基于Word Spotting技术的蒙古文古籍图像检索中的特征选择  

Feature selection in word spotting technology for retrieving historical Mongolian document images

作　　者：魏宏喜[1] 高光来[1] 

机构地区：[1]内蒙古大学计算机学院,呼和浩特010021

出　　处：《计算机应用》2011年第11期3038-3041,共4页journal of Computer Applications

基　　金：国家自然科学基金资助项目(60865003);教育部"春晖"计划项目(Z2009-1-01012)

摘　　要：设计了一个基于word spotting技术的蒙古文《甘珠尔经》图像检索的系统框架。在充分分析了蒙古文《甘珠尔经》中手写单词图像特点的基础上,提出了采用轮廓特征、投影特征和笔划穿越数目来表示单词图像。在由5500个单词图像构成的数据集上进行对比实验,确定了最佳的特征组合,平均准确率(MAP)能达到78.79%,R-Precision能达到73.01%。实验结果表明,所选择的特征是合理的、有效的。A systematic frame for retrieving Mongolian Kanjur images by word spotting technology was designed.Some features including profile feature,projection feature and back-to-ink transition were adopted to represent Mongolian word images by deeply analyzing characteristics of handwritten Mongolian word images.By doing a lot of comparison experiments on the dataset with 5 500 word images,the best combination of features was determined.Meanwhile,the Mean Average Precision（MAP） of about 78.79% was achieved and the R-Precision 73.01%.The experimental results show that the selective features are valid and effective.

关 键 词：蒙古文古籍图像 甘珠尔经 WORDS POTTING 文档图像检索 轮廓特征 动态时间弯曲 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]