[go: up one dir, main page]

Computer Science ›› 2015, Vol. 42 ›› Issue (Z11): 113-118.

Previous Articles     Next Articles

On-line Handwritten Flowchart Recognition Based on Grammar Description Language

CHEN Quan, SHI Da-peng, FENG Gui-huan, ZHAO Xiao-yan and LUO Bin   

  • Online:2018-11-14 Published:2018-11-14

Abstract: This paper proposed a sketch recognition approach based on a grammar description language.The whole re-cognition process is divided into three steps.Firstly,dynamic programming is used to group strokes into stroke combinations under some spatial and temporal constraints.Secondly,a neural network is employed to classify the candidate symbols.Lastly,the candidate symbols are filtered with the constituent grammar rules,and a grammar parser is utilized to achieve the recognition results.This method has been applied to a freely available database FCinkML.The results demonstrate the effectiveness and efficiency of our approach.

Key words: Sketch understanding,Handwritten flowchart,Grammar description language,Neural network

[1] Fish J,Scrivener S.Amplifying the mind’s eye:Sketching andvisual cognition [J].Leonardo,1990,23(1):117-126
[2] Yeager L S,Webb B J,Lyon R F.Combining neural networksand context-driven search for online,printed handwriting recognition in the Newton[J].AI Magazine,1996,9(1):73-89
[3] Rubine D.Specifying gestures by example[J].SIGGRAPH’ 91,1991,25(4):329-337
[4] Cohen P,Johnston M,McGee D,et al.Quickset:Multimodal interaction for distributed applications[C]∥Proceedings of the Fifth ACM International Multimedia Conference .NY:ACM Press,1997:31-40
[5] Wilcox L D,Bill N S,Nitin S.Dynomite:A Dynamically Orga-nized Ink and Audio Notebook[C]∥Proceedings of CHI’97,1997:186-193
[6] Shilman M,Viola P.Spatial Recognition and Grouping Text and Graphics[C]∥EUROGRAPHICS Workshop on Sketch-based Interface and modeling,2004
[7] 王淑侠,王关峰,高满屯,等.基于时空关系的在线多笔画手绘折线段曲线识别[J].计算机科学,2012,9(9):269-274
[8] 蒋维,张斌,孙正兴.基于自适应HMM的在线草图识别方法[J].计算机科学,2005,32(5):185-189
[9] 张莉莎,孙正兴,周若鸿,等.一种基于图匹配的复杂草图识别方法[J].计算机科学,2004,31(4):147-150
[10] 诸彬,何骅,徐从富.基于草图的程序流程图自动识别翻译系统[J].计算机应用与软件,2010,27(5):55-56
[11] 张小亮,孙根正,廖达雄,等.基于几何特征的在线手绘流程图识别[J].计算机辅助工程,2007,16(1):29-33
[12] 谢强,冯桂焕,孙正兴.基于上下文的在线草图识别方法[J].计算机科学,2007,34(3):216-219
[13] 李建新.一个基于支持向量机的草图识别系统[D].天津大学,2006
[14] Ulgen F,Flavell A,Akamatsu N.On-Line Shape recognitionwith incremental training using binary synaptic weights algorithm[J].Applied Intelligence,1996,6(3):225-240
[15] 靳简明,江红英,王庆人.数学公式识别系统:MatheReader [J].计算机学报,2006,29(11):2018-2026
[16] 袁浩,王军伟,卢章平,等.概念设计中草图类型笔触的语义理解研究[J].计算机应用研究,2013(9):2875-2877
[17] Hammond T.A domain description language for sketch recognition[R].MIT Artificial Intelligence Laboratory,September 2002
[18] Coüasnon B.DMOS,a generic document recognition method:Application to table structure analysis in a general and in a specific way[J].International Journal on Document Analysis and Recognition(IJDAR),2006,8(2):111-122
[19] Awal A M,Feng G,Mouchere H,et al.First experiments on a new online handwritten flowchart database[C]∥/IS&T/SPIE Electronic Imaging.International Society for Optics and Photo-nics,2011
[20] Lemaitre A,Mouchere H,Camillerapp J,et al.Interest of sy-ntactic knowledge for on-line flowchart recognition [M]∥Graphics Recognition.New Trends and Challenges.Springer Berlin Heidelberg,2013:89-98
[21] Bresler M,Prua D,Hlavác V.Modeling flowchart structure reco-gnition as a max-sum problem[C]∥2013 12th International Conference on Document Analysis and Recognition(ICDAR).IEEE,2013:1215-1219
[22] Blagojevic R V.Using data mining for digital ink recognition[D].ResearchSpace@ Auckland,2011
[23] 胡兴鸿,施大鹏,冯桂焕.一种手写图文分离方法[J].计算机与现代化,2013(12):145-148
[24] Holmes G,Donkin A,Witten I H.Weka:A machine learningworkbench[C]∥Proceedings of the 1994 Second Australian and New Zealand Conference on Intelligent Information Systems,1994.IEEE,1994:357-361
[25] Carton C,Lemaitre A,Couasnon B.Fusion of statistical and structural information for flowchart recognition[C]∥2013 12th International Conference on Document Analysis and Recognition(ICDAR),2013

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!