Bangla Hand Written Character and Word Recognition Dataset

This handwritten recognition dataset was manually created for academic and research purposes. The dataset contains 85 distinct classes consisting of both handwritten characters and words.

The samples were collected from 50 students of Daffodil International University, ensuring diversity in handwriting styles, stroke patterns, and writing variations. This diversity helps improve the robustness and generalization ability of machine learning and computer vision models trained on the dataset.

The dataset is designed to support research and development in machine learning, computer vision, and pattern recognition, particularly for handwritten character and word recognition tasks. It can be used for training, testing, and benchmarking various algorithms for handwritten text classification and recognition systems.