文献类型：期刊文献

英文题名：Hyperspectral classification based on spectral-spatial convolutional neural networks

作者：Chen, Congcong[1,2];Jiang, Feng[2];Yang, Chifu[2];Rho, Seungmin[3];Shen, Weizheng[1];Liu, Shaohui[2];Liu, Zhiguo[4]

第一作者：Chen, Congcong

通讯作者：Shen, WZ[1]

机构：[1]Northeast Agr Univ, Sch Elect Engn & Informat, Harbin 150030, Heilongjiang, Peoples R China;[2]Harbin Inst Technol, Harbin 150001, Heilongjiang, Peoples R China;[3]Sungkyul Univ, Dept Media Software, Anyang, South Korea;[4]Beijing Union Univ, Coll Informat Technol, Beijing 100101, Peoples R China

第一机构：Northeast Agr Univ, Sch Elect Engn & Informat, Harbin 150030, Heilongjiang, Peoples R China

通讯机构：[1]corresponding author), Northeast Agr Univ, Sch Elect Engn & Informat, Harbin 150030, Heilongjiang, Peoples R China.

年份：2018

卷号：68

起止页码：165-171

外文期刊名：ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE

收录：;EI(收录号：20174904489569);Scopus(收录号：2-s2.0-85035754548);WOS:【SCI-EXPANDED(收录号:WOS:000423894400015)】；

基金：This work is partially funded by the MOE-Microsoft Key Laboratory of Natural Language Processing and Speech, Harbin Institute of Technology (Grant No. KM20006074), the National Key Research and Development Program of China (2016YED0700204-02), the Major State Basic Research Development Program of China (973 Program 2015CB351804) and the National Natural Science Foundation of China under Grant No. 61572155, 61672188 and 61272386. We would also like to acknowledge NVIDIA Corporation who kindly provided two sets of GPU.

语种：英文

外文关键词：Hyperspectral classification; Convolutional neural network; Support vector machine; Spectral-spatial convolutional neural network; Adaptive window

摘要：Hyperspectral image classification is an important task in remote sensing image analysis. Traditional machine learning techniques are difficult to deal with hyperspectral images directly, because hyperspectral images have too many redundant spectral channels. In this paper we propose a novel method for hyperspectral image classification, by which spectral and spatial features are jointly exploited from hyperspectral images. Firstly, considering the local similarity in spatial domain, we employ a large spatial window to get image blocks from hyperspectral image Secondly, each spectral channel of the image block is filtered to extract their spatial and spectral features, after that the features are merged by convolutional layers. Finally, the fully-connected layers are used to get the classification result. Comparing with other state-of-the-art techniques, the proposed method pays more attention to the correlation of spatial neighborhood by using a large spatial window in the network. In addition, we combine the proposed network with the traditional support vector machine (SVM) classifier to improve the performance of hyperspectral image classification. Moreover, an adaptive method of the spatial window sizes selection is proposed in this paper. Experimental results conducted on the AVIRIS and ROSIS datasets demonstrate that the proposed method outperforms the state-of-the-art techniques. (C) 2017 Elsevier Ltd. All rights reserved.