Webcam Based Robust and Affordable Optical Mark Recognition System for Teachers

Abstract

The growing need for efficient automated grading solutions has driven advancements in optical mark recognition (OMR) systems for multiple-choice assessments. This paper introduces a novel webcam-based OMR system that employs advanced image processing and computer vision techniques to eliminate the dependency on specialized hardware. The proposed system enhances image quality, extracts relevant data, and accurately processes marked responses through a robust pipeline of preprocessing, segmentation, and recognition algorithms. Addressing challenges such as inconsistent handwriting styles and varying lighting conditions, the system demonstrates high accuracy and reliability, achieving an impressive accuracy rate of 100%. Experimental validation highlights significant improvements in grading efficiency, reduced human error, and enhanced consistency when compared to manual grading methods. The scalability of the system makes it applicable to remote learning environments, online exams, and large-scale assessment scenarios. Future research directions include integrating machine learning techniques to extend the system's capabilities to subjective assessments and potential collaborations with educational institutions and online platforms. This research contributes to the field by providing an accessible and scalable automated grading solution that optimizes assessment workflows and improves the educational experience.