Edge detection of ghost imaging based on frequency domain filtering

Abstract

Ghost imaging is a novel method which can reconstruct the imaging via a single-pixel detector. Compared with traditional edge extraction methods, ghost imaging can be used for edge detection of unknown targets without acquiring the target image. Initially, we applied a ghost image edge extraction method using discrete cosine transform and rectangular window function based frequency domain filter. Then, to eliminate the ringing artifact problem, we proposed a novel ghost imaging edge detection method using the Crone operator and Butterworth filter, which can obtain high-quality edge results of unknown targets at a very low sampling rate. Finally, many simulations were performed to validate the proposed strategy, and the results show that the proposed method can get high quality edge information of an unknown target at a low sampling rate.