Structural analysis of handwritten mathematical expressions using contextual information

• Grant number: 13/13535-0
• Support Opportunities: Scholarships abroad - Research Internship - Doctorate (Direct)
• Start date: October 21, 2013
• End date: October 20, 2014
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Nina Sumiko Tomita Hirata
• Grantee: Frank Dennis Julca Aguilar
• Supervisor: Christian Viard-Gaudin
• Host Institution: Instituto de Matemática e Estatística (IME). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Institution abroad: Université de Nantes, France
• Associated to the scholarship: 12/08389-1 - The use of contextual information for the recognition of online handwritten mathematical expressions, BP.DD

Abstract

Online recognition of handwritten mathematical expressions is an interesting and challenging problem in the field of Pattern Recognition. It is interesting due to several application possibilities and it is challenging because it involves a large set of symbol classes and the recognition of symbol arrangements in 2D structure. Structural analysis strongly depends on the correct understanding of relative positions among symbols. Based on recent results of online handwritten mathematical expression recognition contests, it seems that structural analysis is where major improvements are still needed. The aim of this one year long project is to develop new methods for the structural analysis of handwritten mathematical expressions. In particular, we plan to investigate and propose ways to determine the correct relationship between two symbols in an expression, by modeling both structural and semantic information. Structural information at different levels of organization as well as fuzzy relations and context information should be used. The proposed model performance will be compared to the ones reported in the contests, using the same datasets. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

JULCA-AGUILAR, FRANK; HIRATA, NINA S. T.; MOUCHERE, HAROLD; VIARD-GAUDIN, CHRISTIAN; IEEE. Subexpression and Dominant Symbol Histograms for Spatial Relation Classification in Mathematical Expressions. 2016 23RD INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), v. N/A, p. 6-pg., 2016-01-01. (13/13535-0)

HIRATA, NINA S. T.; JULCA-AGUILAR, FRANK D.. Matching based ground-truth annotation for online handwritten mathematical expressions. PATTERN RECOGNITION, v. 48, n. 3, p. 837-848, MAR 2015. (12/08389-1, 13/13535-0)