ScoreSync: Automated Mark Entry App using Image Capture with Computer Vision and Optical Character Recognition
International Journal of Innovative Computing · 28.07.2026 · 16 (1-2) · 287-295
DOI: 10.11113/ijic.v16n1-2.703 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Innovative Computing, online erschienen am 28.07.2026. Verfasst von Muhammad Norazlan Abdul Samad und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Traditional score recording, whereby educators manually key in scores in their respective education hubs, is time-consuming and susceptible to error when the volume of papers is high. To address this, ScoreSync, an automated mark entry app, was developed to streamline the grading and recording process by leveraging computer vision and machine learning. The primary objective is to develop an application capable of capturing images of exam papers using a lens or phone camera, ensuring a clear and accurate representation of score sheets. This would not only reduce an educator’s administrative burdens but also improve data accuracy and efficiency. Created to be applied by Android devices, ScoreSync captures images of graded papers and extracts scores through Optical Character Recognition (OCR) guided by YOLOv8, a Convolutional Neural Network (CNN) model optimised for bounding box detection. Developed through a Rapid Development Model, the app has been through multiple refines when it comes to the Artificial Intelligence (AI) model, transitioning from TensorFlow to PyTorch for greater flexibility and accuracy. The initial build of the ScoreSync is solely for the purpose of extracting graded papers of any kind, with results ranging from 80% to 90% in terms of detecting the scores in the exam sheets. Future improvements include expanding compatibility across platforms, refining the model adaptability towards different exam paper styles, and a setup that would aid the software when a live-capture solution is implemented, which would minimise more time spent rather than having the educators snap each exam sheet.
Zitieren
APA
Samad, M. N. A., Anuar, A. N., & Mansor, H. (2026). ScoreSync: Automated Mark Entry App using Image Capture with Computer Vision and Optical Character Recognition. International Journal of Innovative Computing, 16(1-2), 287-295. https://doi.org/10.11113/ijic.v16n1-2.703
Vancouver
Samad MNA, Anuar AN, Mansor H. ScoreSync: Automated Mark Entry App using Image Capture with Computer Vision and Optical Character Recognition. International Journal of Innovative Computing. 2026;16(1-2):287-295. doi:10.11113/ijic.v16n1-2.703
BibTeX
@article{Samad2026,
author = {Samad, Muhammad Norazlan Abdul and Anuar, Arman Nuri and Mansor, Hafizah},
title = {ScoreSync: Automated Mark Entry App using Image Capture with Computer Vision and Optical Character Recognition},
journal = {International Journal of Innovative Computing},
year = {2026},
volume = {16},
number = {1-2},
pages = {287--295},
doi = {10.11113/ijic.v16n1-2.703}
}
EUVIMED