| 引用本文： |
-
董江山,李成范,赵俊娟,等.基于变分贝叶斯ICA的遥感图像混合像元分析[J].电讯技术，2013，53（10）： - . [点击复制]
- DONG Jiang-shan,LI Cheng-fan,ZHAO Jun-juan,et al.Mixed pixel analysis of remote sensing image based on variational Bayesian ICA method[J].，2013，53（10）： - . [点击复制]
|
|
| 摘要: |
| 混合像元已成为遥感图像处理、分类的难点和重点。独立分量分析(ICA)能够实现图像的去
相关性以及得到相互独立的分量，但是，由于ICA模型的各成分独立性和数据统计分布规律
的不变假设，影响了遥感图像分类精度。针对这一问题，提出了基于变分贝叶斯ICA(VBICA)
的遥感
图像分析方法，并利用遥感图像进行验证，结果表明：VBICA方法提取的独立分量具有均方
根误差小、迭代次数少和稳定性较好的特点；基于VBICA方法的遥感分类精度达到了9155%
，且目视效果较好；VBICA方法突破了ICA的局限性，提高了遥感图像自动分类精度，具有很
好的应用前景。 |
| 关键词: 遥感图像 混合像元 独立分量分析 FastICA 变分贝叶斯ICA 自动分类 |
| DOI： |
|
| 基金项目:国家自然科学基金资助项目（41172303） |
|
| Mixed pixel analysis of remote sensing image based on variational Bayesian ICA method |
| DONG Jiang-shan,LI Cheng-fan,ZHAO Jun-juan,YIN Jing-yuan,SHEN Di,XUE Dan |
| () |
| Abstract: |
| The mixed pixels have become the difficulty and keystone of remote sen
sing image analysis and classification. Independent component analysis (ICA) not
only removes the correlation, but also can obtain mutual independent band image
s. However, the restrictions of independence and the fixed statistical distribut
ion rule of ICA model itself affect the classification accuracy of remote sensin
g images. In order to overcome this problem, the variational Bayesian ICA (VBICA
) method is proposed to analyze remote sensing images, and it is verified by
the true remote sensing images. The empirical results show that: the independen
t components extracted by the VBICA method is characterized by less RMSE, i
teration numbers and good stability; the classification accuracy of remote sensi
ng images reaches 9155%, and the method has preferable visual effect; the VBIC
A method has a big breakthrough on the limitation of traditional ICA method and
improves the automatic classification accuracy of remote sensing images, and has a good application prospect. |
| Key words: remote sensing image mixed pixel independent component analysis(ICA) FastICA var
iational Bayesian ICA automatic classification |
《电讯技术》编辑部 