Using Earth Mover's Distance in the Bag-of-Visual-Words Model for Mathematical Symbol Retrieval
Using Earth Mover's Distance in the Bag-of-Visual-Words Model for Mathematical Symbol Retrieval
复制标题
在视觉词袋模型中使用推土机距离进行数学符号检索
DOI：
10.1109/icdar.2011.263
复制
发表时间：
2011
期刊：
影响因子：
--
通讯作者：
G. Soda
中科院分区：
文献类型：
--
作者：
S. Marinai;Beatrice Miotti;G. Soda
在本文中，地球移动器的距离（EMD）被用作数学符号检索任务中的相似性度量。该方法基于视觉词袋模型。在我们的情况下，从每个符号中提取的特征通过自组织映射（SOM）进行聚类，然后将聚类中的特征的出现累积在视觉词的向量中。后者的向量之间的比较是用EMD进行的，EMD自然允许在距离计算中结合SOM簇的拓扑组织。所提出的方法进行了实验测试的数学符号检索任务，并与余弦相似性和最近提出的一些变种。
In this paper, the Earth Mover's Distance (EMD) is used as a similarity measure in the mathematical symbol retrieval task. The approach is based on the Bag-of-Visual-Words model. In our case the features extracted from each symbol are clustered by means of Self-Organizing Maps (SOM) and then occurrences of features in the clusters are accumulated in a vector of visual words. The comparison between the latter vectors is performed with the EMD which naturally allows to incorporate the topological organization of SOM clusters in the distance computation. The proposed approach is experimentally tested in a mathematical symbol retrieval task and compared with the cosine similarity and with some variants that have been recently proposed.