[1]刘 敏,王国利.手写运动的协作基元合成分析方法[J].智能系统学报,2010,5(05):405-410.[doi:10.3969/j.issn.1673-4785.2010.05.005]
 LIU Min,WANG Guo-li.Handwriting movement analysis by synthesis of synergic primitives[J].CAAI Transactions on Intelligent Systems,2010,5(05):405-410.[doi:10.3969/j.issn.1673-4785.2010.05.005]
点击复制

手写运动的协作基元合成分析方法(/HTML)
分享到:

《智能系统学报》[ISSN:1673-4785/CN:23-1538/TP]

卷:
第5卷
期数:
2010年05期
页码:
405-410
栏目:
学术论文—智能系统
出版日期:
2010-10-25

文章信息/Info

Title:
Handwriting movement analysis by synthesis of synergic primitives
文章编号:
1673-4785(2010)05-0405-06
作者:
刘   敏王国利
中山大学 信息科学与技术学院,广东 广州 510006
Author(s):
LIU Min WANG Guo-li
School of Information Science and Technology, Sun Yatsen University, Guangzhou 510006, China
关键词:
手写运动运动基元合成分析非负矩阵分解生物运动神经控制相似最大化准则
Keywords:
handwriting movement movement primitives synthetic analysis non-negative matrix factorization biological motor control correlative maximization principle
分类号:
TP39
DOI:
10.3969/j.issn.1673-4785.2010.05.005
文献标志码:
A
摘要:
研究一种生物运动神经控制机理与数据合成分析相结合的手写运动分析方法.特别地,将运动协作基元的概念用于手写运动数据分析,研究手写运动的协作基元合成分析方法,建立符合生物运动神经控制规律的手写运动数据理解模式.提出的协作基元合成分析过程由2个交替迭代的优化算法组成:其一,基于非负矩阵因子分解模式估计协作基元及调制幅度;其二,采用相似性最大化准则估计协作基元的激活时间.针对笔画切分的实验研究表明,采用协作基元合成分析方法获得的笔画切分结果,能够很好揭示相邻笔画之间的重叠连接模式,证实了所提方法的有效性.
Abstract:
A novel methodology of handwriting analysis was explored in relation to the biological motor control hypothesis and the data analysis by a synthetic approach. In particular, the main concern was the issue of handwriting data analysis with movement primitives, in which the analysis by a synthetic approach was presented to build a data understanding paradigm that respects the synergic hypothesis of the biological motor control. The proposed synthetic analysis approach was comprised of two alternately iterative optimization algorithms; the non-negative matrix factorization paradigm was applied for primitive decompositions while the correlative maximization principle was employed for estimating activation time of synergic primitive. The experimental studies for stroke segmentation obtained by the synthesis of synergic primitives validate the proposed method by showing that the superposed connection mode of neighboring strokes can be recognized from handwriting data.

参考文献/References:

[1]WOLPERT D M, GHAHRAMANI Z. Computational principles of movement neuroscience[J]. Nature Neuroscience, 2000, 3: 1212-1217.
[2]WINTER D A. Biomechanics and motor control of human movement[M]. 2nd ed. New York: John Wiley and Sons, 2004: 14-44.
[3]JENKINS O C, MATARIC M J. Deriving action and behavior primitives from human motion data[C]//Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems. Lausanne, Switzerland, 2002: 2551-2556.
[4]BIZZI E. Motor primitives and rehabilitation[C]//Proceedings of Virtual Rehabilitation. Venice, Italy, 2007: 20-22.
[5]SCHOMAKER L. From handwriting analysis to pen-computer applications[J]. Electronics & Communication Engineering Journal, 1998, 10(3): 93-102.
[6]MARQUARDT C, MAI N. A computational procedure for movement analysis in handwriting[J]. Journal of Neuroscience Methods, 1994, 52(1): 39-45.
[7]PLAMONDON R, DJIOUA M. A multi-level representation paradigm for handwriting stroke generation[J]. Human Movement Science, 2006, 25(4/5): 586-607.
[8]POLYAKOV F, DRORI R, BEN-SHAUL Y, et al. A compact representation of drawing movements with sequences of parabolic primitives[J]. PLoS Computational Biology, 2009, 5(7): 1-18.
[9]BYE R T. The BUMP model of response planning[D]. Sydney, Australia: The University of New South Wales, 2009.
[10]LATASH M L, SCHOLZ J P, SCHONER G. Toward a new theory of motor synergies[J]. Motor Control, 2007, 11(3): 276-308.
[11]GHAHRAMANI Z. Building blocks of movement[J]. Nature, 2000, 407(12): 682-683.
[12]HOFER J, GRUBER C, SICK B. Biometric analysis of handwriting dynamics using a script generator model[C]//Proceedings of IEEE Mountain Workshop on Adaptive and Learning Systems. Logan, USA, 2006: 36-41.
[13]FLASH T, HANDZEL A. Affine differential geometry analysis of human arm movements[J]. Biological Cybernetics, 2007, 96(6): 577-601.
[14]MUSSA-IVALDI F A, BIZZI E. Motor learning through the combination of primitives[J]. Philosophical Transactions: Biological Sciences, 2000, 355(1404): 1755-1769.
[15]HOLTSBERG A, STERNBY J. Segmentation-based recognition: United States, US 2006/0110041 A1[P]. 2005-11-09.
[16]NOVAK K E, MILLER L E, HOUK J C. The use of over-lapping submovements in the control of rapid hand movements[J]. Experimental Brain Research, 2002, 144: 351-364.
[17]GLASER E M, RUCHKIN D S. Principles of neuronbiological signal analysis[M]. New York: Academic Press, 1976: 239-245.
[18]ROHRER B, FASOLI S, KREBS H I, et al. Movement smoothness changes during stroke recovery[J]. Journal of Neuroscience, 2002, 22 (18): 8297- 8304.
[19]VINJAMURI R, CRAMMOND D J, KONDZIOLKA D, et al. Extraction of sources of tremor in hand movements of patients with movement disorders[J]. IEEE Transactions on Information Technology in Biomedicine, 2009, 13(1): 49-56.
[20]DIPIETRO L, KREBS H I, FASOLI S E, et al. Changing motor synergies in chronic stroke[J]. J. Neurophysiology, 2007, 98 (2): 757-768.
[21]LEE J, YOON H S, SOH J, et al. Using geometric extrema for segment-to-segment characteristics comparison in online signature verification[J]. Pattern Recognition, 2004, 37: 93-103.

备注/Memo

备注/Memo:
收稿日期:2009-12-10.
基金项目:国家自然科学基金资助项目(60775055).
通信作者:王国利. E-mail: isswgl@mail.sysu.edu.cn.
作者简介:
刘    敏,女,1976年生,博士研究生,主要研究方向为生物信息处理.
王国利,男,1965年生,教授、博士生导师,德国洪堡学者,中国人工智能学会智能空天系统专业委员会委员,主要研究方向为信息获取与信息处理. 主持完成省部级科研项目6项,获得省部级科技进步奖2次.
更新日期/Last Update: 2010-11-26