Improving the signal-to-noise ratio of thermal ghost imaging based on positive–negative intensity correlation

• Identifying different characteristics of positive and negative correlation in HBT.
• Cancelling noise by utilizing the symmetry of positive and negative correlations.
• Comparing to existing schemes, over 60% SNR improvement in the image results.

Ghost imaging with thermal light is a topic in optical imaging that has aroused great interest in recent years. However, the imaging quality must be greatly improved before the technology can be transferred from the lab to engineering applications. By means of correspondence ghost imaging (CGI) with a pseudo-thermal light source and appropriate sorting of the intensity fluctuations of the signal and reference beams, we obtain the positive and negative Hanbury Brown and Twiss intensity correlation characteristics of the optical field. Then, for ghost imaging of a transmissive binary object, we find that by subtracting the negative from the positive fluctuation frames of the presorted reference detector signals, the signal-to-noise ratio can be effectively increased, with almost all the background noise eliminated. Our results show that, compared with the generic CGI technique, the signal-to-noise ratio can be increased by nearly 60%.