Underwater Image Feature Extraction and Classification Using a Multi-Scale Vision Transformer with Cross-Scale Biased Attention Fusion
Applied Sciences · 27.05.2026 · 16 (11) · 5358
DOI: 10.3390/app16115358 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 27.05.2026. Verfasst von Abdullah Faiz und 2 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: provincial Natural Science Foundation of Shanxi, National Nature Science Foundation of China, Shanxi Key Research and Development Program.
Abstract (Originalsprache)
Underwater image analysis is affected by light scattering, wavelength-dependent attenuation, low contrast, and suspended particles, which reduce the discriminative visual features. Current multi-scale Vision Transformers are not well-suited to these degradations because they cannot effectively fuse features across scales to achieve accurate classification. Although Vision Transformers (ViTs) can model long-range interactions, single-scale patch tokenization remains suboptimal for underwater images, where both fine-grained textures and global structures are important. This study proposes a Multi-Scale Vision Transformer (MS-ViT) with Cross-Scale Biased Attention Fusion (CSBAF) for underwater image classification. Before transformer encoding, the CSBAF introduces a learnable source–target scale-pair bias and an input-dependent scale-reliability gate. This differs from standard multi-scale fusion and cross-attention methods, which mainly concatenate features or exchange information between scale branches. The proposed design enables the model to emphasize reliable scales while suppressing degraded-scale responses. A hybrid dataset containing 14,000 images from the Roboflow Aquarium and RUIE datasets across five classes was used for evaluation. MS-ViT with CSBAF achieved 88.9% accuracy and an 88.8% F1-score, outperforming the CNN baseline by 7.6% and state-of-the-art transformer models, including UWFormer, DP-ViT, and CvT, by 2.3–4.2%. Ablation studies showed a 1.7% accuracy improvement over simple multi-scale concatenation, whereas cross-dataset testing achieved 84.4% accuracy, indicating reasonable cross-dataset robustness. These results demonstrate that explicit scale–aware fusion can improve transformer-based underwater visual understanding.
Förderung
- provincial Natural Science Foundation of Shanxi
- National Nature Science Foundation of China
- Shanxi Key Research and Development Program
- China Postdoctoral Science Foundation
Zitieren
APA
Faiz, A., Li, K., & Chen, P. (2026). Underwater Image Feature Extraction and Classification Using a Multi-Scale Vision Transformer with Cross-Scale Biased Attention Fusion. Applied Sciences, 16(11), 5358. https://doi.org/10.3390/app16115358
Vancouver
Faiz A, Li K, Chen P. Underwater Image Feature Extraction and Classification Using a Multi-Scale Vision Transformer with Cross-Scale Biased Attention Fusion. Applied Sciences. 2026;16(11):5358. doi:10.3390/app16115358
BibTeX
@article{Faiz2026,
author = {Faiz, Abdullah and Li, Kun and Chen, Ping},
title = {Underwater Image Feature Extraction and Classification Using a Multi-Scale Vision Transformer with Cross-Scale Biased Attention Fusion},
journal = {Applied Sciences},
year = {2026},
volume = {16},
number = {11},
pages = {5358},
doi = {10.3390/app16115358}
}
EUVIMED