聊城大学学报：自然科学版
聊城大學學報：自然科學版
료성대학학보：자연과학판
Journal of Liaocheng University:Natural Science Edition
2012年
4期
5-10
,共6页
正则化MFA%目标函数为减式的MFA%基于零空间的MFA%AVIRIS遥感图像
正則化MFA%目標函數為減式的MFA%基于零空間的MFA%AVIRIS遙感圖像
정칙화MFA%목표함수위감식적MFA%기우령공간적MFA%AVIRIS요감도상
regularized MFA%MFA with subtractive objective function%MFA based on the null space%AVIRIS remote sensing image
在间隔Fisher分析（MFA）的基础上，提出了三种改进算法：正则化MFA，目标函数为减式的MFA和基于零空间的MFA．为了检验算法的有效性和先进性，针对AVIRIS遥感图像进行了实验．实验结果表明，所提算法均优于MFA．
在間隔Fisher分析（MFA）的基礎上，提齣瞭三種改進算法：正則化MFA，目標函數為減式的MFA和基于零空間的MFA．為瞭檢驗算法的有效性和先進性，針對AVIRIS遙感圖像進行瞭實驗．實驗結果錶明，所提算法均優于MFA．
재간격Fisher분석（MFA）적기출상，제출료삼충개진산법：정칙화MFA，목표함수위감식적MFA화기우령공간적MFA．위료검험산법적유효성화선진성，침대AVIRIS요감도상진행료실험．실험결과표명，소제산법균우우MFA．
Three kind of improved algorithms, based on marginal Fisher analysis (MFA), are proposed in this paper, which are regularized marginal Fisher analysis, marginal Fisher analysis with subtractive objective function and marginal Fisher analysis based on the null space. Experiments on AVIRIS remote sensing image are made in order to test the validity and advantage of the methods presented in this paper. The experimental results demonstrate that our proposed algorithms are better than MFA.
