TY - GEN
T1 - A Meta-Learning Framework for Few-Shot Classification of Remote Sensing Scene
AU - Bai, Cloud
N1 - Funding Information:
∗Ying Li is the corresponding author. This work was supported in part by the National Natural Science Foundation of China under Grant 61871460; in part by the Shaanxi Provincial Key Research and Development Program under Grant 2020KW-003; in part by the Fundamental Research Funds for the Central Universities under Grant 3102019ghxm016.
Publisher Copyright:
© 2021 IEEE
PY - 2021/6/6
Y1 - 2021/6/6
N2 - While achieving remarkable success in remote sensing (RS) scene classification for the past few years, convolutional neural network (CNN) based methods suffer from the demand for large amounts of training data. The bottleneck in prediction accuracy has shifted from data processing limits toward a lack of ground truth samples, usually collected manually by experienced experts. In this work, we provide a meta-learning framework for few-shot classification of RS scene. Under the umbrella of meta-learning, we show it is possible to learn much information about a new category from only 1 or 5 samples. The proposed method is based on Prototypical Networks with a pre-trained stage and a learnable similarity metric. The experimental results show that our method outperforms three state-of-the-art few-shot algorithms and one typical CNN-based method, D-CNN, on two challenging datasets: NWPU-RESISC45 and RSD46-WHU.
AB - While achieving remarkable success in remote sensing (RS) scene classification for the past few years, convolutional neural network (CNN) based methods suffer from the demand for large amounts of training data. The bottleneck in prediction accuracy has shifted from data processing limits toward a lack of ground truth samples, usually collected manually by experienced experts. In this work, we provide a meta-learning framework for few-shot classification of RS scene. Under the umbrella of meta-learning, we show it is possible to learn much information about a new category from only 1 or 5 samples. The proposed method is based on Prototypical Networks with a pre-trained stage and a learnable similarity metric. The experimental results show that our method outperforms three state-of-the-art few-shot algorithms and one typical CNN-based method, D-CNN, on two challenging datasets: NWPU-RESISC45 and RSD46-WHU.
KW - Few-shot learning
KW - Meta-learning
KW - Remote sensing
KW - Scene classification
UR - http://www.scopus.com/inward/record.url?scp=85115173609&partnerID=8YFLogxK
U2 - 10.1109/icassp39728.2021.9413971
DO - 10.1109/icassp39728.2021.9413971
M3 - Conference Proceeding (Non-Journal item)
VL - 2021-June
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 4590
EP - 4594
BT - ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
ER -