Multiscale Deep Spatial Feature Extraction Using Virtual RGB Image for Hyperspectral Imagery Classification

• Published in: Remote sensing (Basel, Switzerland) Vol. 12; no. 2; p. 280
• Main Authors: Liu, Liqin, Shi, Zhenwei, Pan, Bin, Zhang, Ning, Luo, Huanlin, Lan, Xianchao
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.01.2020
• Subjects: Annotations; Classification; Deep learning; Feature extraction; fully convolutional networks (fcn); hyperspectral image classification; hyperspectral imagery; Hyperspectral imaging; Image classification; Machine learning; multiscale spatial feature; Networks; Neural networks; Pixels; Principal components analysis; processing technology; Remote sensing; Semantics; Support vector machines; virtual rgb image

In recent years, deep learning technology has been widely used in the field of hyperspectral image classification and achieved good performance. However, deep learning networks need a large amount of training samples, which conflicts with the limited labeled samples of hyperspectral images. Traditio...