Efficient Approach to Using CNN-Based Pre-trained Models in Bangla Handwritten Digit Recognition

Due to digitalization in everyday life, the need for automatically recognizing handwritten digits is increasing. Handwritten digit recognition essential numerous applications various industries. Bengali ranks fifth largest language world with 265 million speakers (Native and non-native combined) 4 percent of population speaks Bengali. complexity writing terms variety shape, size, style, researchers did not get better accuracy using Supervised machine learning algorithms date. Moreover, fewer studies have been done on Bangla (BHwDR). In this paper, we proposed a novel CNN-based pre-trained model which includes Resnet-50, Inception-v3, EfficientNetB0 NumtaDB dataset 17 thousand instances 10 classes.. The Result outperformed performance other models date 97% 10-digit classes. Furthermore, evaluated result or our research while suggesting future study