Leveraging a Custom Multi Modal OCR System for Marksheet Data Extraction

In the current educational landscape, the transition towards digitalization has become crucial. However, the manual entry of data from traditional physical marksheets into digital systems remains a significant bottleneck in the process. Educators often spend significant amount of time and effort entering data into electronic systems, which could be better utilized for more productive tasks.In this paper, we offer a solution to this problem. We propose an automated system that leverages Optical Character Recognition (OCR) Technology to streamline the extraction of essential information from marksheets. We discuss about the Dataset, preprocessing on the input image, Multi Modal OCR system Structure and other implementation details. We conclude the paper by a comparative study of our Multi Modal OCR system with other Commercially available OCRs like EasyOCR, PaddleOCR and PyTesseract.