Spectral-Spatial Hyperspectral Image Classification with Superpixel Pattern and Extreme Learning Machine

Spectral-Spatial Hyperspectral Image Classification with Superpixel Pattern and Extreme Learning Machine
复制标题

具有超像素模式和极限学习机的谱空间高光谱图像分类

DOI:
10.3390/rs11171983
复制
发表时间:
2019-08
期刊:
影响因子:
5
通讯作者:
Cai Zhihua
Cai Zhihua
中科院分区:
工程技术2区
文献类型:
--
作者:
Zhang Yongshan;Jiang Xinwei;Wang Xinxin;Cai Zhihua

文献摘要

参考文献

被引文献

相似文献

高光谱图像的光谱-空间分类是近年来遥感领域的研究热点。众所周知,在遥感应用中,光谱特征是基本信息,空间模式提供补充信息。同时利用光谱特征和空间模式,可以充分挖掘高光谱图像的应用价值,提高分类性能。在实际应用中,空间模式可以被提取出来表示一条线、一组点或图像纹理,它们代表了HSI的局部或全局空间特征。在本文中,我们提出了一个光谱-空间HSI分类模型的基础上超像素模式(SP)和基于核的极端学习机(KELM),称为SP-KELM,识别HSI中的像素的土地覆盖。在SP-KELM模型中,超像素模式特征提取的高级主成分分析(PCA),这是基于超像素分割的HSI和用于表示空间信息。然后,KELM方法是一个分类器,在建议的光谱空间模型与原始光谱特征和提取的空间模式特征。在三个公开的HSI数据集上的实验结果验证了所提出的SP-KELM模型的有效性,与谱方法相比,性能提高了10%。
Spectral-spatial classification of hyperspectral images (HSIs) has recently attracted great attention in the research domain of remote sensing. It is well-known that, in remote sensing applications, spectral features are the fundamental information and spatial patterns provide the complementary information. With both spectral features and spatial patterns, hyperspectral image (HSI) applications can be fully explored and the classification performance can be greatly improved. In reality, spatial patterns can be extracted to represent a line, a clustering of points or image texture, which denote the local or global spatial characteristic of HSIs. In this paper, we propose a spectral-spatial HSI classification model based on superpixel pattern (SP) and kernel based extreme learning machine (KELM), called SP-KELM, to identify the land covers of pixels in HSIs. In the proposed SP-KELM model, superpixel pattern features are extracted by an advanced principal component analysis (PCA), which is based on superpixel segmentation in HSIs and used to denote spatial information. The KELM method is then employed to be a classifier in the proposed spectral-spatial model with both the original spectral features and the extracted spatial pattern features. Experimental results on three publicly available HSI datasets verify the effectiveness of the proposed SP-KELM model, with the performance improvement of 10% over the spectral approaches.
DOI: 10.1109/jstsp.2015.2417833
发表时间: 2015-09-01
影响因子: 7.5
作者:
Laparra, Valero;Malo, Jesus;Camps-Valls, Gustau
通讯作者: Camps-Valls, Gustau
DOI: 10.3390/rs9020139
发表时间: 2017-02
期刊: Remote. Sens.
影响因子: --
作者:
Shuzhen Zhang;Shutao Li;Wei Fu;Leyuan Fang
通讯作者: Shuzhen Zhang;Shutao Li;Wei Fu;Leyuan Fang
DOI: 10.1109/lgrs.2014.2380313
发表时间: 2015-01
影响因子: 4.8
作者:
T. Priya;S. Prasad;Hao Wu
通讯作者: T. Priya;S. Prasad;Hao Wu
DOI: 10.1109/igarss.2011.6049567
发表时间: 2011-07
期刊: 2011 IEEE International Geoscience and Remote Sensing Symposium
影响因子: --
作者:
Md. Ali Hossain;M. Pickering;X. Jia
通讯作者: Md. Ali Hossain;M. Pickering;X. Jia
DOI: 10.1109/igarss.2013.6723581
发表时间: 2013-07
期刊: 2013 IEEE International Geoscience and Remote Sensing Symposium - IGARSS
影响因子: --
作者:
Shanshan Li;X. Jia;Bing Zhang
通讯作者: Shanshan Li;X. Jia;Bing Zhang