EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Noise-Robust Deep Learning Framework with Hierarchical Attention for Rock Image Recognition Under Inaccurate Supervision

Jiangbing Sun, Xinyi Zhu, Yan Zhang, Wei Qian, Hongbing Zhang, Yihang Ge, Zhenyi Song

Applied Sciences · 10.08.2026 · 16 (16) · 7968

DOI: 10.3390/app16167968 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 10.08.2026. Verfasst von Jiangbing Sun und 6 weiteren Autor:innen. Die Arbeit nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Rock image recognition based on deep learning serves as a critical task in intelligent geological analysis and core documentation. Although recent advances in deep learning have established the technical foundation for rock image identification, its practical deployment remains constrained by two main bottlenecks: intricate feature representations and ubiquitous label noise. Complexity of rock image features limits the effectiveness of intelligent model applications. Additionally, manual mislabeling and coarse marking issues during lithology annotation are often overlooked, leading to label noise in training datasets. To resolve these challenges, this study proposes a robust learning framework that integrates a novelty-driven feature extraction module with noise-resistant loss functions. Specifically, a Rock Hierarchical Heterogeneous Attention (RHHA) module is designed for enhancing rock feature extraction capabilities by applying distinct spatial and channel attention biases at shallow and deep layers of the network. Furthermore, Symmetric Cross Entropy (SCE) loss is used for loss calculation to enhance the robustness of models under noisy conditions. Comparative experiments were carried out on a real-world dataset of metamorphic rock images from northern Jiangsu Province, China. The results show that the proposed training framework achieves the best performance in rock image identification tasks, outperforming baseline and other comparative methods. The RHHA module outperforms other existing attention modules in capturing the texture and semantic features of rocks. In particular, the SCE loss effectively mitigates overfitting and maintains excellent generalization capability in the presence of noisy labels. The proposed framework holds promise for providing new insights into rock image recognition tasks.

Zitieren

APA

Sun, J., Zhu, X., Zhang, Y., Qian, W., Zhang, H., Ge, Y., & Song, Z. (2026). A Noise-Robust Deep Learning Framework with Hierarchical Attention for Rock Image Recognition Under Inaccurate Supervision. Applied Sciences, 16(16), 7968. https://doi.org/10.3390/app16167968

Vancouver

Sun J, Zhu X, Zhang Y, Qian W, Zhang H, Ge Y, et al. A Noise-Robust Deep Learning Framework with Hierarchical Attention for Rock Image Recognition Under Inaccurate Supervision. Applied Sciences. 2026;16(16):7968. doi:10.3390/app16167968

BibTeX
@article{Sun2026,
  author = {Sun, Jiangbing and Zhu, Xinyi and Zhang, Yan and Qian, Wei and Zhang, Hongbing and Ge, Yihang and Song, Zhenyi},
  title = {A Noise-Robust Deep Learning Framework with Hierarchical Attention for Rock Image Recognition Under Inaccurate Supervision},
  journal = {Applied Sciences},
  year = {2026},
  volume = {16},
  number = {16},
  pages = {7968},
  doi = {10.3390/app16167968}
}
Kostenlos starten →