论文部分内容阅读
稀疏化是压缩感知理论的关键,对信号恰当的稀疏表示能提高恢复的精度.本文提出一种基于参数字典的稀疏表示方法,把参数字典设计作为一个优化问题来分析,通过交替迭代的方式求得参数方程的可行解,进而生成参数字典.本文的参数字典设计方法较其他方法而言能获得较优的近似解,且该方法产生的优化字典更符合紧框架特性.
Sparseness is the key of compressive sensing theory, and the appropriate sparse representation of signal can improve the precision of restoration.In this paper, we propose a sparse representation method based on parameter dictionary, and analyze the design of parameter dictionary as an optimization problem, by means of alternating iteration Obtain the feasible solution of the parametric equation, and then generate the parameter dictionary.The parameter dictionary design method in this paper can obtain better approximate solution than other methods, and the optimized dictionary produced by this method is more consistent with the tight frame characteristic.