Feature-Space Oversampling for Addressing Class Imbalance in SAR Ship Classification

TL;DR

This paper introduces two novel feature-space oversampling algorithms to improve SAR ship classification on imbalanced datasets, demonstrating significant F1-score improvements across multiple models and datasets.

Contribution

The paper proposes two new oversampling algorithms inspired by M2m, tailored for SAR data, and evaluates their effectiveness on public datasets with multiple models.

Findings

Significant F1-score improvements with the new methods.

Effective across different models and datasets.

Enhanced classification of underrepresented classes.

Abstract

SAR ship classification faces the challenge of long-tailed datasets, which complicates the classification of underrepresented classes. Oversampling methods have proven effective in addressing class imbalance in optical data. In this paper, we evaluated the effect of oversampling in the feature space for SAR ship classification. We propose two novel algorithms inspired by the Major-to-minor (M2m) method M2m$_f$, M2m$_u$. The algorithms are tested on two public datasets, OpenSARShip (6 classes) and FuSARShip (9 classes), using three state-of-the-art models as feature extractors: ViT, VGG16, and ResNet50. Additionally, we also analyzed the impact of oversampling methods on different class sizes. The results demonstrated the effectiveness of our novel methods over the original M2m and baselines, with an average F1-score increase of 8.82% for FuSARShip and 4.44% for OpenSARShip.