No Detail Found!

Enhanced LeNet Model for Chinese Calligraphy Style Recognition

Author(s)
Luyao Sun Department of Computer Science, Faculty of Computer Science and Information Technology, Universiti Putra Malaysia (UPM), Serdang, Malaysia
Noridayu Manshor Department of Computer Science, Faculty of Computer Science and Information Technology, Universiti Putra Malaysia (UPM), Serdang, Malaysia
Fatimah Khalid Department of Multimedia, Faculty of Computer Science and Information Technology, Universiti Putra Malaysia (UPM), Serdang, Malaysia
Abstract
The development of Chinese characters over centuries has a lengthy history and is of great value for research. However, examining Chinese calligraphic characters can be challenging and requires a comprehensive understanding of Chinese history. This paper presents the CCSR (Chinese Calligraphy Style Recognition) model for recognizing calligraphic styles, which is based on the LeNet-5 model with a new Concat layer that includes weight mean and scaling operations. The proposed CCSR model effectively classifies calligraphic styles and outperforms traditional recognition methods in terms of accuracy and efficiency. Compared to conventional feature extraction methods, the CCSR model achieves a recognition accuracy of 98.35%, significantly higher than the best traditional methods. When compared with existing deep learning models, such as those incorporating Squeeze-and-Excitation (SE) and Convolutional Block Attention Module (CBAM) modules, the CCSR model not only achieves competitive or superior accuracy but also reduces training time by up to 28.5 minutes. Furthermore, the CCSR model demonstrates its robustness by successfully classifying calligraphic styles from a challenging dataset that contains low-resolution and poorly preserved images. The model maintains an accuracy of 90.79%, outperforming existing methods in difficult recognition conditions. These results highlight the CCSR model's superior performance in both accuracy and training efficiency, particularly in challenging scenarios involving low-quality data.
Publication Details
Page(s) 680-695
Published Journal: Pakistan Journal of Life and Social Sciences, Volume: 23, Issue: 2, Year: 2025

Request the full text from PASTIC

Keywords
deep learning, , LeNet-5 Model, Style Recognition, Chinese calligraphy Concat Layer
References
Munaye Admass W. S.,Bogale Y. Y.,G. A. Y. Y. .2024 .Convolutional neural networks and histogramoriented gradients: A hybrid approach for automatic mango disease detection and classification. International Journal of Information Technology, 16(2) : 817-829.
Alwagdani M. S. Jaha E. S. .2023 .Deep learning-based child handwritten Arabic character recognition and handwriting discrimination. Sensors, 23(15) : 6774.
Burge M. J. Burger W. .2022 .Scale-invariant feature transform (SIFT). Digital Image Processing: An Algorithmic Introduction, : 709-763.
Su Chen Y. S.,Li G.,H. G. .2016 .Machine learning for calligraphy styles recognition. CS229 Final Project, : .
Dai Fei W.,Li W.,Zou C.,Xiong J.,H. J. .2010 .Weight mean and last BN layer mimic the dynamic of batch normalization. , : .
Luen Gang Z. L.,Cheong L. C.,L. K. L. C. .2023 .The Aesthetic Structure of Cursive Script. International Journal of Academic Research in Business and Social Sciences, 13(7) : 1159-1169.
He K.,Sun J. .2015 .Convolutional neural networks at constrained time cost. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, : 5353-5360.
Li Hou Q.,Kang C.,Zhao M.,X. M. .2021 .Intelligent model for speech recognition based on SVM: A case study on English language. Journal of Intelligent & Fuzzy Systems, 40(2) : 2721-2731.
Liu Hu H.,Guan A.,Qian Q.,Li H.,Chen X.,Zhou S.,Q. S. .2022 .Adaptively customizing activation functions for various layers. IEEE Transactions on Neural Networks and Learning Systems, 34(9) : 6096-6107.
Huang Jin J.,C. Jin J. .2023 .China academic digital libraries over 20 years: From digital collaboration to digital knowledge service alliance. International Federation of Library Associations and Institutions (IFLA)., : .
Wijesiriwardana Karunarathne D.,Nishantha C.,Kumara G.,M. G. .2024 .Efficiency and accuracy in Sinhala handwritten character recognition: A Gabor-initialized CNN perspective. Sri Lanka Journal of Technology, 5(1) : 13-24.
Doshi Kurani A.,Vakharia P.,Shah A.,M. A. .2023 .A comprehensive comparative study of artificial neural network (ANN) and support vector machines (SVM) on stock forecasting. Annals of Data Science, 10(1) : 183-208.
Li B. .2016 .Convolution Neural Network for Traditional Chinese Calligraphy Recognition. CS231N Final Project. , : .
Fu Li G.,Gao L.,Fang C.,Zhao W.,Shi G.,Cui F.,Y. F. .2022 .Multi-class detection of kiwifruit flower and its distribution identification in orchard based on YOLOv5l and Euclidean distance. Computers and Electronics in Agriculture, 201 : 107342.
Luo X.,Meng Q.,Chen W.,Wang Y. .2021 .Towards effective batch normalization in the Path Space for ReLU networks. Uncertainty in Artificial Intelligence, : 834-843.
Darmawan Nugraha G. S.,Dwiyansaputra M. I.,R. M. I.,Resnet- M. I. .2023 .50 in Arabic handwriting pattern recognition. Kinetik: Game Technology, Information System, 8 : .
Raju G. .2008 .Wavelet transform and projection profiles in handwritten character recognition-A performance analysis. 2008 16th International Conference on Advanced Computing and Communications, : 309-314.
Schoemans M.,Aref W. G.,Zimányi E.,Sakr M. .2024 .Multi-Entry Generalized Search Trees for Indexing Trajectories. Proceedings of the 32nd ACM International Conference on Advances in Geographic Information Systems, : 421-431.
Segu M.,Tonioni A.,and Tombari F. A. .2023 .Batch normalization embeddings for deep domain generalization. Pattern Recognition, 135 : 109115.
Wang K.,Yi Y.,Peng Tang Z.,J. Tang Z. .2020 .Multi-scene ancient Chinese text recognition with deep coupled alignments. Applied Soft Computing, 108 : 107475.
Ramos Weigang W.,Brom R. M.,Li P. C.,D. L. P. C. .2025 .Threshold study for hanzi image recognition: Defining character and component limits in Chinese, Japanese, and Korean script processing. International Journal of Asian Language Processing, 35(1) : 2450011.
Ladgham Zaibi A.,Sakly A.,A. A. .2021 .A lightweight model for traffic sign classification based on enhanced LeNet‐5 network. Journal of Sensors, (1) : 8870529.
Yu Zhang J.,W. Zhang J.,Li Z.,J. Z. .2021 .Attention-enhanced CNN for Chinese calligraphy styles classification. 2021 IEEE 7th International Conference on Virtual Reality (ICVR), : 352-358.
Liu Zhang Y.,He Y.,J. Y. .2013 .Recognition of calligraphy style based on global feature descriptor. 2013 IEEE International Conference on Multimedia and Expo (ICME), : 1-6.
Zhu C.-H. ,Gan H.,Wang J.-P. H. .2014 .Research on classification and recognition of Chinese characters based on FSVM offline handwriting. Computer Engineering and Applications, 23 : .
0

Citations

0

Downloads

46

Views