会议论文

Mei, Shaohui

英文

2017 INTERNATIONAL CONFERENCE ON DIGITAL IMAGE COMPUTING - TECHNIQUES AND APPLICATIONS (DICTA)

2017

695-701

International Conference on Digital Image Computing - Techniques and Applications (DICTA)

NOV 29-DEC 01, 2017

Sydney, AUSTRALIA

[Ji, Jingyu;Mei, Shaohui;Liu, Xiao;Li, Xu] Northwestern Polytech Univ, Sch Elect & Informat, Xian, Shaanxi, Peoples R China.^[Zeng, Shan] Wuhan Polytech Univ, Coll Math & Comp Sci, Wuhan, Hubei, Peoples R China.^[Wang, Zhiyong] Univ Sydney, Sch Informat Technol, Sydney, NSW 2006, Australia.

IEEE, Australian Govt, Dept Def, Def Sci & Technol Grp, Canon Informat Syst Res Australia Pty Ltd, Univ Sydney, APRS, IAPR

Guo, Y Li, H Cai, W Murshed, M Wang, Z Gao, J Feng, DD

345 E 47TH ST, NEW YORK, NY 10017 USA

IEEE

978-1-5386-2839-3

National Natural Science Foundation of ChinaNational Natural Science Foundation of China (NSFC) [61671383, 61301195]; Fundamental Research Funds for the Central UniversitiesFundamental Research Funds for the Central Universities [3102016ZB012, 3102016ZB029]; Hubei natural science foundationNatural Science Foundation of Hubei Province [2017CF-B500]; ARC grantsAustralian Research Council

本校为其他机构

Convolutional Neural Network (CNN) has achieved remarkable progresses in hyperspectral image (HSI) classification. However, how to effectively implement spatial context that has been demonstrated to he useful for HSI classification is still an open issue. Existing CNNs for hyperspectral classification are restricted into a small scale due to small-scale inputs and limited training samples. Therefore, in this paper, two different methods are proposed to integrate both spatial context and spectral signature into CNN based HSI classification: I). fixed kernels in which weights are determined by p...