LIU Xu-han, SHEN Xiao-hong, YAO Hai-yang, et al. Image compressed sensing based on tent chaos measurement matrix[J]. 2014, 33(9): 26-28+31. DOI: 10.13873/J.1000-9787(2014)09-0026-03.
基于帐篷混沌观测矩阵的图像压缩感知
摘要
针对目前典型的随机观测矩阵在不确定性、硬件实现、存储方面的缺陷
利用混沌序列优异的伪随机性
设计了一种基于帐篷混沌序列的观测矩阵。该矩阵的参数和初始值一旦确定
每一时刻的状态是可以完全重现的
克服了随机观测矩阵需要大量实验求平均来降低不确定性的缺点。仿真实验中
分别利用帐篷混沌观测矩阵和3种典型的随机观测矩阵对不同数据量的图像信号进行压缩与重构
结果表明:所设计的帐篷混沌观测矩阵不但重构性能优于随机观测矩阵
而且也克服了随机观测矩阵在硬件实现、存储方面的缺陷
大大减少了存储空间和传输带宽的压力。
Abstract
Aiming at deficiencies in uncertainty
hardware implementation and storage of the typical random measurement matrix currently
design a kind of measurement matrix based on tent chaos sequence by utilizing the excellent pseudo-random of the chaotic sequence. Once parameters and initial values of the matrix is determined
each moment of the state can be completely recreated
overcome the defect of random measurement matrix that needs a large number of experiments to reduce the uncertainty. Through compression and reconstruction of different amounts of image signal by utilizing the tent chaos measurement matrix and three typical random measurement matrixes
the results prove that the reconstruction performance of tent chaos measurement matrix is superior to typical random measurement matrix
and the deficiencies of hardware implementation and storage of random measurement matrixes are overcome
reduce storage space and transmission bandwidth pressure greatly.