A5089850943- Handwritten Text Recognition Techniques
- Natural Language Processing Techniques
- Mathematics, Computing, and Information Processing
- Algorithms and Data Compression
- Image Processing and 3D Reconstruction
- Advanced Image and Video Retrieval Techniques
- Image Retrieval and Classification Techniques
- Text and Document Classification Technologies
- Ichthyology and Marine Biology
- Multimedia Communication and Technology
- Health Education and Validation
- Robotics and Automated Systems
- Business, Education, Mathematics Research
- Coastal and Marine Management
- Advanced Malware Detection Techniques
- Interactive and Immersive Displays
- Underwater Acoustics Research
- Music and Audio Processing
- Augmented Reality Applications
- User Authentication and Security Systems
Barcelona Institute for Science and Technology
2018
Universitat Politècnica de València
2010-2016
Centro Tecnológico de Investigación, Desarrollo e Innovación en tecnologías de la Información y las Comunicaciones (TIC)
2013
In this work, a system for recognition of printed mathematical expressions has been developed. Hence, statistical framework based on two-dimensional stochastic context-free grammars defined. This formal allows to jointly tackle the segmentation, symbol and structural analysis expression by computing its most probable parsing. order test approach reproducible comparable experiment carried out over large publicly available (InftyCDB-1) database. Results are reported using well-defined global...
In mathematical expression recognition, symbol classification is a crucial step. Numerous approaches for recognizing handwritten math symbols have been published, but most of them are either an online approach or hybrid approach. There absence study focused on offline features recognition. Furthermore, many papers provide results difficult to compare. this paper we assess the performance several well-known task. We also test novel set based polar histograms and vertical repositioning method...
We consider the difficult problem of classifying spatial relationships between symbols and subexpressions in handwritten mathematical expressions. first improve existing geometric features based on bounding boxes center points, normalizing them using distance centers two or question. then propose a novel feature set for layout classification, polar histograms computed over points strokes. A series experiments are presented which Support Vector Machine is used with these new to classify five...
Automatic recognition of printed mathematical symbols is a fundamental problem for expressions. Several classification techniques has been previously used, but there are very few works that compare different on the same database and with experimental conditions. In this work we have tested classical novelty symbol two databases.
Recognition of on-line handwritten mathematical symbols has been tackled using different methods, but the recognition rates achieved until now still leave room for improvement. Many published approaches are based on hidden Markov models, and some them use off-line information extracted from data. In this paper, we present a set hybrid features that combine both information. Lately, recurrent neural networks have demonstrated to obtain good results they outperformed models in several sequence...
Several approaches have been proposed to tackle the problem of mathematical expression recognition, and automatic methods for performance evaluation are required. Mathematical expressions usually encoded as a LaTeX string or tree (MathML) purpose, but these formats do not enforce uniqueness. Consequently, given that there can be several representations syntactically different semantically equivalent, biased. Given recognition its ground-truth tree, error is computed by comparing them. In...
Online services are often protected with captchas that typically must be solved by typing on a keyboard. Now smartphones and tablets increasingly being used to browse the web, new best suited touch-capable devices should devised, because entering text soft keyboards is usually uncomfortable error-prone. This article contributes solving this issue μcaptcha, novel captcha scheme tell humans computers apart means of math handwriting input. Instead keyboard, user retypes mathematical expression...
There are thousands of Unicode characters and hence it can be hard to visually find a particular one. For this reason, we aimed at developing tool that allows handwrite character receive list the most similar candidates input. This will integrated in math editor which handles more than 5,000 different characters. Since no public datasets were found fit our needs, crowdsourced acquisition online handwritten data for training purposes. We developed neural network combining convolutional layers...