Student Uniform Detection with Face Recognition

Abstract

This project presents an advanced system for detecting student uniforms in educational environments, leveraging the YOLOv8 (You Only Look Once, version 8) object detection algorithm. By harnessing YOLOv8�s real-time capabilities, we integrate facial recognition using Haar Cascade and Local Binary Pattern Histograms (LBPH) to ensure robust identification and dress code compliance. The system is trained on a meticulously curated dataset, focusing on uniform components like shirts, ties, and pants, as well as individual facial features. This dual detection approach not only enforces dress codes but also enhances campus security by verifying student identities, thus ensuring that only authorized individuals access school premises. Furthermore, the system simplifies attendance monitoring by automating uniform checks, reducing administrative workload, and improving accuracy. Challenges such as optimizing model performance under varied lighting conditions and improving detection accuracy have been addressed through extensive training and data augmentation. The report details our methodology, system architecture, challenges faced, and the effectiveness of this integrated solution, offering a practical tool tailored to the needs of modern educational institutions in computer vision applications.