A Deep Learning Approach for Recognizing Bengali Characters and Numbers

Bengali is one of the world’s most widely spoken languages. However, it is a challenging issue to recognize Bengali Alphabet and Numbers from handwritten documents. It has 50 fundamental alphabets and 10 Fundamental Numbers. Because of their diverse sizes, similarities between alphabets, and distinct writing methods, recognizing the Bengali handwritten Alphabet and Numbers is a difficult task. In this research, a deep learning-based approach is proposed where Convolutional Neural Network (CNN) and Bidirectional LSTM (BiLSTM) are combined to form the CNN-BiLSTM Model, and the classification of that model is followed by a Multi Support Vector Machine (M-SVM). The model works in a good manner to recognize Bengali Characters as well as Bengali Numbers. It also visualized how the models independently interact. This study also introduced a dataset named Bengali Handwritten Alphabet and Numbers (BHAN-2024) dataset which was applied to evaluate the perfection of the proposed system. A Morphological Operation is used before testing the input image which assists the model in removing the unnecessary parts from the input image. This study solves the recognition problem of Bengali characters and numbers using the same model with higher accuracy. The experimental findings show that the suggested model recognizes Bengali handwritten Alphabets as well as Numbers with an average accuracy of 97.08%.