基于量子关联成像的图像重构算法采样数研究

量子关联成像技术采用单点强度探测，存贮信息量大，成像速度慢，需研究快速图像重构成像算法。对量子关联成像技术图像重构算法中的统计迭代法和压缩感知算法的采样次数进行了仿真分析，压缩感知算法采用二维离散余弦变换（DCT）将图像稀疏化，高斯随机矩阵作为测量矩阵，正交匹配追踪（OMP）算法对图像进行重构。结果表明：图像越大，重构图像需要的采样次数和采样时间越长，采用压缩感知算法能有效减少采样次数，从而提高系统成像速度。因此，研究量子关联成像的图像重构算法，减少图像的采样次数，对提高成像速度具有重要意义。

Sampling number of image reconstruction arithmetic based on quantum correlated imaging

Quantum correlated imaging technology adopts single-point intensity detecting, huge information storage, slow imaging speed, so faster image reconstruction algorithm is required. The simulation of samples with image reconstructing algorithm is based on statistical arithmetic and compressed sensing respectively. The inputs of the compressed sensing algorithm are sparse images which are calculated with Discrete Cosine Transform (DCT) and gauss random matrices, reconstructing image with Orthogonal Matching Pursuit (OMP). The result shows that the correlated imaging algorithm based on compressed sensing can lessen the number of measurements and save data space and speed. Therefore, the study of quantum correlated imaging image reconstruction algorithm has great significance for lessening the number of samples and improving imaging speed.