CROHME_2011 | 2014 Lei Hu , Richard Pospesel

CROHME_2011 is a C# library. CROHME_2011 has no bugs, it has no vulnerabilities, it has a Strong Copyleft License and it has low support. You can download it from GitHub.

Copyright (c) 2011-2014 Lei Hu, Richard Pospesel, Kevin Hart, Richard Zanibbi. This file is part of DPRL CROHME 2011. DPRL CROHME 2011 is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version. DPRL CROHME 2011 is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with DPRL CROHME 2011. If not, see This document is about DPRL's submission for the CROHME 2011. CROHME is the abbreviation of Competition on Recognition of Online Handwritten Mathematical Expression. The DPRL CROHME 2011 has a feed-feedforward architecture. For the segmentation, The time-ordered sequence of strokes for an expression is broken down into groups of three; for n strokes, the system produces a list of stroke subsets, i.e. S = ((s1, s2, s3), …, (sn-2, sn-1,sn)); if n is not divisible by 3, the final stroke subset may have only one or two strokes. For each three-stroke group, an HMM classifier is used to compute the highest classification probability for each possible segment, and then the segmentation that maximizes the harmonic mean of the resulting class probabilities is selected for the stroke group. The HMM used for classification is the same as the paper HMM-Based Recognition of Online Handwritten Mathematical Symbols Using Segmental K-means Initialization and a Modified Pen-up/down Feature. The details about preprocessing, feature selection and HMM algorithm can be found in the paper. The source code about the HMM classifier can be found at HMM classifier source code. The spatial relation between symbols is determined using two probabilistic quadratic classifiers. The details can be found in the paper Baseline Extraction-Driven Parsing of Handwritten Mathematical Expressions. Finally, once symbols are segmented and recognized, the DRACULAE parser recovers the expression structure using default parameters. More details about the parser can be found at DRACULAE parser.