基于改进的独立分量分析的图像分离技术  

Research of image separation based on improved independent component analysis

作　　者：王翠茹[1] 杜鹃[1] 张鑫林[1] 

机构地区：[1]华北电力大学计算机科学与技术系,保定071003

出　　处：《仪器仪表学报》2006年第z1期785-786,797,共3页Chinese Journal of Scientific Instrument

摘　　要：本文主要将改进的ICA优秀算法优化并运用于多通道混合图像分离与处理中,通过同基于特征分析的线性变换技术的对比,以及实验并探讨噪声对混合图像的ICA分离的影响,取得了一些有价值的仿真结果。It plays an important part in image processing and becomes more and more essential.Some improved advance ICA algorithms were implemented in our experiment of image separation.Then we got some useful data and results about ICA in image separation experiment though the contrast to the algorithms based on linearly transformed technique of FA(Factor Analysis) and discussed the influence of noise added to the mixed images.

关 键 词：独立分量分析 ERICA FPICA 累积量 峭度 

分 类 号：TH7-55[机械工程—仪器科学与技术]