MSNet: A Multiple Supervision Network for Remote Sensing Scene Classification

Nanqing Liu

doi:10.1109/LGRS.2020.3043020

MSNet: A Multiple Supervision Network for Remote Sensing Scene Classification

2022-01-01

6

PRE

AI

DOI：10.1109/LGRS.2020.3043020

Original

Original request for help

Save

Abstract

En 中文

Remote sensing scene classification is a complex task due to large intraclass variations in object appearances with a small number of samples per class and high interclass similarities due to shared objects in different classes, which usually cause model overfitting and high interclass confusion. To address these challenges, a multiple supervision approach, called multiple supervision network (MSNet), consisting of the ResNet-50 backbone, a feature discriminative branch (FDB), and a feature confusion branch (FCB) is proposed in this letter. The FDB selects discriminative features per class and suppresses peaks in feature maps to examine more informative regions with lower feature magnitudes. Meanwhile, the FCB reduces overfitting by introducing confusion to the input of a fully connected layer which also enhances the robust features. The FDB and FCB are only used in training of the backbone and not used in inference. Thus, the proposed method does not introduce additional computing time on the backbone while it significantly boosts its performance in scene classification. The experimental results show that MSNet outperforms the methods considered in this letter.

Keywords:

Training

Remote sensing

Generative adversarial networks

Feature extraction

Gallium nitride

Visualization

Information science

Convolutional neural network (CNN)

deep learning

generative adversarial network (GAN)

remote sensing scene classification (RSSC)

MSNet: A Multiple Supervision Network for Remote Sensing Scene Classification

Abstract

AI Summary

Journal

Organization