Improving the signal-to-noise ratio of computational ghost imaging of a reflective object with a rough surface by Hadamard modulated light field

We propose an optimization scheme to improve the reconstruction quality of computational ghost imaging (GI) of a reflective target with a rough surface by using the Hadamard modulation light field (HCGI). By comparison with computational GI with a traditional Gaussian light field (GCGI), the signal-to-noise ratio of GCGI is quite bad, and it is difficult to distinguish the imaging signal from the background when the surface roughness of the object is higher, while a ghost image with better quality can be obtained by HCGI. The difference is explained by comparing the distribution of the correlation coefficient. Additionally, it is found that HCGI has better noise robustness in comparison with GI with other random coded patterns.