在线手绘态势图的分割和识别

系统仿真学报 ›› 2016, Vol. 28 ›› Issue (8): 1884-1891.

在线手绘态势图的分割和识别

邓维^1,2, 吴玲达¹, 张友根³, 杨超¹

1.装备学院复杂电子系统仿真实验室,北京 101416;
2.75711部队,广州 510515;
3.国防信息学院信息系统系,武汉 430010

收稿日期:2015-02-03 修回日期:2015-03-19 出版日期:2016-08-08 发布日期:2020-08-17
作者简介:邓维(1986-),男,湖北武汉,博士,研究方向为空间信息处理;吴玲达(1962-),女,上海,博士,教授,研究方向为空间信息处理。
基金资助:
核高基国家科技重大专项(2013ZX01045- 004-002)

Segmentation and Recognition of Hand-drawn Situation Maps

Deng Wei^1,2, Wu Lingda¹, Zhang Yougen³, Yang Chao¹

1. Science and Technology on Complex Electronic System Simulation Laboratory, Equipment Academy, Beijing 101416, China;
2. Unit 75711 of PLA, Guangzhou 510515, China;
3. Department of Information Systems, Academy of National Defense Information, Wuhan 430010, China

Received:2015-02-03 Revised:2015-03-19 Online:2016-08-08 Published:2020-08-17

摘要/Abstract

摘要： 当前对在线手绘军标识别的研究通常以单个孤立的样本作为识别对象。在实际的笔式标绘应用中,以整幅手绘态势图作为对象进行分割和识别有助于保持用户思维连贯性,但对识别算法提出了更大的挑战。分析了手绘态势图的特点和分割难点,提出一种基于动态规划的分割和识别方法。通过基于最小生成树的笔画聚类对态势图进行粗分割;通过置信度转换和融合的方法综合利用单图符识别信息和几何信息,计算候选分割路径的评价准则;采用动态规划搜索最优分割路径。实验表明了方法的有效性,该方法有助于扩展草图识别的应用范围。

关键词: 草图识别, 分割, 态势图, 军标图符, 动态规划

Abstract: Most of current research on online sketched military symbol recognition concerns the isolated samples. In pen-based situation marking systems, users prefer segmenting and recognizing the full sketched situation maps. However, it is a more difficult problem. The characteristics of hand-drawn situation maps and difficulties to segment them were analyzed, and a dynamic programming-based approach was proposed. A hand-drawn situation map was segmented coarsely by minimum spanning tree-based stroke clustering. The candidate segmentation path was evaluated by synthesizing geometric analysis and isolated symbol classifier together, using confidence transformation and fusion. The dynamic programming algorithm was used to search the optimal segmentation and recognition result. Experimental results demonstrate that the proposed method is effective, which extends the applications of sketch recognition.

Key words: sketch recognition, segmentation, situation maps, military symbols, dynamic programming

中图分类号:

TP391

邓维, 吴玲达, 张友根, 杨超. 在线手绘态势图的分割和识别[J]. 系统仿真学报, 2016, 28(8): 1884-1891.

Deng Wei, Wu Lingda, Zhang Yougen, Yang Chao. Segmentation and Recognition of Hand-drawn Situation Maps[J]. Journal of System Simulation, 2016, 28(8): 1884-1891.

参考文献 25

[1]	陈鸿, 汤晓安, 杨耀明, 等. 基于位移映射的非规则军队标号绘制算法[J]. 计算机辅助设计与图形学学报, 2011, 23(5): 797-804.
[2]	吴玲达, 张友根, 邓维, 等. 面向手绘军标图形的旋转自由识别方法[J]. 中国图象图形学报, 2014, 19(3): 456-462.
[3]	邓维, 吴玲达, 张友根, 等. 采用快速图元匹配的手绘非规则军标识别[J]. 计算机辅助设计与图形学学报, 2014, 26(10): 1835-1843.
[4]	Hammond T, Logsdon D, Peschel J, et al.A sketch recognition interface that recognizes hundreds of shapes in course-of-action diagrams[C]// Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI). Atlanta, USA: ACM, 2010: 4213-4218.
[5]	Cummings D, Fymat S, Hammond T.Sketch-based interface for interaction with unmanned air vehicles[C]// Proceedings of the ACM SIGCHI Conference on Human Factors in Computing Systems. Austin, Texas, USA: ACM, 2012.
[6]	Davis R.Magic paper: sketch-understanding research[J]. IEEE Computer (S 0018-9162), 2007, 40(9): 34-41.
[7]	Stahovich T F, Peterson E J, Lin H.An efficient, classification-based approach for grouping pen strokes into objects[J]. Computers & Graphics (S0097-8493) , 2014, 42(8): 14-30.
[8]	Kara L B, Stahovich T F.Hierarchical parsing and recognition of hand-sketched diagrams[C]// Proceedings of the 17th annual ACM symposium on User interface software and technology. USA: ACM, 2004: 13-22.
[9]	Gennari L, Kara L B, Stahovich T F, et al.Combining geometry and domain knowledge to interpret hand-drawn diagrams[J]. Computers & Graphics (S0097-8493), 2005, 29(4): 547-562.
[10]	Jiang Y, Tian F, Wang X, et al.Structuring and manipulating hand-drawn concept maps[C]// Proceedings of the 14th International Conference on Intelligent User Interfaces. USA: ACM, 2009: 457-462.
[11]	Ouyang T Y, Davis R.ChemInk: a natural real-time recognition system for chemical drawings[C]// Proceedings of the 16th International Conference on Intelligent User Interface (IUI). USA: ACM, 2011: 267-276.
[12]	Feng Guihuan, Viard-gaudin C, Sun Zhengxing. On-line hand-drawn electric circuit diagram recognition using 2D dynamic programming[J]. Pattern Recognition (S0031-3203), 2009, 42(12): 3215-3223.
[13]	Bresler M, Pruša D, Hlavác V.Simultaneous Segmentation and Recognition of Graphical Symbols using a Composite Descriptor[C]// 18th Computer Vision Winter Workshop (CVWW). Vienna, Austria: Pattern Recognition and Image Processing Group, 2013: 16-23.
[14]	Arandjelovic Relja, Sezgin Tevfik Metin.Sketch recognition by fusion of temporal and image-based features[J]. Pattern Recognition (S0031-3203), 2011, 44(6): 1225-1234.
[15]	Sun Zhengbang, Wang Changhu, Zhang Liqing, et al.Free Hand-Drawn Sketch Segmentation[C]// ECCV. Berlin, Germany: Springer Berlin Heidelberg, 2012: 626-639.
[16]	Peterson E J, Stahovich T F, Doi E, et al. Grouping strokes into shapes in hand-drawn diagrams [C]// AAAI 2010. USA: AIAA, 2010.
[17]	Stevens P C, Blagojevic R, Plimmer B.Supervised machine learning for grouping sketch diagram strokes[C]// Proceedings of the International Symposium on Sketch-Based Interfaces and Modeling. USA: ACM, 2013: 43-50.
[18]	王长虎, 芮勇. 数字墨水技术: 数字世界中的神来之笔[J]. 中国计算机学会通讯, 2014, 10(7): 26-30.
[19]	张友根, 吴玲达, 邓维, 等. 融合时空上下文的手绘笔画图文分类[J]. 电子与信息学报, 2013, 35(1): 113-118.
[20]	于金伦, 周祥东, 刘成林. 手写字符串识别搜索算法研究[J]. 模式识别与人工智能, 2009, 22(2): 182-187.
[21]	Liu Chenglin, Hao Hongwei, Sako H.Confidence transformation for combining classifiers[J]. Pattern Analysis and Applications (S1433-7541), 2004, 7(1): 2-17.
[22]	Liu Cheng-lin.Classifier combination based on confidence transformation[J]. Pattern Recognition (S0031-3203), 2005, 38(1): 11-28.
[23]	Wang Qiufeng, Yin Fei, Liu Chenglin.Improving handwritten Chinese text recognition by confidence transformation[C]// International Conference on Document Analysis and Recognition. Washington, DC, USA: IEEE Computer Society, 2011: 518-522.
[24]	谢毓湘, 栾悉道, 陈丹雯, 等. 一种基于局部不变特征的图像特定场景检测方法[J]. 国防科技大学学报, 2013, 35(6): 78-83.
[25]	Herold J, Stahovich T F.A machine learning approach to automatic stroke segmentation[J]. Computer & Graphics (S0097-8493), 2014, 38(2): 357-364.