Research
My research interests are basically directed towards pattern recognition and machine learning. Here are some topics I’ve been working on the last 20 years
Handwriting recognition, segmentation, and verification.
Signature and Writer Verification/Identification.
Classification and Regression models (including deep learning models)
Dynamic Selection of Classifiers/Regressors.
Concept Drift.
Dissimilarity Pattern Recognition.
Constrative Learning.
Texture Classification.
Academic Cooperation
Over the past few years, I have had the privilege of collaborating with researchers hailing from diverse corners of the world. My work has been supported by multiple funding agencies, reflecting the interdisciplinary nature and significance of my investigations. Here are some of my collaborations:
SticAmsud [2024-2025]
SticAmsud [2019-2020]
SticAmsud [2014-2015]
FAPs-INRIA-CNRS [2017-2019]
ORAND S/A (CONICYT, Chile) [2012-2013]
Intel Strategic Research Alliance [2012-2015]
IBM Research with Paulo Cavalin [2013-2018]
CAPES/FCT (INESC-Porto) with Fabien Gouyon [2009-2012]
Université du Quebec/ETS (Canada) with Prof. Robert Sabourin, Prof. Alessandro Koerich, and Prof. Rafael Menelau Cruz [2005- ]
Université de Rouen (France) with Prof. Laurent Heutte [2015-]
UFPE, with Prof. George Cavalcanti [2015-]
PUCPR, with Prof Alceu S. Britto Jr [2004-]
Datasets
Below are several datasets that my students have built over the past few years.
Dataset | Category | Download |
---|---|---|
Forest Species Database- Macroscopic | Texture | link |
Forest Species Database - Microscopic | Texture | link |
Charcoal Dataset | Texture | link |
Brazilian Forensic Letter Database | Handwriting | link |
Endangered Forest Species | Texture | link |
#DeOlhoNosCorais | Texture | link1 link2 |
Parking Lot Database | Texture, Object | link |
Fault Analysis Database | Tabular, Regression | link |
Breast Cancer Histopathological Database - BreakHis | Texture, Medical Image | link |
SignWriting | Handwriting | link |
Dopamine Release | Texture | link |