A Novel Method for Hyperspectral Image Classification Based on Laplacian Eigenmap Pixels Distribution-Flow
A Novel Method for Hyperspectral Image Classification Based on Laplacian Eigenmap Pixels Distribution-Flow
复制标题
基于拉普拉斯特征图像素分布流的高光谱图像分类新方法
DOI：
10.1109/jstars.2013.2259470
复制
发表时间：
2013-05
影响因子：
5.5
通讯作者：
Zheng, Yaoguo
中科院分区：
文献类型：
--
作者：
Hou, Biao;Zhang, Xiangrong;Ye, Qiang;Zheng, Yaoguo
高光谱图像的准确分类是许多实际应用的重要任务。提出了一种新的基于流形学习算法的高光谱图像分类方法，该方法主要有三个方面的贡献：1)提出了一种新的拉普拉斯特征映射像素分布流(LE PD-FLOW)方法，该方法构造了一种新的联合空间-像素特征距离(JSPCD)度量来提高分类精度，并结合光谱特征和空间特征使用了合适的加权因子来区分不同类别的数据点；2)给出了每个流形映射的调整策略，不仅可以更好地显示结果，而且可以将映射结果与适当的度量进行比较；3)针对小尺度和大尺度的分类问题，分别提出了单阈值方法和多阈值方法，以获取可用于分类的有用边界点。通过调整这两种高光谱图像特征的权重，可以很容易地得到预期的分类结果。利用LE PD-FLOW可以发现分类边界上像素点的变化，从而可以高精度地对高光谱数据进行标记。实验结果表明，该方法对高光谱图像的分类是有效的。
The accurate classification of hyperspectral images is an important task for many practical applications. In this paper, a new method for hyperspectral image classification is proposed based on manifold learning algorithm, The approach introduced here presents three major contributions: 1) a new Laplacian eigenmap pixels distribution-flow (LE PD-Flow) is proposed for hyperspectral image analysis, in which, a new joint spatial-pixel characteristics distance (JSPCD) measure is constructed to improve the accuracy of classification and a suitable weighting factor is used to distinguish data points of different classes by combining the spectral feature with the spatial feature; 2) the adjustment strategy of each manifold mappings is addressed, which allows not only better visualization of the results, but also the comparisons of mapping results with an appropriate measurement; 3) in order to get useful boundary points used for classification, single threshold and multiple thresholds method are presented to solve small scale and large scale classification problem, respectively. We can easily obtain the expected classification results by adjusting the weights of the two kinds of feature of hyperspectral image. With the LE PD-Flow, variation of pixels on the boundaries for classification can be found, and then hyperspectral data can be labeled with high accuracy. Experimental results show that the proposed method is effective for classification of hyperspectral image.
登录
查看更多内容
DOI：
10.1109/igarss.2005.1525342
发表时间：
2005-07
期刊：
Proceedings. 2005 IEEE International Geoscience and Remote Sensing Symposium, 2005. IGARSS '05.
影响因子：
--
作者：
T. Han;D. Goodenough
通讯作者：
T. Han;D. Goodenough
影响因子：
3.4
作者：
C. Bachmann;T. Donato
通讯作者：
C. Bachmann;T. Donato
DOI：
10.1109/tgrs.2009.2023983
发表时间：
2010-01-01
影响因子：
8.2
作者：
Bazi, Yakoub;Melgani, Farid
通讯作者：
Melgani, Farid
DOI：
10.1109/jstars.2012.2190266
发表时间：
2012-04-01
影响因子：
5.5
作者：
Braun, Andreas Ch;Weidner, Uwe;Hinz, Stefan
通讯作者：
Hinz, Stefan
DOI：
10.1109/jstsp.2010.2096797
发表时间：
2011-06-01
影响因子：
7.5
作者：
Hung, Chih-Cheng;Kulkarni, Sameer;Kuo, Bor-Chen
通讯作者：
Kuo, Bor-Chen