基于图像压缩传感理论, 在手动式光学单点成像系统的基础上研究了自动式光学单点成像系统。主要介绍了系统中自动编码转盘的设计以及编码块的获取, 采用一系列优化的编码块图案作为测量矩阵, 并利用最小均方差线性估计(MMSE)重构算法进行实验。实验研究表明, 通过7.8% 低采样率即可实现对字符样本的重构。该自动编码转盘系统自动化程度较高, 误差较小, 而且可随意改变测量次数。

An automatic single-point imaging compressive sampling (CS) system was developed in the present work based on the concept of manual CS system. The design of the rotating coding disc and the way for obtaining the information of coding block were presented. A series of optimized coding blocks were used as sensing matrix and the minimum mean square error linear estimate (MMSE) algorithm was employed for image reconstruction. The quality of image reconstruction could be guaranteed when low sample rate of 7.8% was used. It is demonstrated that the proposed automatic system can work better than the manual one, with lower error, higher efficiency and more flexible in measurement times.