J4 ›› 2015, Vol. 32 ›› Issue (1): 24-29.

• 图像与信息处理 • 上一篇    下一篇

单像素成像系统的场景监视算法

方绍正,邓静   

  1. 安徽建筑大学电子与信息工程学院 安徽 合肥 230601
  • 收稿日期:2014-03-31 修回日期:2014-06-30 出版日期:2015-01-28 发布日期:2015-02-04
  • 通讯作者: 方绍正(1978-)硕士,讲师,主要从事高等教育和计算机应用的方面研究。 E-mail:hanfeng4321@163.com

Scene Monitoring Algorithm Based on Single-Pixel Imaging System

Fang Shao-zheng Deng Jing   

  1. Department of Electronic and Information Engineering,Anhui Jianzhu University ,Hefei 230601,China
  • Received:2014-03-31 Revised:2014-06-30 Published:2015-01-28 Online:2015-02-04

摘要: 利用单像素成像系统进行场景监视,对变化场景进行相同的信息调制,利用场景探测强度的差值作为测量值,采用压缩感知算法只需进行一次算法求解即可获取场景变化信息。使用数字仿真技术对提出的和传统方法进行场景变化信息反演获取对比实验。在简单的二值场景变化信息获取实验中,提出的方法在采样率为1%的情况下就可以获取准确的场景变化信息;对8bit灰度场景变化信息进行获取时,由于信息量变大,为了获取较高质量的场景变化信息需要的采样数增加,但在相同采样数下,提出的算法可以获得比传统方法更准确的结果。实验结果表明采用提出的方法可以在低采样率的情况下获取准确场景变化信息,并降低了获取场景变化信息计算复杂度,对单像素成像系统应用于场景监控具有参考价值。

关键词: 图像处理, 单像素相机, 压缩感知, 场景监测, 数字微反射镜

Abstract: Scene monitoring algorithm based on single-pixel imaging system is studied in the paper. The algorithm based on compressed sensing that uses scene detection intensity difference as measured value can obtain a scene change information. The results from the proposed method and traditional method are compared using computer simulation experiments. For a simple binary scene, the accurate change information can be obtained under the sample rate 0.01. For a gray scene, because information becomes large, the number of samples required increases, but at the same number of samples, the proposed method can obtain more accurate results than that of traditional method. Simulation results show that the proposed method can obtain accurate scene change information at low sample rate, and can reduce the computational complexity. It will have a reference value in the scene monitoring based on single-pixel imaging system.

Key words: Image processing, Single pixel camera, Compressed sensing, Scene monitoring, Digital micro-mirror device