Differential Hadamard ghost imaging via single-round detection

The experimental implementation of computational ghost imaging based on orthogonalized patterns usually consists of two steps, corresponding to the illumination of the positive and negative elements. Here, we put forward a novel technique which avoids "negative illumination" and involves a single round of measurements. Our method is based on the properties of the second-order correlation function in Hadamard-based computational ghost imaging. Experimental results show that object information may be effectively recovered with half of the measurements compared to the conventional differential scheme. As a matter of fact, the application of our scheme is not limited to the Hadamard pattern, but it may be applied also to other orthogonal matrices. As such, it may be relevant for the general development of computational ghost imaging.