Theory And Experiment Of Lens Ghost Imaging With Pseudo-Thermal Light

Ghost imaging, also known as coincidence imaging or correlated imaging, is a new imaging method. It breaks through the diffraction limit of traditional optical imaging method and has the characteristic of non-locality. Ghost imaging has been a research hot of optical imaging for its good development prospects in many fields, such as military, medical, and aerospace. With the advancement of the research, the remote detection is desired. Regardless of the length of the detection arm, the length of the reference arm should be shortened in local stand-off detection. So the research of shortening optical path by a lens has a very important significance.Based on the introduction of the basic definition and theory of ghost imaging, the theory and experiment of lens ghost imaging with pseudo-thermal light is studied. The lens ghost imaging function is derived based on ghost imaging theory and impulse response function of linear optical system. Its rationality is analyzed by comparing with the lensless ghost imaging function. Through the ghost imaging simulation of double-slits, the function is further verified. Experiments on lens ghost imaging with pseudo-thermal light in transmission and reflection show that the imaging quality is quite clear when the object is 30 m far. It is found that the imaging resolution can be improved when a lens is placed in the optical path between the round glass and the field stop.In order to improve the image quality, compressed sensing ghost imaging is studied. The main steps of orthogonal matching pursuit algorithm are given based on the theory of compressed sensing. The performance of the algorithm is studied through the MATLAB simulation. By using the ghost imaging experiment data to construct the observation vector and observation matrix in compressed sensing theory, the compressed sensing ghost imaging is implemented. Comparing the compressed sensing ghost imaging with traditional ghost imaging in the case of same sample rate, the results show that the image reconstructed by compressed sensing ghost imaging can obtain the higher quality with fewer experiment data, which laid the groundwork for real time imaging. Finally, the best solution with both parameters is obtained, after comparing the change of visibility and SNR of reconstruction image with the maximum number of iterations and sampling rate.