Integrating Handwritten Digit Recognition with Learning Management Systems for Evaluated Answer Scripts

The manual evaluation and tallying of handwritten answer scripts is an extremely laborious and time-consuming process. With thousands of answer sheets to process, educators face severe challenges in promptly compiling results and providing feedback. Automating the tallying and analysis of evaluated answer sheets through handwritten digit recognition can greatly alleviate this burden. Recent advances in computer vision and deep learning have led to highly accurate algorithms for recognizing handwritten digits. By leveraging these techniques, answer sheets containing marks and scores given by human evaluators can be efficiently tallied and analyzed at scale. Consequently, in an effort to help the researchers overcome the current challenges, we have attempted to establish a foundation for the next studies in the field. The most appropriate and effective approach for digit recognition was determined by reviewing and comprehending the current approaches and techniques for handwritten digit recognition. A total of sixty thousand photos with a pixel size of 28 × 28 were utilized as training sets. The original image was matched with the training sets and photos. It was discovered following thorough study and review that it makes it simple for the staff to add up the graded response scripts and makes student analysis easier. A range of handwritten digit recognition methods were investigated and studied in this work.