Hindi Character Recognition Dataset

The dataset is focused on the classification of handwritten Devanagari script, a significant writing system used primarily in India and Nepal. This script comprises 36 characters representing the sounds of the Sanskrit language and 10 numerical digits. Unlike many Western scripts, Devanagari does not employ capitalization, providing a unique challenge for recognition algorithms. Each character in the Devanagari script is connected by a distinctive horizontal bar, known as the ‘shirorekha,’ which runs along the top of the script, unifying the characters into cohesive words and sentences.