Enhanced CNN-based Feature Extraction and Classification for Chinese Artwork Styles

Article Sidebar

PDF
Published: 2026-01-18
Keywords:
convolutional neural networks, artistic feature extraction, transfer learning, Chinese painting classification

Main Article Content

Jiaying Li
Integrated Marketing Communications, Northwestern University, Chicago, IL, USA

Abstract

This research focuses on optimizing convolutional neural network (CNN) architectures for extracting and classifying visual features in traditional Chinese paintings, which represent a distinctive artistic tradition characterized by brushstroke techniques, ink variations, and compositional nuances. The proposed hierarchical feature extraction framework integrates multi-scale fusion strategies with specialized modules for brushstroke, color, and compositional analysis. By systematically comparing ResNet, VGG, and EfficientNet backbones and combining them with layer-wise fine-tuning, the methodology achieves superior performance with limited training samples. Experimental validation on collections of traditional Chinese paintings demonstrates significant improvements in accuracy over strong CNN baselines, with the best configuration increasing overall accuracy from 82.1% to 93.2%. The framework provides practical solutions for museum digitization and auction-house cataloging.

Article Details

Issue

Vol. 1 No. 2 (2025): 2025 International Conference on Computer Science, Artificial Intelligence and Communication Systems (CSACS 2025)

Section

Articles

How to Cite

Enhanced CNN-based Feature Extraction and Classification for Chinese Artwork Styles. (2026). Journal of Science, Innovation & Social Impact, 1(2), 135-148. https://sagespress.com/index.php/JSISI/article/view/61

References

1. G. Castellano, and G. Vessio, "Deep learning approaches to pattern extraction and recognition in paintings and drawings: An overview," Neural Computing and Applications, vol. 33, no. 19, pp. 12263-12282, 2021. doi: 10.1007/s00521-021-05893-z

2. W. Li, "Enhanced automated art curation using supervised modified CNN for art style classification," Scientific Reports, vol. 15, no. 1, p. 7319, 2025. doi: 10.1038/s41598-025-91671-z

3. M. V. Conde, and K. Turgutlu, "CLIP-Art: Contrastive pre-training for fine-grained art classification," In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 3956-3960. doi: 10.1109/cvprw53098.2021.00444

4. W. Zhao, D. Zhou, X. Qiu, and W. Jiang, "Compare the performance of the models in art classification," PLoS ONE, vol. 16, no. 3, p. e0248414, 2021. doi: 10.1371/journal.pone.0248414

5. C. Sandoval, E. Pirogova, and M. Lech, "Two-stage deep learning approach to the classification of fine-art paintings," IEEE Access, vol. 7, pp. 41770-41781, 2019. doi: 10.1109/access.2019.2907986

6. E. Cetinic, T. Lipic, and S. Grgic, "Fine-tuning convolutional neural networks for fine art classification," Expert Systems with Applications, vol. 114, pp. 107-118, 2018. doi: 10.1016/j.eswa.2018.07.026

7. W. Zhao, W. Jiang, and X. Qiu, "Big transfer learning for fine art classification," Computational Intelligence and Neuroscience, vol. 2022, no. 1, p. 1764606, 2022.

8. H. Ugail, D. G. Stork, H. Edwards, S. C. Seward, and C. Brooke, "Deep transfer learning for visual analysis and attribution of paintings by Raphael," npj Heritage Science, vol. 11, no. 1, p. 268, 2023. doi: 10.1186/s40494-023-01094-0

9. Q. Zhao, and R. Zhang, "Classification of painting styles based on the difference component," Expert Systems with Applications, vol. 259, p. 125287, 2025. doi: 10.1016/j.eswa.2024.125287

10. L. Qiao, X. Guo, and W. Li, "Classification of Chinese and Western painting images based on brushstrokes feature," In International Conference on Entertainment Computing, November, 2020, pp. 325-337. doi: 10.1007/978-3-030-65736-9_30

11. Y. Fu, H. Yu, C. K. Yeh, T. Y. Lee, and J. J. Zhang, "Fast accurate and automatic brushstroke extraction," ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM), vol. 17, no. 2, pp. 1-24, 2021.

12. X. Du, and Y. Cai, "Design of Chinese painting style classification model based on multi-layer aggregation CNN," PeerJ Computer Science, vol. 10, p. e2303, 2024. doi: 10.7717/peerj-cs.2303

13. Y. Y. Huang, S. S. Yu, J. J. Chu, H. H. Fan, and B. B. Du, "Using knowledge graphs and deep learning algorithms to enhance digital cultural heritage management," npj Heritage Science, vol. 11, no. 1, p. 204, 2023. doi: 10.1186/s40494-023-01042-y

14. T. Karras, M. Aittala, J. Hellsten, S. Laine, J. Lehtinen, and T. Aila, "Training generative adversarial networks with limited data," Advances in Neural Information Processing Systems, vol. 33, pp. 12104-12114, 2020.

15. Z. Su, L. Zhang, J. Liu, and S. Wang, "Material classification method of traditional Chinese painting image based on prototypical network," npj Heritage Science, vol. 13, no. 1, p. 377, 2025. doi: 10.1038/s40494-025-01949-8