Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Jurnal Ilmiah Sains
Jonathan Wuntu
Faculty of Engineering, Sam Ratulangi University, , Jl. Kampus Kleak, Manado, Indonesia
Agriculture, Biological Sciences & Forestry Biochemistry, Genetics & Molecular Biology Chemistry Computer Science & IT Mathematics Physics

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

Found 1 Documents
Search

 1 
Mold on Food Product: Comparative Analysis of YOLO Variants for Detecting Rhizopus stolonifer on Bread Vanny Hani Siwi; Jonathan Wuntu; Norrytha Lineke Wuntu; Audy Denny Wuntu
Jurnal Ilmiah Sains Volume 25 Issue 2, October 2025
Publisher : Sam Ratulangi University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35799/jis.v25i2.64398

Abstract

Bread is a staple food that is highly susceptible to fungal contamination, particularly by Rhizopus stolonifer, which poses significant health and food safety risks. Early and accurate detection of mold growth is essential to prevent spoilage and ensure consumer safety. This study presents a comparative analysis of recent YOLO (You Only Look Once) variants, YOLOv8n, YOLOv10n, YOLO11n, and YOLOv12n for detecting Rhizopus stolonifer mold on bread surfaces. This study utilized a mold detection dataset sourced from the Roboflow platform, which contains annotated bread images captured under diverse lighting, texture, and contamination conditions to support robust model training. Each YOLO variant was trained and evaluated under consistent hyperparameters to ensure fairness in comparison. Experimental results indicate that YOLOv8n achieved an mAP50 of 0.472 and mAP50:95 of 0.203; YOLOv10n achieved 0.474 and 0.191, respectively; YOLO11n achieved 0.504 and 0.204; and YOLOv12n achieved 0.503 and 0.224. Among these, YOLO11n demonstrated the highest mAP50 performance, while YOLOv12n attained the best mAP50:95 score, indicating superior detection consistency across varying IoU thresholds. These findings suggest that recent YOLO architectures offer promising potential for real-time and automated detection of Rhizopus stolonifer mold in bread, supporting advancements in intelligent food safety monitoring systems.
Co-Authors Audy Denny Wuntu Norrytha Lineke Wuntu Vanny Hani Siwi