RS-MetaNet: Deep meta metric learning for few-shot remote sensing scene classification
This addresses the challenge of classifying remote sensing scenes with very limited labeled data, which is crucial for applications like environmental monitoring, but the approach is incremental as it builds on meta-learning concepts in a domain-specific context.
The paper tackles the problem of few-shot remote sensing scene classification, where existing methods overfit due to sample-level learning, and proposes RS-MetaNet, a meta-learning approach that organizes training at the task level and introduces a Balance Loss function to improve generalization. The method achieves state-of-the-art results on three datasets with only 1-20 labeled samples.
Training a modern deep neural network on massive labeled samples is the main paradigm in solving the scene classification problem for remote sensing, but learning from only a few data points remains a challenge. Existing methods for few-shot remote sensing scene classification are performed in a sample-level manner, resulting in easy overfitting of learned features to individual samples and inadequate generalization of learned category segmentation surfaces. To solve this problem, learning should be organized at the task level rather than the sample level. Learning on tasks sampled from a task family can help tune learning algorithms to perform well on new tasks sampled in that family. Therefore, we propose a simple but effective method, called RS-MetaNet, to resolve the issues related to few-shot remote sensing scene classification in the real world. On the one hand, RS-MetaNet raises the level of learning from the sample to the task by organizing training in a meta way, and it learns to learn a metric space that can well classify remote sensing scenes from a series of tasks. We also propose a new loss function, called Balance Loss, which maximizes the generalization ability of the model to new samples by maximizing the distance between different categories, providing the scenes in different categories with better linear segmentation planes while ensuring model fit. The experimental results on three open and challenging remote sensing datasets, UCMerced\_LandUse, NWPU-RESISC45, and Aerial Image Data, demonstrate that our proposed RS-MetaNet method achieves state-of-the-art results in cases where there are only 1-20 labeled samples.