Single deep neural network trained on synthetic EMNIST-derived data performs joint detection and classification of handwritten Latin letters and reaches 88.28 percent on real exam forms.
EAI Endorsed Trans
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Intelligent Character Recognition of Handwritten Forms with Deep Neural Networks
Single deep neural network trained on synthetic EMNIST-derived data performs joint detection and classification of handwritten Latin letters and reaches 88.28 percent on real exam forms.