Automation of Examination Evaluation using Matlab-Based Optical Mark Recognition (OMR) System

Mr. U. Pradeep Kumar1, Bhogadi Sai Sowjanya2, Appikonda Kumar Raju3, Pinninti Pavani4, Chandaka Prasanth Kumar5, Tanakala Tejaswini6
1Assistant Professor, Department of ECE & Sanketika Vidya Parishad Engineering College, Visakhapatnam,India
2,3,4,5,6 UG Students Department of ECE &Sanketika Vidya Parishad Engineering College, Visakhapatnam,India

Abstract - Multiple Choice Questions (MCQs) are widely used in modern examinations. Traditional Optical Mark Recognition (OMR) systems rely on expensive dedicated hardware or Python/OpenCV implementations that suffer from lighting sensitivity and library dependencies.This paper presents a complete MATLAB-based OMR evaluation system that processes scanned answer sheets through an integrated image- ocessing pipeline and a user-friendly Graphical User Interface (GUI). The system converts RGB images to grayscale, applies median filtering for noise removal, uses adaptive thresholding and morphological operations for bubble isolation, performs Canny edge detection, and employs region-based pixel density analysis (threshold , fill ratio ) to detect filled bubbles. Detected answers are automatically compared with an Excel answer key. The GUI, developed using MATLAB GUIDE, supports image loading, answer-key import, evaluation, and result display (roll number, marks, test ID, and performance remarks). The proposed system achieves 89–94% accuracy on standard OMR sheets, eliminates hardware dependency, and provides a robust, cost-effective solution for educational institutions.Key Words: Optical Mark Recognition (OMR), MATLAB Image Processing, adaptive thresholding, morphological operations, GUI, examination automation, MCQ evaluation