TY - GEN
T1 - Lahnet
T2 - 38th Annual IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018
AU - Hua, Yuansheng
AU - Mou, Lichao
AU - Zhu, Xiao Xiang
N1 - Publisher Copyright:
© 2018 IEEE.
PY - 2018/10/31
Y1 - 2018/10/31
N2 - In this paper, we proposed an innovative end-to-end convolutional neural network (CNN), which is trained to learn how to fuse multi-level features for aerial scene classification. Instead of using only coarse semantic features as conventional CNNs, we resort to first hierarchically extracting dense highlevel features and then element-wise fusing them with lowlevel features to build a comprehensive feature representation, which contains not only high-level semantic information but also fine-grained low-level details, for scene classification. The network is evaluated on two broadly used aerial scene datasets, UCM and AID. The experimental results indicate that the proposed LAHNet performs superiorly compared to the existing benchmark methods. Furthermore, visualization of the fused features presents an intuitive illustration of the remarkable improvement.
AB - In this paper, we proposed an innovative end-to-end convolutional neural network (CNN), which is trained to learn how to fuse multi-level features for aerial scene classification. Instead of using only coarse semantic features as conventional CNNs, we resort to first hierarchically extracting dense highlevel features and then element-wise fusing them with lowlevel features to build a comprehensive feature representation, which contains not only high-level semantic information but also fine-grained low-level details, for scene classification. The network is evaluated on two broadly used aerial scene datasets, UCM and AID. The experimental results indicate that the proposed LAHNet performs superiorly compared to the existing benchmark methods. Furthermore, visualization of the fused features presents an intuitive illustration of the remarkable improvement.
KW - Aerial scene classification
KW - Convolutional neural network (CNN)
KW - Feature fusion
UR - http://www.scopus.com/inward/record.url?scp=85060983215&partnerID=8YFLogxK
U2 - 10.1109/IGARSS.2018.8519576
DO - 10.1109/IGARSS.2018.8519576
M3 - Conference contribution
AN - SCOPUS:85060983215
T3 - International Geoscience and Remote Sensing Symposium (IGARSS)
SP - 4728
EP - 4731
BT - 2018 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 22 July 2018 through 27 July 2018
ER -