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.

Using Earth Mover's Distance in the Bag-of-Visual-Words Model for Mathematical Symbol Retrieval / Simone Marinai;Beatrice Miotti;Giovanni Soda. - STAMPA. - (2011), pp. 1309-1313. (Intervento presentato al convegno International Conference on Document Analysis and Recognition tenutosi a Beijing (China) nel Sept. 2011) [10.1109/ICDAR.2011.263].

Using Earth Mover's Distance in the Bag-of-Visual-Words Model for Mathematical Symbol Retrieval

MARINAI, SIMONE;MIOTTI, BEATRICE;SODA, GIOVANNI
2011

Abstract

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.
2011
Proceedings International Conference on Document Analysis and Recognition 2011
International Conference on Document Analysis and Recognition
Beijing (China)
Sept. 2011
Simone Marinai;Beatrice Miotti;Giovanni Soda
File in questo prodotto:
File Dimensione Formato  
icdar11_miotti-CR.pdf

accesso aperto

Tipologia: Versione finale referata (Postprint, Accepted manuscript)
Licenza: Tutti i diritti riservati
Dimensione 231.2 kB
Formato Adobe PDF
231.2 kB Adobe PDF

I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificatore per citare o creare un link a questa risorsa: https://hdl.handle.net/2158/597400
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 14
  • ???jsp.display-item.citation.isi??? 10
social impact