Influence Of Source Inhomogeneity On Compressed Sensing-Based Thz Imaging

In this paper, we investigate the effect of source inhomogeneity on THz imaging. The numerical analysis are conducted through obtaining the compressed data from a compressive sensing (CS) based THz imaging system in the optical-design software Zemax. The analysis results demonstrate that Peak Signal to Noise Ratio (PSNR) of the system is dramatically destroyed by inhomogeneous distribution of THz source. The combination of the real light distribution in the optimized reconstruction procedure is then proposed and its efficiency is demonstrated by simulation results. It is expected that the results in this paper may provide guidance for the high quality CS-based THz imaging in real-world application.