Improving the noise immunity of 3D computational ghost imaging

Abstract

Computational ghost imaging (CGI) can build a three-dimensional (3D) image using the reconstructed shading images. However, it could be easily affected by the noise accumulated in the 3D reconstruction. More importantly, the selection of initial growing position will also affect the quality of the formed 3D image significantly. In this paper, we apply the technique of sub-pixel displacement to achieve smooth shading images in noisy environments and propose a method for selecting the optimal initial growing position to preserve the stereo feature of the object. We demonstrate that the surfaces of the reconstructed 3D images are more accurate using our proposed method as compared to the ones achieved by previously used methods. Our research would promote the development of 3D imaging using CGI in noisy environments.