摘要
为了对新型的采样定理:压缩感知理论CS(Compressed Sensing)进行深入的研究并将其应用于图像压缩编码,对压缩感知中的贪婪重构算法进行综述并分析该类算法的优缺点。通过理论分析与仿真实验,对比各算法的性能与效率。该类算法中StOMP需要人为地进行参数设置,而人为设置的参数值往往使算法的重建效果较差。针对该问题,利用粒子群优化算法对StOMP中的参数进行配置,以此来提高StOMP的重建效果。实验表明经过参数配置后的StOMP算法在重构效果上平均提高了2.62 dB,最大能提高13.63 dB。
In order to carry out in-depth research on novel sampling theorem： the compressed sensing（ CS）,and to apply it to image compression coding,we give an overview on the greedy reconstruction algorithms in CS and analyse the advantages and disadvantages of each algorithm. According to theoretical analysis and simulation experiments,we give a comparison on performances and efficiencies of each algorithm.Among them,the StOMP algorithm shall set parameter values artificially,which usually lowers down the reconstruction effect of the algorithm.To solve this problem,we use PSO algorithm to configure the parameters of StOMP so as to improve the reconstruction quality of the algorithm.Experiment illustrates that the StOMP algorithm with parameters configured can increase the reconstruction effect by 2. 62 dB in average and the maximum improvement reaches 13. 63 dB.
出处
《计算机应用与软件》
CSCD
2016年第4期258-261,285,共5页
Computer Applications and Software
基金
广东省教育厅高等院校学科建设专项资金项目(12ZK0362)
关键词
压缩感知
重构算法
匹配追踪
粒子群算法
Compressed sensing
Reconstruction algorithm
Matching pursuit
Particle swarm optimisation（PSO）
