NumtaDB - Assembled Bengali Handwritten Digits

TL;DR

NumtaDB is a comprehensive, bias-free dataset of over 85,000 handwritten Bengali digit images designed to facilitate benchmarking and development of digit recognition algorithms.

Contribution

The paper introduces NumtaDB, a large, publicly available dataset of Bengali handwritten digits, including its collection, curation process, and key statistics.

Findings

Dataset contains over 85,000 images.

Biases from location, gender, and age are minimized.

Provides a valuable resource for Bengali digit recognition research.

Abstract

To benchmark Bengali digit recognition algorithms, a large publicly available dataset is required which is free from biases originating from geographical location, gender, and age. With this aim in mind, NumtaDB, a dataset consisting of more than 85,000 images of hand-written Bengali digits, has been assembled. This paper documents the collection and curation process of numerals along with the salient statistics of the dataset.