Multiple Speckle Patterns Ghost Imaging And 3D Computational Ghost Imaging

Ghost imaging, also called correlated imaging, is a new imaging technique which can retrieve the images of an object with high-quality in a different beam path to the object by the spatial correlation of the light field. It has been draw widespread attention in recent years. The equipment requirement, the reconstructed image quality and reconstruction time are important elements on the realization of ghost imaging system. So the study on ghost imaging schemes with simple imaging system, good performance and less imaging time has an important application prospect.First of all, a new ghost imaging scheme based on multiple speckle patterns and differential method is proposed in this thesis, named multiple speckle patterns differential ghost imaging. The scheme extracts the differential object information as the image object information which could be recovered with high quality by using the slow detectors. The theoretical analysis and simulation results show that the scheme decreases the high temporal resolution requirements for detectors by continuously detecting multiple independent speckle patterns, and improves the Signal-to-Noise rate performance by using differential method which has an important effect on the elimination of the background and other noise.Based on the above scheme, a multiple speckle patterns differential compressive ghost imaging scheme is further proposed with compressive sensing reconstruction algorithm. The scheme can effectively decrease the detection equipment requirement, improve the image quality and reduce the reconstruction time.The numerical simulation results show that the peak Signal-to-Noise rate performance is improved more than 11 dB, and the reconstruction data is decreased 75% to the results with the above scheme.At last, a 3D compressive computational ghost imaging scheme which used only three single-pixel detectors for 3D reconstruction is proposed. In the scheme, the three single-pixel detectors are used to fast recover 2D images of object with high quality individually, and 3D image of the object is reconstructed from the different shading information in the three images. The simulation results show that,the mean square error performance,using the proposed scheme with low measurements, is reduced more than 57%, in comparison with those results using the 3D computational ghost imaging scheme. The proposed 3D compressive computational ghost imaging scheme have improved the quality of 3D structure reconstruction, and have effectively decreased the sampling times, and the reconstruction time.