Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal MATRIX : JURNAL MANAJEMEN TEKNOLOGI DAN INFORMATIKA
Yansah, Muhammad Kahfi
Unknown Affiliation
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Humanoid object detection moving in open space using YOLOv8 Yansah, Muhammad Kahfi; Dijaya, Rohman; Setiawan, Hamzah; Sumarno, Sumarno
Matrix : Jurnal Manajemen Teknologi dan Informatika Vol. 15 No. 2 (2025): Matrix: Jurnal Manajemen Teknologi dan Informatika
Publisher : Unit Publikasi Ilmiah, P3M, Politeknik Negeri Bali

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31940/matrix.v15i2.60-71

Abstract

This study explores the application of the YOLOv8 algorithm in detecting humanoid objects in an open space environment, with a special focus on school areas such as parking lots. The main objective is to develop an intelligent system that can accurately identify students based on four uniform classifications: none, grey, batik, and department-specific uniforms. The system is designed to function effectively in real-time by analyzing image and video data. The research methodology begins with data acquisition using CCTV footage, followed by annotation and preprocessing using Roboflow. The dataset consists of 314 images with 1,649 labeled bounding boxes, which are then divided into training and validation sets. A yaml configuration file is created to interact with the YOLOv8 model. Training is performed using YOLOv8s variants, with experimental variations in image size, batch size, and epochs to optimize model performance. The evaluation results show that the model achieves a precision of 0.86, a recall of 0.92, and a mean Average Precision (mAP@0.50) of 0.93. Furthermore, visual testing confirms the system's ability to detect students with a total detection accuracy of 85%. Some minor errors were observed in distinguishing between visually similar classes, such as batik and department uniforms. These results demonstrate the robustness and reliability of YOLOv8 in dynamic real-world environments. This study concludes that YOLOv8 can be effectively applied to educational settings for surveillance or monitoring systems. Future research will focus on improving accuracy by expanding the dataset and incorporating more diverse categories of humanoid objects.
Co-Authors Rohman Dijaya Setiawan, Hamzah Sumarno .