De-noising research on terahertz holographic reconstructed image based on weighted nuclear norm minimization and connected domains

Wenshu Ma,Qi Li,Yutong Wang,Jianye Lu
DOI: https://doi.org/10.1117/12.2500302
2018-07-24
Abstract:The terahertz Gabor inline holographic reconstructed images have the characteristics of large backgrounds and small targets, which are different from the standard images of conventional research. Thus, we carried out the connected area disposal method before the Weighted Nuclear Norm Minimization (WNNM) in this paper. Connected area disposal can split the image into multiple independent regions to denoise the large background. The WNNM denoising method can preserve the details of the targets well. The numerical analysis and experimental researches of the denoising results proved that the proposed method in this paper can get a better denoising effect than WNNM algorithm alone.
Computer Science
What problem does this paper attempt to address?