[1]LUO Yanlong,BI Xiaojun,WU Licheng,et al.Dongba pictographs recognition based on improved residual learning[J].CAAI Transactions on Intelligent Systems,2022,17(1):79-87.[doi:10.11992/tis.202112009]
Copy

Dongba pictographs recognition based on improved residual learning

CAAI Transactions on Intelligent Systems[ISSN 1673-4785/CN 23-1538/TP] Volume: 17 Number of periods: 2022 1 Page number: 79-87 Column: 学术论文—智能系统 Public date: 2022-01-05
Title:
Dongba pictographs recognition based on improved residual learning
Author(s):
LUO Yanlong1;  BI Xiaojun2;  WU Licheng2;  LI Xiali2
1. College of Information and Communication Engineering, Harbin Engineering University, Harbin 150001, China;
2. School of Information Engineering, Minzu University of China, Beijing 100081, China
Keywords:
deep learning; Dongba pictographs; image recognition; build dataset; ResNet model; residual shortcut connection; improved down-sampling; recognition accuracy
CLC:
TP18
DOI:
10.11992/tis.202112009
Abstract:
Dongba pictographs recognition based on deep learning model has better recognition effect than that of traditional algorithms. However, these methods have disadvantages such as small number of recognizable Dongba pictographs and low recognition accuracy. Aiming at these problems, in this study, we build a novel dataset of Dongba pictographs that contains 1387 Dongba pictographs and more than 220 thousand images. Therefore, the number of recognizable Dongba pictographs is greatly increased and the Dongba pictographs recognition accuracy is improved. Since Dongba pictographs are characterized by high similarity and random writing, ResNet is adopted as an improved network structure. Moreover, we design a residual shortcut connection and the number of convolutional layers and introduce the max-pooling into the ResNet to improve down-sampling. The experimental results demonstrate that the improved ResNet model can recognize more Dongba characters, and has achieved the highest recognition accuracy 98.65% in our dataset.
References:
[1] GUO Hai, ZHAO Jingying, LI Xiaoniu. Preprocessing method for NaXi pictographs character recognition using wavelet transform[J]. International journal of digital content technology and its applications, 2010, 4(3): 117–131.
[2] GUO Hai, ZHAO Jingying. Research on feature extraction for character recognition of NaXi pictograph[J]. Journal of computers, 2011, 6(5): 947–954.
[3] DA Mingjun, ZHAO Jingying, SUO Guojie, et al. Online handwritten Naxi pictograph digits recognition system using coarse grid[C]//Computer science for environmental engineering and EcoInformatics. Berlin, German: Springer, 2011: 390?396.
[4] HSU C W, LIN C J. A comparison of methods for multiclass support vector machines[J]. IEEE transactions on neural networks, 2002, 13(2): 415–425.
[5] BREIMAN L. Random forests[J]. Machine learning, 2001, 45(1): 5–32.
[6] 徐小力, 蒋章雷, 吴国新, 等. 基于拓扑特征和投影法的东巴象形文识别方法研究[J]. 电子测量与仪器学报, 2017, 31(1): 150–154
XU Xiaoli, JIANG Zhanglei, WU Guoxin, et al. Identification method of Dongba pictograph based on topological characteristic and projection method[J]. Journal of electronic measurement and instrumentation, 2017, 31(1): 150–154
[7] 杨玉婷, 康厚良. 东巴象形文字特征曲线提取算法研究[J]. 图学学报, 2019, 40(3): 591–599
YANG Yuting, KANG Houliang. Research on the extracting algorithm of dongba hieroglyphic feature curves[J]. Journal of graphics, 2019, 40(3): 591–599
[8] 张泽晖. 基于卷积神经网络的东巴文字分类与识别[D]. 昆明: 云南大学, 2019.
ZHANG Zehui. Classification and recognition of dongba characters based on convolutional neural network[D]. Kunming: Yunnan University, 2019.
[9] WU Guoxin, LIU Xiuli, JIANG Zhanglei, et al. Dongba classical ancient books image classification method based on ReN-Softplus convolution residual neural network[C]//2019 14th IEEE International Conference on Electronic Measurement & Instruments. New York, USA: IEEE, 2019: 398?404.
[10] 谢裕睿, 董建娥. 基于ResNet网络的东巴象形文字识别研究[J]. 计算机时代, 2021(1): 6–10
XIE Yurui, DONG Jian’e. Research on Dongba hieroglyph recognition using ResNet network[J]. Computer era, 2021(1): 6–10
[11] 李霖灿. 纳西族象形标音文字字典[M]. 昆明: 云南民族出版社, 2001.
LI Lincan. Naxi Pictographs and Transcription Characters Dictionary[M]. Kunming: Yunnan Nationalities Publishing House, 2001.
[12] 方国瑜, 和志武. 纳西象形文字谱[M]. 昆明: 云南人民出版社, 1981.
FANG Guoyu, HE Zhiwu. Naxi pictograph character chart[M]. Kunming: Yunnan People’s Publishing House, 1981.
[13] BUSLAEV A, PARINOV A, KHVEDCHENYA E, et al. Albumentations: fast and flexible image augmentations[EB/OL]. (2018-09-18) [2021-11-30].https://arxiv.org/abs/1809.06839.
[14] ZHANG Hongyi, MOUSTAPHA Cisseet, YANN N Dauphin, et al. Mixup: beyond empirical risk minimization[EB/OL]. (2017-10-25) [2021-11-30].https://arxiv.org/abs/1710.09412.
[15] DAN Hendrycks, NORMAN Mu, EKIN D Cubuk, et al. AugMix: A simple data processing method to improve robustness and uncertainty[EB/OL]. (2019-12-05) [2021-11-30].https://arxiv.org/abs/1912.02781.
[16] HE Tong, ZHANG Zhi, ZHANG Hang, et al. Bag of tricks for image classification with convolutional neural networks[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York, USA: IEEE, 2019: 558?567.
[17] KHAN A, SOHAIL A, ZAHOORA U, et al. A survey of the recent architectures of deep convolutional neural networks[J]. Artificial intelligence review, 2020, 53(8): 5455–5516.
[18] XIE Saining, GIRSHICK R, DOLLáR P, et al. Aggregated residual transformations for deep neural networks[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition. New York, USA: IEEE, 2017: 5987?5995.
[19] TAN Mingxing, LE Quoc V. EfficientNet: rethinking model scaling for convolutional neural networks[EB/OL]. (2019-05-28) [2021-11-30].https://arxiv.org/abs/1905.11946.
[20] HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition. New York, USA: IEEE, 2016: 770?778.
[21] LIN Min, CHEN Qiang, YAN Shuicheng. Network in network[EB/OL]. (2013-12-16) [2021-11-30].https://arxiv.org/abs/1312.4400.
[22] HUANG Gao, LIU Shichen, MAATEN L V D, et al. CondenseNet: an efficient DenseNet using learned group convolutions[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York, USA: IEEE, 2018: 2752?2761.
[23] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[EB/OL]. (2014-09-04) [2021-11-30].https://arxiv.org/abs/1409.1556v3.
[24] IOFFE Sergey, SZEGEDY Christian. Batch normalization: accelerating deep network training by reducing internal covariate shift [C]//2015 International Conference on Machine Learning. New York, USA: ACM, 2015: 448?456.
[25] VINOD Nair, GEOFFREY E Hinton. Rectified Linear Units Improve Restricted Boltzmann Machines[C]//2010 International Conference on Machine Learning. New York, USA: ACM, 2010: 807?814.
Similar References:

Memo

-

Last Update: 1900-01-01

Copyright © CAAI Transactions on Intelligent Systems