An Image Processing Oriented Optical Mark Reader

Abstract

This project is a development of an optical mark reader that can be used for counting the examination score from the multiple-choice answer sheet. The system is developed based on PC-type micro computer connecting to an image scanner or web cam. The image of sheet taken from webcam or scanner is converted into binary image using adaptive binarization method; skew correction is done by detecting line using though transform. Binary image is cut into Roll number region and Answer region using reference square fixed on answer sheet. The system operation scan be distinguished into two models:learning mode and operation mode. In the learning mode,system separates each static block in a sheet using horizontal and vertical projection after doing morphological operation. Each block is divided into ten equal parts and mark area for each question and mark cells for each option of each question. Operational mode is used to check and decide the answer. For answer checking purpose, the number of black pixels in each answer block is counted and decision is created from comparison of number of black pixel between different cells of each question. Similar process is used to extract Roll number. Score is calculated by comparing correct answer and marked answer. Finally, the list of Roll number and scores can be exported into data sheet.