[1]马龙龙,刘成林.基于统计部首模型的联机手写汉字识别方法[J].智能系统学报,2010,5(5):385-391.[doi:10.3969/j.issn.1673-4785.2010.05.002]
MA Long-long,LIU Cheng-lin.On-line handwritten Chinese character recognition using statistical radical models[J].CAAI Transactions on Intelligent Systems,2010,5(5):385-391.[doi:10.3969/j.issn.1673-4785.2010.05.002]
点击复制
MA Long-long,LIU Cheng-lin.On-line handwritten Chinese character recognition using statistical radical models[J].CAAI Transactions on Intelligent Systems,2010,5(5):385-391.[doi:10.3969/j.issn.1673-4785.2010.05.002]
基于统计部首模型的联机手写汉字识别方法
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
5
期数:
2010年第5期
页码:
385-391
栏目:
学术论文—机器感知与模式识别
出版日期:
2010-10-25
- Title:
- On-line handwritten Chinese character recognition using statistical radical models
- 文章编号:
- 1673-4785(2010)05-0385-07
- Keywords:
- on-line handwritten Chinese character recognition; statistical radical model; hierarchical structure; over segmentation; path search; radical recognition
- 分类号:
- TP391.4
- 文献标志码:
- A
- 摘要:
- 利用汉字的部首层次结构有助于减小字符识别器的存储空间和提高泛化性、适应性，但部首分割一直是一个难点.提出一种新的基于部首的联机手写汉字识别方法,该方法把部首形状信息和几何信息集成到识别框架中,在组合搜索过程中利用字符-部首的层次结构字典引导部首的分割与识别,从而提高部首分割的准确率.为克服部首间的连笔,引入角点检测提取子笔划.部首识别采用统计分类器，模型参数通过自学习得到.在字符识别中,采用了2种不同的字典表示以及相应的不同搜索算法.该方法已用于左右与上下结构的字符集，实验结果表明了该方法的有效性.
- Abstract:
- The hierarchical radical structure of Chinese characters can be explored to reduce the number of parameters in character recognition, as well as to improve the generalization ability and adaptability. However, the segmentation of radicals from characters has long been a difficult problem. A new radical-based approach for online handwritten Chinese character recognition was proposed. The approach integrated appearance-based radical recognition and geometric context into a principled framework using a hierarchical character-radical dictionary to guide radical segmentation and recognition during the path search process for the purpose of increasing the accuracy of radical segmentation. The parameters of statistical radical models were estimated in embedded learning. To overcome the connection of strokes between radicals, corner points were detected to extract sub-strokes. For character recognition, two dictionary representation schemes and accordingly different search algorithms were used. The effectiveness of the proposed approach has been demonstrated on Chinese characters of left-right and up-down structures.
- 参考文献/References:
-
［1］LIU Chenglin, JAEGER S, NAKAGAWA M. Online handwritten Chinese character recognition: the state of the art［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(2): 198-213.
［2］KWON J, SHIN B, KIM J H. Recognition of on-line cursive Korean characters combining statistical and structural methods［J］. Pattern Recognition, 1997, 30(8): 1255-1263.
［3］KANG K W, KIM J H. Utilization of hierarchical stochastic relationship modeling for Hangul character recognition［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(9): 1185-1196.
［4］OTA I, YAMAMOTO R, NISHIMOTO T, et al. Online handwritten kanji recognition based on inter-stroke grammar［C］//Proceedings of the 9th International Conference on Document Analysis and Recognition. Curitiba, Brazil, 2007: 1188-1192.
［5］KAWAMURA A, YURA K, HAYAMA T, et al. On-line recognition of freely handwritten Japanese characters using directional feature densities［C］//Proceedings of the 11th International Conference on Pattern Recognition. Hague, Netherlands, 1992, 2: 183-186.
［6］HAMANAKA M, YAMADA K, TSUKUMO J. On-line Japanese character recognition experiments by an off-line method based on normalized-cooperated feature extraction［C］//Proceedings of the 2nd International Conference on Document Analysis and Recognition. Tsukuba, Japan, 1993: 204-207.
［7］KIM H J, KIM K H, KIM S K, et al. On-line recognition of handwritten Chinese characters based on hidden Markov models［J］. Pattern Recognition, 1997, 30(9): 1489-1499.
［8］NAKAI M, AKIRA N, SHIMODAIRA H, et al. Substroke approach to HMM-based on-line kanji handwriting recognition［C］//Proceedings of the 6th International Conference on Document Analysis and Recognition. Seattle, USA, 2001: 491-495.
［9］ZHANG Yaodong, LIU Peng, SOONG F K. Minimum error discriminative training for radical-based online Chinese handwriting recognition［C］//Proceedings of the 9th International Conference on Document Analysis and Recognition. Curitiba, Brazil, 2007: 53-57.
［10］LIU Yingjian, ZHANG Liqin, DAI Juwei. A new approach to on-line handwriting Chinese character recognition［C］//Proceedings of the 2nd International Conference on Document Analysis and Recognition. Tsukuba, Japan, 1993: 192-195.
［11］CHELLAPILLA K, SIMARD P. A new radical based approach to offline handwritten East-Asian character recognition［C］//Proceedings of the 10th International Workshop on Frontiers in Handwriting Recognition. La Baule, France, 2006: 261-266.
［12］LIU Chenglin, KOGA M, FUJISAWA H. Lexicon-driven segmentation and recognition of handwritten character strings for Japanese address reading［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(11): 1425-1437.
［13］MA Longlong, LIU Chenglin. A new radical-based approach to online handwritten Chinese character recognition［C］//Proceedings of the 19th International Conference on Pattern Recognition. Tampa, USA, 2008: 1-4.
［14］MA Longlong, LIU Chenglin. On-line handwritten Chinese character recognition based on nested segmentation of radicals［C］//Proceedings of 2009 Chinese Conference on Pattern Recogniton & First CJK Joint Workshop on Pattern Recognition. Nanjing, China, 2009: 1-5.
［15］LIU Chenglin, ZHOU Xiangdong. Online Japanese character recognition using trajectory-based normalization and direction feature extraction［C］//Proceedings of the 10th International Workshop on Frontiers in Handwriting Recognition. La Baule, France, 2006: 217-222.
［16］KIMURA F, TAKASHINA K, TSURUOKA S, et al. Modified quadratic discriminant functions and the application to Chinese character recognition［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1987, 9(1): 149-153.
备注/Memo
收稿日期：2009-11-13.
基金项目：国家自然科学基金资助项目(60775004,60825301).
通信作者：马龙龙.E-mail: longma@nlpr.ia.ac.cn.
作者简介：马龙龙，男，1981年生，博士研究生，主要研究方向为联机手写汉字识别.刘成林，男，1967年生，研究员、博士生导师，主要研究方向为模式识别和文字识别.2005年获得IAPR/ICDAR青年科学家奖，发表学术论文90余篇.
更新日期/Last Update:
2010-11-24