An enhanced synthetic oversampling framework with self-supervised contrastive learning for multi-class image imbalance
Class imbalance significantly affects the performance of machine learning and deep learning classifiers, especially in image recognition tasks where certain classes are underrepresented. Traditional synthetic oversampling methods, while helpful, often fail to address the complexities of real-world d...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | en |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://ir.uitm.edu.my/id/eprint/132623/1/132623.pdf https://ir.uitm.edu.my/id/eprint/132623/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|