Publication: Developing a facial recognition system for oku smart parking
| datacite.subject.fos | oecd::Engineering and technology::Electrical engineering, Electronic engineering, Information engineering::Electrical and electronic engineering | |
| dc.contributor.author | Nur Masyitah binti Mohd Idris | |
| dc.date.accessioned | 2025-05-28T09:14:39Z | |
| dc.date.available | 2025-05-28T09:14:39Z | |
| dc.date.issued | 2024-07 | |
| dc.description.abstract | In Malaysia, individuals identified as "Orang Kurang Upaya" (OKU) encounter significant challenges accessing parking facilities, exacerbated by issues with existing RFID card systems prone to misuse. This project aims to improve accessibility for OKU individuals with physical disabilities or blindness by implementing a real-time face recognition system. Utilizing the Multi-task Cascaded Convolutional Network (MTCNN) model integrated with OpenCV, the system verifies users' identities from video frames or digital images. Deep learning techniques embedded in the MTCNN enhance accuracy by processing multiple layers of data, overcoming challenges such as pose variations, occlusions, and changing lighting conditions typical in real-world environments. Methodologically, rigorous testing evaluated the system's performance across diverse conditions, revealing robust operation in both controlled settings and unpredictable scenarios. numerical results confirm the system's high accuracy in face detection and recognition, essential for ensuring secure and convenient access to designated OKU parking spaces without requiring physical contact or RFID cards. The system's ability to learn and adapt from extensive image datasets contributes to its reliability in identifying authorized users swiftly and effectively. This research highlights the transformative potential of facial recognition technology in ddressing accessibility and security concerns in OKU smart parking systems. By providing a non-intrusive yet reliable means of identity verification, the proposed system aims to mitigate access challenges faced by OKU individuals, promoting inclusivity and enhancing overall user experience in parking facilities. | |
| dc.identifier.uri | https://erepo.usm.my/handle/123456789/21979 | |
| dc.language.iso | en | |
| dc.title | Developing a facial recognition system for oku smart parking | |
| dc.type | Resource Types::text::report::technical report | |
| dspace.entity.type | Publication | |
| oairecerif.author.affiliation | Universiti Sains Malaysia |