Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal International Journal of Electronics and Communications Systems
Riswansyah , Muh Fikra Junian
Unknown Affiliation
Computer Science & IT Electrical & Electronics Engineering

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

Found 1 Documents
Search

 1 
Lightweight Image-Based Mold Detection System for Real-Time Bread Quality Monitoring Using Artificial Neural Networks (ANN) Saputra, Nikola; Ilyas, Muh.; Riswansyah , Muh Fikra Junian; Kaswar, Andi Baso; Lamada, Mustari S.
International Journal of Electronics and Communications Systems Vol. 5 No. 2 (2025): International Journal of Electronics and Communications System
Publisher : Universitas Islam Negeri Raden Intan Lampung, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24042/ijecs.v5i2.28706

Abstract

Mold contamination of white bread is an ongoing challenge for quality monitoring, while conventional visual inspection remains unreliable for early and consistent detection. This study aims to propose a lightweight image-based mold detection system for white bread oriented towards real-time quality monitoring using Artificial Neural Networks (ANNs). An experimental workflow combining digital image acquisition, pre-processing, Otsu-based segmentation, morphological refinement, multicolor color space feature extraction, and an Artificial Neural Network (ANN) classifier is implemented. Results indicate that color information is the dominant discriminatory cue for mold identification, while texture descriptors provide complementary structural information that improves class separation. The RGB+HSV+LAB combination achieved the highest performance, with a training accuracy of 97.91 percent and a testing accuracy of 96.66 percent. These findings demonstrate that effective mold classification can be achieved without relying on deep or computationally intensive architectures when the feature representation is well-designed. In conclusion, a lightweight, feature-centric ANN (Artificial Neural Network) is sufficient for reliable classification of mold growth levels on white bread. This study confirms that a compact, feature-based learning strategy is sufficient for reliable classification of mold on white bread, providing a technically efficient basis for a vision-based food quality assessment system.
Co-Authors Andi Baso Kaswar Andi Baso Kaswar Ilyas, Muh. Saputra, Nikola