独立成分分析研究综述
n Overview of independent component analysis research
独立成分分析(ICA)是人工智能理论中一种新的数据处理方法，目的在于从未知源信号的观测混合信号中分离或抽取相互统计独立的源信号。将ICA用来处理盲源分离问题，已成功地应用于语音信号处理、通信、人脸识别、图像特征提取、神经计算和医学信号处理等众多领域。本文简要介绍和评述了ICA的产生背景、定义、分类，并且就独立成分分析、盲源分离的若干算法及其应用进行了一些研究。
Independent component analysis (ICA) is a new data processing method in artificial intelligence theory and its purpose is separating independent source signals from an unknown mixed source. ICA is used to deal with separation of blind source and has been successfully applied to voice signal processing, communications, face recognition, image feature extraction, neural computing and medical signal processing and other fields. This paper introduces and reviews the background, definition, classification of ICA, and study on the algorithms and applications of independent component analysis and separation of blind source.
王旭、丁世飞
计算技术、计算机技术通信电子技术应用
人工智能理论独立成分分析盲源分离盲信号处理极大似然估计极大后验估计
artificial intelligence theoryindependent component analysisseparation of blind sourceblind signal processingmaximum similarity estimationmaximum posterior estimation
王旭,丁世飞.独立成分分析研究综述[EB/OL].(2011-03-03)[2025-08-16].http://www.paper.edu.cn/releasepaper/content/201103-105.点此复制
评论