Active Ensemble Deep Learning for Polarimetric Synthetic Aperture Radar Image Classification

• Published in: IEEE geoscience and remote sensing letters Vol. 18; no. 9; pp. 1580 - 1584
• Main Authors: Liu, Sheng-Jie, Luo, Haowen, Shi, Qian
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.09.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Active learning; Classification; Convergence; convolutional neural network (CNN); Deep learning; ensemble learning; Hyperspectral imaging; Image classification; Machine learning; Neural networks; Radar; Radar imaging; Radar polarimetry; Remote sensing; SAR (radar); Synthetic aperture radar; Training; Training data
• ISSN: 1545-598X; 1558-0571
• DOI: 10.1109/LGRS.2020.3005076

Although deep learning has achieved great success in the image-classification tasks, its performance is subject to the quantity and quality of the training samples. For the classification of the polarimetric synthetic aperture radar (PolSAR) images, it is nearly impossible to annotate the images from visual interpretation. Therefore, it is urgent for remote-sensing scientists to develop new techniques for PolSAR image classification under the condition of very few training samples. In this letter, we take the advantage of active learning and propose active ensemble deep learning (AEDL) for PolSAR image classification. We first show that only 35% of the predicted labels of the deep-learning model's snapshots near its convergence were exactly the same. The disagreement between the snapshots is nonnegligible. From the perspective of multiview learning, the snapshots together serve as a good committee to evaluate the importance of the unlabeled instances. Using the snapshot committee to give out the informativeness of the unlabeled data, the proposed AEDL achieved better performance on two real PolSAR images than the standard active learning strategies. It achieved the same classification accuracy with only 86% and 55% of the training samples compared to the breaking tie active learning and random selection for the Flevoland data set.