Translator Disclaimer
16 June 2020 Semisupervised hyperspectral imagery classification based on a three-dimensional convolutional adversarial autoencoder model with low sample requirements
Author Affiliations +
Abstract

Although there are many state-of-the-art methods for hyperspectral classification, data deficiency is a problem that should be addressed before popularizing hyperspectral technology. To solve this problem, it is worth exploring methods based on small datasets. Inspired by the advanced deep learning classification methods and the autoencoder structure, we propose a structure named three-dimensional convolutional adversarial autoencoder that combines the two processes for semisupervised hyperspectral classification. Our experiments show its utility in data-deficient situations, and our study analyzes its advantages and disadvantages, and points out a probable direction toward optimization.

© 2020 Society of Photo-Optical Instrumentation Engineers (SPIE) 1931-3195/2020/$28.00 © 2020 SPIE
Zeyu Cao, Xiaorun Li, and Liaoying Zhao "Semisupervised hyperspectral imagery classification based on a three-dimensional convolutional adversarial autoencoder model with low sample requirements," Journal of Applied Remote Sensing 14(2), 024522 (16 June 2020). https://doi.org/10.1117/1.JRS.14.024522
Received: 24 February 2020; Accepted: 1 June 2020; Published: 16 June 2020
JOURNAL ARTICLE
11 PAGES


SHARE
Advertisement
Advertisement
Back to Top