Comparative Analysis of Classification Algorithms for Printed script

A robust system for translating Telugu text from images to English is developed using Optical Character Recognition (OCR) and deep learning techniques.Tesseract OCR is utilized for detecting Telugu script, followed by translation with Facebook's mBART, a multilingual sequenceto-sequence model.Google Translate is also used for comparison purposes.The translation model's performance is evaluated with the BLEU score, which measures the similarity between machine translations and human-generated references.The system was tested on a dataset of Telugu images with corresponding English meanings.Various preprocessing techniques, such as resizing, filtering, and text segmentation, were applied to improve OCR accuracy.The outcomes demonstrate the effectiveness of integrating traditional OCR methods with advanced neural machine translation, providing a scalable solution for Telugu-to-English translation across various applications like document digitization, education, and accessibility.