Moving Target Imaging via Computational Ghost Imaging Combined With Artificial Bee Colony Optimization
Yuanjin Yu, Jiali Zheng, Shizhuang Chen, Zhaohua Yang
Abstract
When a reasonable image is captured using the method of computational ghost imaging (CGI), it usually requires a number of illuminations. For example, to fully sample an <inline-formula> <tex-math notation="LaTeX">$N \times N$ </tex-math></inline-formula> pixel object, traditional CGI needs <inline-formula> <tex-math notation="LaTeX">$N^{2}$ </tex-math></inline-formula> illuminations. When the target is dynamic and its lateral motion distance exceeds the range of a single pixel, the correlation between the detected signal and the illumination pattern is lost, which degrades the imaging quality. To overcome the imaging problem of moving objects, we propose an image reconstruction method based on the artificial bee colony (ABC) algorithm, which is used to estimate the motion velocity, and the image is reconstructed by actively shifting patterns according to the estimated velocity. Numerical simulations and experiments demonstrate the effectiveness of the proposed method. The results show that the velocity of the target can be retrieved using the ABC algorithm and that the image can be reasonably reconstructed.