基于散斑图滤波的差分鬼成像优化算法

针对传统鬼成像重构算法采样次数多、重建图像质 量差的问题,提出了一种基于散斑 图滤波的差分鬼成像优化方法。该方法在差分鬼成像的关联阶段先对随机散斑图进行相应的 滤波处理；再将原始桶探测器值与处理后的散斑图作关联运算,得出降噪处理后的重建鬼像 ,最后通过图像融合的方法补充重建图像的细节信息。数值仿真结果表明,与传统差分鬼成 像相比,在相同采样次数下重建差分图像的峰值信噪比提高了2-4 dB。实验结果表明该 算法能够很好地兼顾采样次数和图像质量间矛盾关系,对低采样次数下的鬼成像也能取得良 好的恢复效果。

Differential ghost imaging optimization algorithm based on speckle image filtering

Aiming at the problems of the traditional ghost imaging reconstruction algorithm with many sampling numbers and poor reconstructed image quality,a di fferential ghost imaging optimization method based on speckle image filtering is proposed.In the correlation stage of differential ghost imaging,this method f irst performs corresponding filtering processing on the random speckle pattern; then associates the original bucket detector value with the processed speckle pa ttern to obtain the reconstructed ghost image after noise reduction.Finally,th e detailed information of the reconstructed image is supplemented by the method of image fusion.Numerical simulation results show that compared with traditiona l differential ghost imaging,the peak signal-to-noise ratio of the reconstruc te d differential image in this paper is improved by 2-4dB under the same sampling numbers.Experimental results show that the algorithm can take into account the contradiction between sampling numbers and image quality.It can also achieve go od recovery effects for ghost imaging under low sampling numbers.