热光关联成像的均方收敛无偏估计方法

关联成像实验和理论证明了光可以用特殊的方式传递信息。文中采用半经典理论,按照赝热光场、光电探测和关联计算的成像过程,系统分析了关联成像的理论基础,并给出了系统的视场、空间分辨率和对比度。在此基础上,对传统关联成像的线性计算方法进行了改进,使之为物体透射函数的均方收敛无偏估计。相应的计算鬼成像仿真实验表明:在相同测量次数下,尤其是低于奈奎斯特采样频率时,该方法显著提高了鬼像的峰值性噪比并有效抑制了背景噪声。

Mean square convergence unbiased estimation of thermal light correlated imaging

The theory that light can transmit information in a unique way has been proved by the experiment and theory of correlated imaging. In this paper, the principles of correlated imaging were discussed in semi-classical interpretations. In the view of pseudo-thermal light field, photoelectric detection and correlated computation, the imaging process was analyzed. Field of view, spatial resolution and contrast of the system were given. On this basis, the traditional linear correlation algorithm was improved to make the ghost image a mean square convergence unbiased estimation of the object transmission function. The corresponding computational ghost imaging experiment measurement indicates that under the same number, especially less than the Nyquist frequency, the PSNR is significantly improved and background noise is effectively suppressed compared with the traditional algorithm.