基于改进卷积神经网络的高光谱图像特征提取方法
详细信息 查看官网全文
- 英文论文题名：Improved Convolutional Neural Network Based Hyperspectral image feature extraction method
- 论文作者：沈飞 ; 曹鹏 ; 龙华保 ; 谢长生
- 英文论文作者：Fei Shen ; Peng Cao ; Huabao Long ; Changsheng Xie ; Shanghai Institute of Spaceflight Control Technology Infrared Detection Technology Research & Development Center ; CASC
- 年：2016
- 作者机构：上海航天控制技术研究所中国航天科技集团公司红外探测技术研发中心;
- 论文关键词：高光谱图像 ; 深度学习 ; 卷积网络 ; 稀疏约束
- 英文论文关键词：hyperspectral image ; deep learning ; convolutional neural network ; sparse constraint
- 会议召开时间：2016-12-06
- 会议录名称：2016年红外、遥感技术与应用研讨会暨交叉学科论坛论文集
- 语种：中文
- 分类号：TP183;TP751
- 学会代码：SHHH
- 会议名称：2016年红外、遥感技术与应用研讨会暨交叉学科论坛
- 会议地点：中国江苏常州
- 主办单位：上海市红外与遥感学会、上海市传感技术学会、中国科学院上海技术物理研究所
- 学会名称：上海市红外与遥感学会
- 页数：4
- 文件大小：474k
- 原文格式：D
- 会议级别：地方
摘要
针对高光谱复杂数据,提出了一种基于稀疏约束的深度卷积网络以处理高光谱图像中高维复杂像元特征提取与分类的方法。在卷积网络中引入无监督的稀疏特征学习有利于光谱数据特征的获得,从而有助于深度卷积网络能够更有效地提取出光谱数据中的特征信息。在实验中,该方法被应用在两个高光谱数据集中,通过所提取得特征信息,均获得了良好的分类效果。
According to the characters of complex hyperspectral data, the idea of sparsity is introduced to deep convolutional neural network(CNN) to handle feature extraction and classification problems. Combining sparse unsupervised learning method with neural network model, it is possible to get a good, sparse representation of the spectral information so that deep CNN model could extract feature information hierarchically and effectively. In the experiment, two hyperspectral data sets are applied for this simulation, and the results demonstrate fine classification performance by the features obtained by the proposed method.
According to the characters of complex hyperspectral data, the idea of sparsity is introduced to deep convolutional neural network(CNN) to handle feature extraction and classification problems. Combining sparse unsupervised learning method with neural network model, it is possible to get a good, sparse representation of the spectral information so that deep CNN model could extract feature information hierarchically and effectively. In the experiment, two hyperspectral data sets are applied for this simulation, and the results demonstrate fine classification performance by the features obtained by the proposed method.
引文
[1]WANG Li-guo,ZHAO Chun-hui.Hyperspectral Image Processing[M].Beijing:National Defense Industry Press(王立国,赵春晖.高光谱图像处理技术.北京:国防工业出版社),2013:1-33.
[2]HE Lin,PAN Quan,DI Wei,et al.Research Advance on Target Detection for Hyperspectral Imagery[J].Chinese Journal of Electronics(贺霖,潘泉,邸韡,李远清.高光谱图像目标检测研究进展.电子学报),2009,37(9):2016-2024.
[3]WU Chun-peng,FAN Wei,HE Yuan,et al.Handwritten Character Recognition by Alternately Trained Relaxation Convolutional Neural Network[C]//Proceedings of ICFHR,2014:291-296.
[4]CHEN Li,WANG Song,FAN Wei,et al.Reconstruction combined training for convolutional neural networks on character recognition[C]//Proceedings of ICDAR,2015:431–435.
[5]CAI Meng,SHI Yong-zhe,KANG Jian,et al.Convolutional maxout neural networks for low-resource speech recognition[C]//Proceedings of ISCSLP,2014:133-137.
[6]OSSAMA A H,MOHAMED A R,JIANG Hui,et al.Convolutional neural networks for speech recognition[J].IEEE Transactions on Audio,Speech,and Language Processing,2014,22(10):1533-1545.
[7]LAWRENCE S,GILES C L,TSOI A C,et al.Face recognition:A convolutional neural-networks approach[J].IEEE Transactions on Neural Networks,1997,8(1):98-113.
[8]ROMERO A,RADEVA P,GATTA C.Meta-parameter free unsupervisedsparse feature learning.[J].IEEE Transactions on Pattern Analysis Machine Intelligence,2015,37(8):1716-1722.
[9]ZHAO Meng-yuan,WANG Dong,ZHANG Zhi-yong,et al.Music removal by convolutional denoising autoencoder in speech recognition[J].Proceedings of APSIPA,2015:338-341.
[10]LI Zu-he,FAN Yang-yu,WANG Feng-qin.Convolutional Autoencoder-based Color Image Classification using Chroma Subsampling in YCb Cr Space[J].Proceedings of CISP,2015,351-355.
[11]REN Xiao-hang,CHEN Kai,YANG Xiao-kang,et al.A new unsupervised convolutional neural network model for Chinese scene text detection[C]//IEEE China Summit and International Conference on China SIP,2015:428–432.
[12]JIMENEZ R L O,ARZUAGA C E,VELEZ R M.Unsupervised linear feature-extraction methods and their effects in the classification of high-dimensional data[J].IEEE Transactions on Geoscienceand Remote Sensing,2007,45(2):469-483.
[13]LECUN Y,BOTTOU L,BENGIO Y,et al.Gradient-Based Learning Applied to Document Recognition[J].Proceedings of the IEEE,1998,86(11):2278-2324.
[14]LIANG Xi-jun.Some Applications of Sparse Optimization in Machine Learning[D].Dalian University of Technology(梁锡军.稀疏优化在机器学习中的若干应用.大连理工大学),2013:1-7.
[15]BEN W,TOLHURST D J.Characterizing the sparseness of neural codes[J].Network Computation in Neural Systems,2001,12(3):255-270.
[16]HU Fan,XIA Gui-song,HU Jing-wen,et al.Unsupervised Deep Feature Extraction for Remote Sensing Image Classification[J].Remote Sensing,2015,7(11):14680-14707.
[17]CHEN Yu-shi,LIN Zhou-han,ZHAO Xing,et al.Deep learning-based classification of hyperspectral data[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2014,7(6):2094-2107.
[2]HE Lin,PAN Quan,DI Wei,et al.Research Advance on Target Detection for Hyperspectral Imagery[J].Chinese Journal of Electronics(贺霖,潘泉,邸韡,李远清.高光谱图像目标检测研究进展.电子学报),2009,37(9):2016-2024.
[3]WU Chun-peng,FAN Wei,HE Yuan,et al.Handwritten Character Recognition by Alternately Trained Relaxation Convolutional Neural Network[C]//Proceedings of ICFHR,2014:291-296.
[4]CHEN Li,WANG Song,FAN Wei,et al.Reconstruction combined training for convolutional neural networks on character recognition[C]//Proceedings of ICDAR,2015:431–435.
[5]CAI Meng,SHI Yong-zhe,KANG Jian,et al.Convolutional maxout neural networks for low-resource speech recognition[C]//Proceedings of ISCSLP,2014:133-137.
[6]OSSAMA A H,MOHAMED A R,JIANG Hui,et al.Convolutional neural networks for speech recognition[J].IEEE Transactions on Audio,Speech,and Language Processing,2014,22(10):1533-1545.
[7]LAWRENCE S,GILES C L,TSOI A C,et al.Face recognition:A convolutional neural-networks approach[J].IEEE Transactions on Neural Networks,1997,8(1):98-113.
[8]ROMERO A,RADEVA P,GATTA C.Meta-parameter free unsupervisedsparse feature learning.[J].IEEE Transactions on Pattern Analysis Machine Intelligence,2015,37(8):1716-1722.
[9]ZHAO Meng-yuan,WANG Dong,ZHANG Zhi-yong,et al.Music removal by convolutional denoising autoencoder in speech recognition[J].Proceedings of APSIPA,2015:338-341.
[10]LI Zu-he,FAN Yang-yu,WANG Feng-qin.Convolutional Autoencoder-based Color Image Classification using Chroma Subsampling in YCb Cr Space[J].Proceedings of CISP,2015,351-355.
[11]REN Xiao-hang,CHEN Kai,YANG Xiao-kang,et al.A new unsupervised convolutional neural network model for Chinese scene text detection[C]//IEEE China Summit and International Conference on China SIP,2015:428–432.
[12]JIMENEZ R L O,ARZUAGA C E,VELEZ R M.Unsupervised linear feature-extraction methods and their effects in the classification of high-dimensional data[J].IEEE Transactions on Geoscienceand Remote Sensing,2007,45(2):469-483.
[13]LECUN Y,BOTTOU L,BENGIO Y,et al.Gradient-Based Learning Applied to Document Recognition[J].Proceedings of the IEEE,1998,86(11):2278-2324.
[14]LIANG Xi-jun.Some Applications of Sparse Optimization in Machine Learning[D].Dalian University of Technology(梁锡军.稀疏优化在机器学习中的若干应用.大连理工大学),2013:1-7.
[15]BEN W,TOLHURST D J.Characterizing the sparseness of neural codes[J].Network Computation in Neural Systems,2001,12(3):255-270.
[16]HU Fan,XIA Gui-song,HU Jing-wen,et al.Unsupervised Deep Feature Extraction for Remote Sensing Image Classification[J].Remote Sensing,2015,7(11):14680-14707.
[17]CHEN Yu-shi,LIN Zhou-han,ZHAO Xing,et al.Deep learning-based classification of hyperspectral data[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2014,7(6):2094-2107.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |