Bagus Setyawan, Aviv Yuniar Rahman, Istiadi
Tempe in Indonesia is a traditional food produced from the fermentation process of soybeans. Tempeh itself is very good for meeting very high protein needs. In Indonesia, tempeh is very common in various regions. Tempe has become a staple food in Indonesia. Ingredients include increased protein, folate, vitamin B12, and many others. Tempe that will be consumed will be chosen directly by consumers without thinking about the image of freshness in tempeh. This makes some limitations on inconsistent results on the image of tempeh taste. This study aims to detect the quality of tempeh maturity using YOLO Tiny-v4. Another objective of this research is to provide a tempeh image maturity detection system that can later be used to speed up the tempeh selection process. Not only that, this tempeh maturity quality detection system can be used in real-time in its application. It can be concluded that the detection results using YOLO Tiny-v4 on the detection of raw, ripe, and rotten tempeh have a maximum mean average precision (mAP) of 1.00. This proves from previous research that the YOLO method can be applied to detect the quality of tempeh maturity, and in this case, the YOLO Tiny-v4 method can improve the results of previous studies. © 2023 IEEE.
Universitas Widyagama Malang, Department of Informatic Engineering, Malang, Indonesia
Research at a Glance
Register to unlock
Topics & SDG Alignment
Register to unlock
Collaboration
Register to unlock
Author Profile (Selected)
Register to unlock
References Overview
Register to unlock
Journal & Source
Register to unlock
Metadata & Integrity
Register to unlock