全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

基于新颖子波变换的高光谱遥感图像特征提取

DOI: 10.11834/jig.20091019

Keywords: 高光谱影像,光谱特征,子波变换,Q准则,特征提取

Full-Text   Cite this paper   Add to My Lib

Abstract:

提出了一种新颖的用于高光谱遥感图像特征提取的子波变换算法。与二进小波变换按恒Q准则划分频域不同的是,该算法通过改变相邻子波的带宽比,可以实现更为灵活的频域划分。采用子波能量的离散余弦变换作为特征矢量,然后进行无监督C均值聚类实验和有监督RBF(径向基函数)神经网络分类实验。实验结果表明,子波变换能量的离散余弦变换特征可以有效地描述光谱曲线特征,且正确分类率高于传统的小波变换。

References

[1]  Li Jiang.Liner Unmixing of Hyperspectral Signals Via Wavelet Feature Extraction[D].Mississippi State,USA,Mississippi State University,2002:18-24.
[2]  Benediktsson J A,Sveinssen J R,Kolbeinn Amos.Classification and feature extraction of AVIRIS data[A].In:Proceedings of International Conference on Geoscience and Remote Sensing[C],Florence,Italy,1995,3:1194-1205.
[3]  Hu Guang-shu.Digital Signal Processing-method,Algorithm and Realization[M].Beijing:Tsinghua University Press,1997:23-28.[胡广书.数字信号处理-理论、算法与实现[M].北京:清华大学出版社,1997:23-28.]
[4]  Wang Cheng,Menenti M,Li Zhao-liang.Modified principal component analysis for feature selection of hyperspectral imagery[A].In:Proceedings of International Conference on Geoscience and Remote Sensing[C],Toulouse,France,2003,6:3781-3783.
[5]  Mallat S.A Theory for multi-resolution signal decomposition:The wavelet representation[A].In:Proceedings of IEEE Transactions on Pattern Anaysis and Machine Intelligence.1989.11:674-693.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133