Compressed sensing MR image reconstruction based on nonlocal total variation and partially known support. (Chinese. English summary) Zbl 1363.92028

Zhao, Di; Du, Huiqian; Han, Yu; Mei, Wenbo

Trans. Beijing Inst. Technol. 36, No. 3, 308-313 (2016).

Summary: By exploiting the similarity of structure between the reference and the target images, a novel compressed sensing (CS)-based reconstruction method is proposed for MR image. Indexes of the \(L\) largest wavelet coefficients of the reference image are extracted and regarded as the known part of the desired target image’s support, and the \(l_1\) norm of the wavelet coefficients belonging to the complement to the known support is constrained. Furthermore, the nonlocal total variation (NLTV) is utilized as a regularization term to construct the objective function. Then the target image is reconstructed via a fast composite splitting algorithm (FCSA). Experimental results demonstrate that the proposed method can preserve edges and details while suppressing noise efficiently. It outperforms conventional CS-MRI and other similar reconstruction methods under the same sampling rate.

MSC:

92C55	Biomedical imaging and signal processing
94A08	Image processing (compression, reconstruction, etc.) in information and communication theory

Keywords:

magnetic resonance imaging; compressed sensing; nonlocal total variation; fast composite splitting algorithm