Abstract
Recent researches introduced fast, compact and efficient convolutional neural networks (CNNs) for offline handwritten Chinese character recognition (HCCR). However, many of them did not address the problem of network interpretability. We propose a new architecture of a deep CNN with high recognition performance which is capable of learning deep features for visualization. A special characteristic of our model is the bottleneck layers which enable us to retain its expressiveness while reducing the number of multiply-accumulate operations and the required storage. We introduce a modification of global weighted average pooling (GWAP)—global weighted output average pooling (GWOAP). This paper demonstrates how they allow us to calculate class activation maps (CAMs) in order to indicate the most relevant input character image regions used by our CNN to identify a certain class. Evaluating on the ICDAR-2013 offline HCCR competition dataset, we show that our model enables a relative 0.83% error reduction while having 49% fewer parameters and the same computational cost compared to the current state-of-the-art single-network method trained only on handwritten data. Our solution outperforms even recent residual learning approaches.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Abadi M, Barham P, Chen J, Chen Z, Davis A, Dean J, Devin M, Ghemawat S, Irving G, Isard M et al (2016) Tensorflow: a system for large-scale machine learning. OSDI 16:265–283
Al-Janabi S (2018) Smart system to create an optimal higher education environment using ida and iots. Int J Comput Appl. https://doi.org/10.1080/1206212x.2018.1512460
Al-Janabi S, Abaid Mahdi M (2019) Evaluation prediction techniques to achievement an optimal biomedical analysis. Int J Grid Utility Comput. https://doi.org/10.1504/ijguc.2019.10020511
Al-Janabi S, Alkaim AF (2019) A nifty collaborative analysis to predicting a novel tool (DRFLLS) for missing values estimation. Soft Comput 1–15
Al-Janabi S, Salman M, Fanfakh A (2018a) Recommendation system to improve time management for people in education environments. J Eng Appl Sci 13:10182–10193
Al-Janabi S, Salman MA, Mohammad M (2018b) Multi-level network construction based on intelligent big data analysis. In: International conference on bigdata and smart digital environment. Springer, pp 102–118
Ali SH (2012) A novel tool (FP-KC) for handle the three main dimensions reduction and association rule mining. In: 2012 6th International conference on sciences of electronics. Technologies of information and telecommunications (SETIT). IEEE, pp 951–961
Arqub OA, Mohammed AS, Momani S, Hayat T (2016) Numerical solutions of fuzzy differential equations using reproducing kernel Hilbert space method. Soft Comput 20(8):3283–3302
Arqub OA, Al-Smadi M, Momani S, Hayat T (2017) Application of reproducing kernel algorithm for solving second-order, two-point fuzzy boundary value problems. Soft Comput 21(23):7191–7206
Chen TQ, Rubanova Y, Bettencourt J, Duvenaud DK (2018) Neural ordinary differential equations. In: Advances in neural information processing systems, pp 6572–6583
Cheng C, Zhang XY, Shao XH, Zhou XD (2016) Handwritten Chinese character recognition by joint classification and similarity ranking. In: 2016 15th International conference on frontiers in handwriting recognition (ICFHR). IEEE, pp 507–511
Chollet F et al (2015) Keras. https://keras.io
Cireşan D, Meier U, Schmidhuber J (2012) Multi-column deep neural networks for image classification. arXiv preprint arXiv:1202.2745
He K, Zhang X, Ren S, Sun J (2015) Delving deep into rectifiers: surpassing human-level performance on imagenet classification. In: Proceedings of the IEEE international conference on computer vision, pp 1026–1034
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 770–778
Ioffe S, Szegedy C (2015) Batch normalization: accelerating deep network training by reducing internal covariate shift. arXiv preprint arXiv:1502.03167
Kenan K, Ali SH, Patel A (2015) Rapid lossless compression of short text messages. Comput Stand Interfaces 37:53–59. https://doi.org/10.1016/j.csi.2014.05.005. http://www.sciencedirect.com/science/article/pii/S0920548914000737
Kimura F, Takashina K, Tsuruoka S, Miyake Y (1987) Modified quadratic discriminant functions and the application to Chinese character recognition. IEEE Trans Pattern Anal Mach Intell 1:149–153
Li F, Shen Q, Li Y, Mac Parthaláin N (2016) Handwritten Chinese character recognition using fuzzy image alignment. Soft Comput 20(8):2939–2949
Li Z, Teng N, Jin M, Lu H (2018) Building efficient CNN architecture for offline handwritten Chinese character recognition. Int J Doc Anal Recognit (IJDAR) 21(4):233–240
Lin M, Chen Q, Yan S (2013) Network in network. arXiv preprint arXiv:1312.4400
Liu CL, Yin F, Wang DH, Wang QF (2011) CASIA online and offline Chinese handwriting databases. In: 2011 International conference on document analysis and recognition (ICDAR). IEEE, pp 37–41
Liu CL, Yin F, Wang DH, Wang QF (2013) Online and offline handwritten Chinese character recognition: benchmarking on new databases. Pattern Recognit 46(1):155–162
Lu S, Wei X, Lu Y (2015) Cost-sensitive MQDF classifier for handwritten Chinese address recognition. In: 2015 13th International conference on document analysis and recognition (ICDAR). IEEE, pp 76–80
Qin Z, Yu F, Liu C, Chen X (2018) How convolutional neural networks see the world—a survey of convolutional neural network visualization methods. Math Found Comput 1(2):149–180
Saravanan B, Mohanraj V, Senthilkumar J (2019) A fuzzy entropy technique for dimensionality reduction in recommender systems using deep learning. Soft Comput 23(8):2575–2583
Silver D, Huang A, Maddison CJ, Guez A, Sifre L, van den Driessche G, Schrittwieser J, Antonoglou I, Panneershelvam V, Lanctot M, Dieleman S, Grewe D, Nham J, Kalchbrenner N, Sutskever I, Lillicrap T, Leach M, Kavukcuoglu K, Graepel T, Hassabis D (2016) Mastering the game of Go with deep neural networks and tree search. Nature 529:484–503
Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 15(1):1929–1958
Xiao X, Jin L, Yang Y, Yang W, Sun J, Chang T (2017) Building fast and compact convolutional neural networks for offline handwritten Chinese character recognition. Pattern Recognit 72:72–81
Yang X, He D, Zhou Z, Kifer D, Giles CL (2017) Improving offline handwritten Chinese character recognition by iterative refinement. In: 2017 14th IAPR international conference on document analysis and recognition (ICDAR). IEEE, pp 5–10
Yin F, Wang QF, Zhang XY, Liu CL (2013) ICDAR 2013 Chinese handwriting recognition competition. In: 2013 12th International conference on document analysis and recognition (ICDAR). IEEE, pp 1464–1470
Zhang Y (2015) Deep convolutional network for handwritten Chinese character recognition. Computer Science Department, Stanford University
Zhang XY, Bengio Y, Liu CL (2017) Online and offline handwritten Chinese character recognition: a comprehensive study and new benchmark. Pattern Recognit 61:348–360
Zhang Y, Liang S, Nie S, Liu W, Peng S (2018) Robust offline handwritten character recognition through exploring writer-independent features under the guidance of printed data. Pattern Recognit Lett 106:20–26
Zhong Z, Jin L, Xie Z (2015) High performance offline handwritten Chinese character recognition using googlenet and directional feature maps. In: 2015 13th International conference on document analysis and recognition (ICDAR). IEEE, pp 846–850
Zhong Z, Zhang XY, Yin F, Liu CL (2016) Handwritten Chinese character recognition with spatial transformer and deep residual networks. In: 2016 23rd International conference on pattern recognition (ICPR). IEEE, pp 3440–3445
Zhou B, Khosla A, Lapedriza A, Oliva A, Torralba A (2016) Learning deep features for discriminative localization. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 2921–2929
Acknowledgements
This work is supported by National Natural Science Foundation of China under Grant No. 61472123 and Hunan Provincial Natural Science Foundation under Grant No. 2018JJ2064. We would like to express our gratitude to the China Scholarship Council for giving the first author an opportunity to obtain master’s degree at Hunan University under Chinese Government Scholarship.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
All authors declare that they have no conflict of interest regarding the publication of this paper.
Additional information
Communicated by Mu-Yen Chen.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Melnyk, P., You, Z. & Li, K. A high-performance CNN method for offline handwritten Chinese character recognition and visualization. Soft Comput 24, 7977–7987 (2020). https://doi.org/10.1007/s00500-019-04083-3
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-019-04083-3