Compressive Ghost Imaging of the Moving Object Using the Low-Order Moments

Authors:
Open Access
Publisher website
Google Scholar
Abstract
Ghost imaging reconstructs the image based on the second-order correlation of the repeatedly measured light fields. When the observed object is moving, the consecutive sampling procedure leads to a motion blur in the reconstructed images. To overcome this defect, we propose a novel method of ghost imaging to obtain the motion information of moving object with a small number of measurements, in which the object could be regarded as relatively static. Our method exploits the idea of compressive sensing for a superior image reconstruction, combining with the low-order moments of the images to directly extract the motion information, which has the advantage of saving time and computation. With the gradual motion estimation and compensation during the imaging process, the experimental results show the proposed method could effectively overcome the motion blur, also possessing the advantage of reducing the necessary measurement number for each motion estimation and improving the reconstructed image quality.
Funding Information
  • National Natural Science Foundation of China (61631014, 61401036, 61471051)
  • National Science Fund for Distinguished Young Scholars of China (61225003)
  • the BUPT Excellent Ph.D. Students Foundation (CX2019224)

This publication has 26 references indexed in Scilit: