Can I read Data Augmentation for Arabic HTR on EtoBox?

Data Augmentation for Arabic HTR by braintumor2019 is a document available to read on EtoBox.

What is Data Augmentation for Arabic HTR about?

This paper presents a new data augmentation technique for Offline Arabic Handwritten Text Recognition (HTR) using Moving Least Squares (MLS) to enhance the training datasets. The authors demonstrate that their method, combined with a Convolutional Recurrent Neural Network (CRNN), significantly improves recognition performance on the Arabic IFN/ENIT database, especially with smaller input sizes. The results indicate that the MLS technique generates realistic augmented images that help the recognition system

Author

braintumor2019

Language

EN