A Machine Learning and OMR -Based Authenticated Coverless Image Steganography

Schools and universities use the Optical Mark Recognition (OMR) system frequently to scan multiple-choice tests. In this procedure, we provided an automated method based on digital image processing that is utilised to correct Multiple Choice Questions (MCQ). The steganography method involves hiding sensitive information within carriers like digital images, text, and other types of media. Current image steganography techniques, producing the steno-image as a result. These adjustments enable stegno analysis algorithms to recognise the hidden message that is embedded. In order to address this issue, a coverless data concealing idea is suggested. This is an innovative and quick technology that enables the processing of hundreds or thousands of OMR answer sheets. The technique relies on a key points detection algorithm and the development of a template answer sheet. The suggested solution uses thresholds, vertical and horizontal projections, and ways to automatically calculate the number of right answers. The suggested strategy can identify multiple choices or none at all. The study's findings include 100% accuracy on five different MCQ paper formats and less than one second processing time for each exam. The effectiveness of the suggested system will also be evaluated against various scholarly studies.