广义半迭代硬阈值追踪算法及在“鬼”成像中的应用  

Generalized Semi-Iterative Hard Thresholding Pursuit Algorithm and Its Application in Ghost Imaging

作　　者：赵生妹[1] 唐文娟[1] 郑宝玉[1] 

机构地区：[1]南京邮电大学信号处理与传输研究院,南京210003

出　　处：《北京邮电大学学报》2016年第1期117-121,共5页Journal of Beijing University of Posts and Telecommunications

基　　金：国家自然科学基金项目(61271238;61475075;61271240);教育部高等学校博士学科点专项科研基金项目(20123223110003)

摘　　要：提出了一种广义半迭代硬阈值追踪重建算法,综合了广义硬阈值追踪重建算法和半迭代思想的优点,修正了目标函数寻求最优解的搜索方向,获得了多项式的加速收敛,且不需要已知信号稀疏度.数值仿真结果表明,该算法在重构概率、峰值信噪比、信噪比、匹配度等方面的性能明显提高,且在"鬼"成像中的应用性能明显优于广义硬阈值追踪算法.A recovery algorithm,named generalized semi-iterative hard thresholding pursuit,was proposed. The algorithm modifies the searching direction of objective function with semi-iterative idea,has a polynomial acceleration convergence,and can be used for the signal without sparse information. It is shown that the proposed algorithm did improve the performance in term of reconstruction probability,peak signal-to-noise ratio,signal-to-noise ratio,and matching rate. The reconstruction quality in ghost imaging with the proposed algorithm is superior to that with the generalized hard thresholding method.

关 键 词：重构算法 硬阈值追踪 半迭代 “鬼”成像 

分 类 号：TN911.7[电子电信—通信与信息系统]