LDA单词图像表示的蒙古文古籍图像关键词检索方法  

LDA-Based Word Image Representation for Keyword Spotting on Historical Mongolian Documents

作　　者：白淑霞[1] 鲍玉来[1] Bai Shuxia;Bao Yulai(Library,Inner Mongolia University,Hohhot 010021,China)

机构地区：[1]内蒙古大学图书馆,内蒙古呼和浩特010021

出　　处：《现代情报》2017年第7期51-54,88,共5页Journal of Modern Information

基　　金：国家自然科学基金项目"基于领域本体的蒙古文数字资源整合机制研究"(项目编号:71163029)

摘　　要：[目的 ]为了克服传统视觉词袋方法(Bag-of-Visual-Words)中忽略视觉单词间的空间关系和语义信息等问题。[方法 ]本文提出一种与视觉语言模型相结合的基于LDA主题模型,并采用查询似然模型实现检索。[结果 ]实验数据表明,本文所提出的基于LDA的表示方法可以高效、准确地解决蒙古文古籍的关键词检索问题。[结论 ]同时,该方法的性能比Bo VW方法有显著提高。[Objective]In order to overcome the problem of ignoring the spatial relations and semantic information between visual words in traditional visual word bag(Bag-of-Visual-Words).[Methods]In this paper,a LDA-based topic model was adopted which was linearly combined with a visual language model for each word image.And the basic query likelihood model was used for realizing the procedure of retrieval.[Results]The experimental results on our dataset showed that the proposed LDA-based representation approach could effi ciently and accurately attain to the aim of keyword spotting on a collection of historical Mongolian documents.[Conclusions]Meanwhile,the proposed approach improved the performance signifi cantly than the original BoVW approach.

关 键 词：隐含狄利克雷分配(LDA) 主题模型 视觉语言模型 蒙古文古籍 关键词检索 查询似然模型 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]