[1]马少平,金奕江.基于多Agent系统的脱机手写体汉字识别[J].智能系统学报,2009,4(5):398-405.[doi:10.3969/j.issn.1673-4785.2009.05.003]
　MA Shao-ping .,JIN Yi-jiang ...Offline recognition of hand-written Chinese characters based on a multi-Agent system[J].CAAI Transactions on Intelligent Systems,2009,4(5):398-405.[doi:10.3969/j.issn.1673-4785.2009.05.003]
点击复制

基于多Agent系统的脱机手写体汉字识别

《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷: 4 期数: 2009年第5期 页码: 398-405 栏目: 学术论文—机器感知与模式识别 出版日期: 2009-10-25
Title:
Offline recognition of hand-written Chinese characters based on a multi-Agent system
文章编号:
1673-4785(2009)05-0398-08
作者:
马少平1,2,3，金奕江1,2,3
1.清华大学计算机科学与技术系，北京100084；2.清华大学智能技术与系统国家重点实验室，北京100084；3.清华大学清华信息科学与技术国家实验室（筹），北京100084
Author(s):
MA Shao-ping 1,2.3， JIN Yi-jiang 1.2.3
1. Department of Computer Science and Technology, Tsinghua University， Beijing 100084， China; 2.State Key Lab of Intelligent Technology and Systems, Tsinghua University， Beijing 100084, China； 3.Tsinghua National Laboratory of Information Science and Technology, Tsinghua University， Beijing 100084, China
关键词:
汉字识别; 多Agent系统; 类市场模型; 模糊综合; 辩论协商规则
Keywords:
hand-written chinese character;  multi-Agent system;  market-like model;  fuzzy synthetic;  debate-negotiation rules
分类号:
TP391.4
DOI:
10.3969/j.issn.1673-4785.2009.05.003
文献标志码:
A
摘要:
由于脱机手写体汉字的多样性和随意性，识别起来具有很大的难度，依靠单一的特征很难实现高准确率的识别.引入多Agent的概念，将多种知识统一于多Agent系统之中，给出了一个面向脱机手写体汉字识别的多Agent类市场模型，提出了一种模糊综合方法和辩论协商规则，实现了一个基于多Agent系统的脱机手写体汉字识别系统.初步测试结果显示出系统的有效性.
Abstract:
Due to the diversity and randomness of Chinese characters, it is difficult for offline handwritten Chinese character recognition to perform well when based solely on analysis of a single feature. In order to solve this problem, a multiAgent based recognition method was proposed. It merges a variety of knowledge into a marketlike model. A comprehensive approach using fuzzy rules to provide consultation and debate rules between Agents was also incorporated. With this proposed method, a multiAgent offline handwritten Chinese character recognition system was constructed. Preliminary experimental results showed the effectiveness of this system.
参考文献/References:
［1］陈?? 静，穆志纯，孙筱倩. 计算机模拟汉字字形认知过程的研究［J］. 智能系统学报， 2008, 3(3): 216-221.
CHEN Jing, MU Zhichun, SUN Xiaoqian. Computer simulation of the cognition of Chinese characters［J］. CAAI Transactions on Intelligent Systems, 2008, 3(3): 216-221.
［2］张忻中. 汉字识别技术［M］. 北京：清华大学出版社，1992: 31-41, 125-160.
［3］DU Qingdong, LIU Jie. A new neural fusion recognition method with multi-Agent［C］//Proceedings of the Third International Conference on International Information Hiding and Multimedia Signal Processing (IIH-MSP 2007). Washington DC: IEEE Computer Society， 2007: 127-130.
［4］ZHU Xiaoyan.Multiple neural networks model and its application in pattern recognition［C］//IEEE International Conference on Neural Information Processing.Beijing, China, 1995: 966-969.
［5］张永慧，刘昌平，罗??? 公，等.技术综合集成在模式识别中的应用［J］.计算机学报， 1995, 18(19): 678-685.
ZHANG Yonghui, LIU Changping, LUO Gong, et al. Integration comprehensive techniques in pattern recognition［J］. Chinese Journal of Computers, 1995, 18(19): 678-685.
［6］马少平，夏?? 莹，朱小燕. 基于模糊方向线素特征的手写体汉字识别［J］. 清华大学学报：自然科学版，1997, 37(3): 42-45.
MA Shaoping, XIA Ying, ZHU Xiaoyan. Handwritten Chinese characters recognizing based on fuzzy directional line element feature［J］. Journal of Tsinghua University: Sci & Tech, 1997, 37(3): 42-45.
［7］夏??? 莹，马少平，常新功，等. 基于统计的汉字识别文本的自动后处理方法［J］. 模式识别与人工智能, 1996, 9(2): 172-178.
XIA Ying, MA Shaoping, CHANG Xingong, et al. The method of automatic postprocessing based statistical probabilities for Chinese recognition text［J］. Pattern Recognition and Artificial Intelligence, 1996, 9(2): 172-178.
［8］RUSSELL S, NORVING P. 人工智能——一种现代方法［M］. 姜?? 哲，金奕江，张??? 敏，等，译. 2版. 北京：人民邮电出版社，2004: 26-42.
［9］LIU Jiming. 多智能体原理与技术［M］. 靳小龙, 张世武, LIU Jiming，译. 北京：清华大学出版社, 2003: 1-17, 43-64.
［10］HUYNH T D, JENNINGS N R, SHOADBOLT N R. An integrated trust and reputation model for open multi-Agent systems［J］. Autonomous Agents and Multi-Agent Systems, 2006, 13(2): 119-154.
［11］PANAIT L, LUKE S. Cooperative multi-Agent learning: the state of the art［J］. Autonomous Agents and MultiAgent Systems, 2005, 11(3): 387-434.
［12］王立春，陈世福. 多Agent多问题协商模型［J］. 软件学报， 2002, 13(8): 1637-1643.
?WANG Lichun, CHEN Shifu. A multiAgent multi-issue negotiation model［J］. Journal of Software, 2002, 13(8): 1637-1643.
［13］张德喜，马少平，朱绍文，等. 基于统计与神经元方法相结合的手写体相似字识别［J］. 中文信息学报， 1999, 13(3): 33-39.
ZHANG Dexi, MA Shaoping, ZHU Shaowen, et al. Handwritten similar Chinese characters recognition based on combining statistics with neural networks method［J］. Journal of Chinese Information Processing, 1999, 13(3): 33-39.
相似文献/References:
[1]蒋建国,吴 琼,夏 娜.自适应粒子群算法求解Agent联盟[J].智能系统学报,2007,2(2):69.
　JIANG Jian-guo,WU Qiong,XIA Na.Solving Agent coalition using adaptive particle swarm optimization algorithm[J].CAAI Transactions on Intelligent Systems,2007,2():69.
[2]杨 力,许泽明,曹谢东,等.基于MAS的U-B-M三层总体框架下信息集成平台的研究与应用[J].智能系统学报,2007,2(5):58.
　YANG Li,XU Ze-ming,CAO Xie-dong,et al.Integrating diverse information platforms with MASbased UBM threelayer frameworks[J].CAAI Transactions on Intelligent Systems,2007,2():58.

备注/Memo

作者简介：
马少平, 男, 1961年生, 教授, 博士生导师，主要研究方向为智能信息处理、信息检索、汉字识别与后处理以及中文古籍数字化.承担过多项国家自然科学基金、“863”项目、“973”项目及国际合作项目，在脱机手写体汉字识别和后处理方面达到了国际先进水平，“脱机手写体汉字与数字识别系统”1998年1月获得国家教委科技进步二等奖.发表学术论文70余篇,出版专著2部.

金奕江，男，1970年生，工程师,主要研究方向为汉字识别、信息检索与处理.发表学术论文10余篇.

更新日期/Last Update: 2009-12-29
Copyright © 《 智能系统学报》 编辑部
地址：（150001）黑龙江省哈尔滨市南岗区南通大街145-1号楼 电话：0451- 82534001、82518134　邮箱：tis@vip.sina.com