TY - GEN
T1 - Hyperspectral Image Classification Based on Long Short Term Memory Network
AU - Liu, Jin
AU - Zhang, Xiangrong
AU - Zhang, Jingyan
AU - An, Jinliang
AU - Li, Chen
AU - Gao, Li
N1 - Publisher Copyright:
© 2018 IEEE.
PY - 2018/12/31
Y1 - 2018/12/31
N2 - In the task of hyperspectral image classification, how to learn features of hyperspectral image is the important and difficulty issue which may directly affect the classification results. Inspired by the idea of natural language processing, in this paper, we propose a local space long short-term memory network based hyperspectral image classification, which constructs sequential features in the local area of hyperspectral images. This method is based on the integration features of two traditional low-level features, and from these integration features to extract sequential features of the center sample in the local space, then use the long short-term memory network to learn high-level semantic features, finally use them to classify image. This method can not only obtain more representative and discriminative high-level semantic features, and through constructing the local space sequence to enhance positive impact of the useful pixels, inhibit negative effects of useless pixels, it improves the classification accuracy.
AB - In the task of hyperspectral image classification, how to learn features of hyperspectral image is the important and difficulty issue which may directly affect the classification results. Inspired by the idea of natural language processing, in this paper, we propose a local space long short-term memory network based hyperspectral image classification, which constructs sequential features in the local area of hyperspectral images. This method is based on the integration features of two traditional low-level features, and from these integration features to extract sequential features of the center sample in the local space, then use the long short-term memory network to learn high-level semantic features, finally use them to classify image. This method can not only obtain more representative and discriminative high-level semantic features, and through constructing the local space sequence to enhance positive impact of the useful pixels, inhibit negative effects of useless pixels, it improves the classification accuracy.
KW - deep learning
KW - hyperspectral image
KW - long short term memory network
KW - semantic features
UR - https://www.scopus.com/pages/publications/85061777973
U2 - 10.1109/EORSA.2018.8598645
DO - 10.1109/EORSA.2018.8598645
M3 - 会议稿件
AN - SCOPUS:85061777973
T3 - 5th International Workshop on Earth Observation and Remote Sensing Applications, EORSA 2018 - Proceedings
BT - 5th International Workshop on Earth Observation and Remote Sensing Applications, EORSA 2018 - Proceedings
A2 - Weng, Qihao
A2 - Gamba, Paolo
A2 - Chang, Ni-Bin
A2 - Wang, Guangxing
A2 - Yao, Wanqiang
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 5th International Workshop on Earth Observation and Remote Sensing Applications, EORSA 2018
Y2 - 18 June 2018 through 20 June 2018
ER -