Fadlol, Muhammad Thoriq
Unknown Affiliation

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

Found 1 Documents
Search

Penerapan Deteksi Titik Api Pada Area graving dock Menggunakan YOLO dan GRAD-CAM Fadlol, Muhammad Thoriq; Khumaidi, Agus; Subiyanto, Lilik; Joko Endrasmono; Mustika Kurnia Mayangsari; Anggarjuna Puncak Pujiputra
Jurnal Elektronika dan Otomasi Industri Vol. 12 No. 1 (2025): Vol 12 No 1 (Mei 2025): Jurnal Elkolind Vol 12 No 1 (Mei 2025)
Publisher : Program Studi Teknik Elektronika Politeknik Negeri Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33795/elkolind.v12i1.7253

Abstract

Fire spot detection in the graving dock area is crucial to prevent potentially harmful fires. This study employs the YOLOs method as a deep learning-based object detection technique to detect fire and sparks in real-time. Despite its high accuracy, visual interpretation of detection results remains challenging. Therefore, the Grad-CAM technique is utilized to generate a heatmap on the detection area of YOLO. The heatmap is calculated using the alpha blending method with a specific transparency factor, resulting in clearer visualization of detected objects. The test results show that the combination of YOLO and Grad-CAM can detect fire with an accuracy of 73%. The heatmap visualization validates the critical areas that contribute to the model's decision, making it suitable for fire monitoring systems in high-risk areas.