Scribe identification for Tang Dynasty Changsha Kiln poetic ceramics via dual-path multi-scale global attention model
Source abstract
Abstract
Abstract Tang Dynasty Changsha Kiln ceramics are iconic cultural heritage featuring unique calligraphy. Current research focuses on textual analysis, neglecting scribe identification due to the absence of public datasets and limitations of expert-dependent methods. We address this gap by constructing the first comprehensive dataset of 1865 character images from 135 artifacts. To enable automated scribe attribution, we develop a dual-path convolutional neural network integrating a multi-scale global attention (MSGA) module, enhancing feature perception via multi-scale fusion and attention. Experiments show MSGA achieves 97.85% precision, significantly outperforming non-local attention baselines (93.75%). Applying our model to museum collections revealed two ceramics in separate institutions that originate from the same scribe—a finding undetected conventionally. This work establishes a quantitative framework for calligraphic style discrimination and ancient handwriting attribution, transforming historical calligraphy study through deep learning methods.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Image Processing and 3D Reconstruction · Aesthetic Perception and Analysis · Handwritten Text Recognition Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (37)
Show source-provided references
1. Kun, Y. Y. Mass-produced decorations: a study on the stylized manifestations in painted porcelain of Changsha Kiln. Des. Res. 14, 164–168 (2024).
2. Jian, M. L. et al. Revisiting the techniques of the colored porcelain of Changsha Kiln in the Tang and Five dynasties. J. Natl. Mus. China 11, 82–93 (2021).
3. Bin, B. C. A study on the artistic features of the shapes of Changsha Kiln Ceramics in the Tang Dynasty. Masterpieces Nat. 6, 32–38 (2021).
4. Qing, L. H. Research on colored drawing and poetry decoration of Changsha Kiln in Tang Dynasty. MSc thesis, Hunan University of Technology (2021).
5. Yu, X. C. The collection, collation and study on the characters of Changsha Kiln Porcelain in Tang and Five Dynasties. PhD thesis, Hunan Normal University (2021).
6. Baosheng, W. A. N. G. & Suzuki, N. Changsha kiln folk ceramics as a medium to express ordinary people’s thoughts. J. Jpn. Soc. Sci. Des. 60, 2_1-2_10 (2013).
7. Si, L. Z. & Xi, J. W. A Study on the folk characteristics of calligraphy in Changsha kiln in Tang dynasty. Ceramics 01, 158–159 (2021).
8. Fang, X. H. Tang Changsha Kiln underglaze poetry and prose decoration. Ceramics 03, 46–48 (2024).
9. Li, L. S. et al. Multi-measure similarity method for interpreting Bronze Inscriptions based on Hu Moment and TF-KSURF. Laser Optoelectron. Prog. 58, 123–131 (2021).
10. Ferdous Wahid, M., Shahriar, F. & Sobuj, S. I. A classical approach to handcrafted feature extraction techniques for bangla handwritten digit recognition. In 2021 International Conference on Electronics, Communications and Information Technology. 1–4 (IEEE, 2021).10.1109/ICECIT54077.2021.9641406 ↗
11. Tobing, T. et al. Character-based writer verification of ancient hebrew square-script manuscripts: on edge-direction feature. Soc. Imaging Sci. Technol. 20, 32 (2023).
12. Javidi, M. & Jampour, M. A deep learning framework for text-independent writer identification. Eng. Appl. Artif. Intell. 95, 103912 (2020).10.1016/j.engappai.2020.103912 ↗
13. Popović, M., Dhali, M. A. & Schomaker, L. Artificial intelligence based writer identification generates new evidence for the unknown scribes of the Dead Sea Scrolls exemplified by the Great Isaiah Scroll (1QIsaa). PloS One 16, e0249769 (2021).10.1371/journal.pone.0249769 ↗
14. Chattopadhyay, S. et al. Surds: self-supervised attention-guided reconstruction and dual triplet loss for writer independent offline signature verification. In Proceedings of the 26th International Conference on Pattern Recognition 1600–1606 (IEEE, 2022).10.1109/ICPR56361.2022.9956442 ↗
15. E. I. Bahi, H. Handwriting recognition in historical manuscripts using a deep learning approach. In International Conference on Advanced Intelligent Systems for Sustainable Development 528–535 (Springer, 2022).10.1007/978-3-031-26384-2_45 ↗
16. Sun, Y. et al. Offline Mongolian handwriting identification based on convolutional neural network. Electronics 13, 111 (2023).10.3390/electronics13010111 ↗
17. Tong, Y. L. et al. Small sample bronze inscription classification algorithm based on morphological features. Laser Optoelectron. Prog. 60, 175–184 (2023).
18. Islam, M. A. & Iacob, I. E. Manuscripts character recognition using machine learning and deeplearning. Modelling 04, 168–188 (2023).10.3390/modelling4020010 ↗
19. Peer, M. & Sablatnig, R. Feature mixing for writer retrieval and identification on papyri fragments. In Proceedings of the 7th International Workshop on Historical Document Imaging and Processing 31–36 (2023).10.1145/3604951.3605515 ↗
20. Moudgil, A. et al. Deep learning for ancient scripts recognition: a CapsNet-LSTM based approach. Alex. Eng. J. 103, 169–179 (2024).10.1016/j.aej.2024.06.007 ↗
21. Nabi, S. T., Kumar, M. & Singh, P. DeepNet-WI: a deep-net model for offline Urdu writer identification. Evol. Syst. 15, 759–769 (2024).10.1007/s12530-023-09504-1 ↗
22. Jing, Y. F. Research on Tang Dynasty Changsha Kiln Porcelain. PhD thesis, University of Chinese Academy of Social Sciences (2021).
23. Editorial Committee of Changsha Kiln. Changsha Kiln: Collection of Works (Hunan Fine Arts Publishing House, 2004).
24. Shi, R. Z. Tang Dynasty Style and Brilliant Colors: Deluxe Volume of Changsha Kiln (Hunan Fine Arts Publishing House, 2008).
25. Jian, M. L. Huxiang Ceramics Vol. 2: Changsha Kiln (Hunan Fine Arts Publishing House, 2009).
26. Shi, R. Z. & Yue, W. Collection of Changsha Kiln Works (Hubei Fine Arts Publishing House, 2004).
27. Ruan, J. et al. EGE-UNet: an efficient group enhanced UNet for skin lesion segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention 481–490 (Springer, 2023).10.1007/978-3-031-43901-8_46 ↗
28. Xu, Y. & Luo, R. Flattening surface based on using contour estimating subdivision surface. Preprint at https://arXiv.org/abs/2212.13489 (2022).
29. Zheng, P. et al. A free-form surface flattening algorithm that minimizes geometric deformation energy. IET Image Process. 16, 2544–2556 (2022).10.1049/ipr2.12508 ↗
30. Wang, X. et al. Non-local neural networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition 7794–7803 (IEEE, 2018).10.1109/CVPR.2018.00813 ↗
31. Zhao, H. & Li, H. Handwriting identification and verification using artificial intelligence-assisted textural features. Sci. Rep. 13, 21739 (2023).10.1038/s41598-023-48789-9 ↗
32. Luo, C., Jin, L. & Sun, Z. Moran: a multi-object rectified attention network for scene text recognition. Pattern Recognit. 90, 109–118 (2019).10.1016/j.patcog.2019.01.020 ↗
33. Upadhyay, R. R. & Singh, K. K. Handwritten signature verification system using hybrid transfer learning approach. Evol. Syst. 15, 2313–2322 (2024).10.1007/s12530-024-09617-1 ↗
34. Simonyan, K. & Zisserman, A. Very deep convolutional networks for large-scale image recognition. Preprint at https://arXiv.org/abs/1409.1556 (2014).
35. Wang, A. et al. RepViT: revisiting mobile CNN from ViT perspective. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 15909–15920 (IEEE, 2018).10.1109/CVPR52733.2024.01506 ↗
36. He, K. et al. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition 770–778 (IEEE, 2016).10.1109/CVPR.2016.90 ↗
37. Srivastava, A., Chanda, S. & Pal, U. Exploiting multi-scale fusion, spatial attention and patch interaction techniques for text-independent writer identification. In Asian Conference on Pattern Recognition 203–217 (Springer, 2021).10.1007/978-3-031-02444-3_15 ↗