 035019 (27 September 2017). https://doi.org/10.1117/1.JRS.11.035019 . Submission: Received: 25 May 2017; Accepted: 31 August 2017
Received: 25 May 2017; Accepted: 31 August 2017; Published: 27 September 2017
JOURNAL ARTICLE
10 PAGES


SHARE
RELATED CONTENT

Appearance-based color face recognition with 3d model
Proceedings of SPIE (March 19 2013)
3D discrete shearlet transform and video denoising
Proceedings of SPIE (September 27 2011)
Multicomposite wavelet estimation
Proceedings of SPIE (September 27 2011)

Back to Top