Optimization of Compressive Ghost Imaging in the Terahertz Range Using Correlation-Based Ranking in Optical System Transmission

One of the current challenges in terahertz (THz) optics is overcoming the limitations of multipixel cameras, which suffer from low sensitivity and resolution. A promising solution lies in the use of single-pixel imaging techniques, employing spatial light modulators (SLMs), particularly the ghost imaging method combined with compressive measurements. This approach enables high-speed imaging comparable to multipixel cameras, while offering a higher signal-to-noise ratio. However, the quality of the reconstructed images is significantly influenced by the characteristics of the illumination patterns generated by the SLM. This paper proposes an optimization strategy for the ghost imaging method, based on ranking orthogonal basis patterns by their correlation coefficients during optical system transmission. This technique improves image quality compared to random pattern selection at the same levels of undersampling in compressive measurement algorithms.