手绘图识别的用户适应性研究  

User Adaptation to On-Line Sketch Graphic Recognition

作　　者：连帅[1] 孙根正[1] 廖达雄[1] 

机构地区：[1]西北工业大学机电学院,陕西西安710072

出　　处：《计算机仿真》2008年第8期189-192,230,共5页Computer Simulation

摘　　要：草图理解的关键问题是允许用户按其习惯方式输入图形,即草图理解的用户适应性。为此提出了一种在线草图识别用户适应性解决方法:首先对用户输入的图形进行预处理;然后利用转角函数的方法进行笔划特征点的抽取;最后用动态用户建模技术进行笔划和复杂图形的识别。具有很强的用户适应性,消除了因用户随意输入带来的识别困难,为草图理解的实现提供支持。实验表明所提出的方法不仅能得到较好的检索结果,而且具有较好的用户适用性,为解决在线草图识别及其用户适应性问题提供了有益的尝试。User adaptation is an important problem in user interface study. A method of user modeling for online composite sketchy shape recognition is presented. The freehand drawing is pretreated firstly, the future point is detected in terms of computable metric, Using this attribute, people can predict the resulting shape more efficiently and make dynamic pruning to the noisy or exceptional nodes. Experiments show that the proposed methods not only can improve the efficiency but also have effective user adaptation, thus providing a beneficial trial for resolving the problems of online sketch identification and its user adaptability.

关 键 词：在线草图识别 用户适应性 用户建模 决策树 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TP391.41[自动化与计算机技术—计算机科学与技术]