CLASSIFICATION OF WHOLE AND BROKEN RICE USING DEEP LEARNING BASED DIGITAL IMAGE PROCESSING TECHNOLOGY USING YOLOV5 METHOD

Muni Chan, I Gusti Ngurah (2023) CLASSIFICATION OF WHOLE AND BROKEN RICE USING DEEP LEARNING BASED DIGITAL IMAGE PROCESSING TECHNOLOGY USING YOLOV5 METHOD. Tugas Akhir thesis, University of Technology.

[img] Text
5160711127_I Gusti Ngurah.pdf
Download (144kB)

Abstract

ABSTRACT Indonesia is a tropical country that has a large potential for agricultural commodities. The rice commodity is the most important commodity in Indonesia because of its role as a staple food where the majority of the Indonesian population consumes rice every day as carbohydrate intake. The problem faced in this study is how to detect the physical characteristics of whole rice and broken rice. A system is needed that can distinguish the physical characteristics of rice by detecting it accurately. The method used in this study is the YOLOv5 algorithm. This study used a total of 160 training image datasets and 40 validation image datasets by labeling before training. This study succeeded in detecting intact rice images as well as broken rice images with a total loss of 0.0045, using 100x epochs with an accuracy of 92.25%. Keywords: Whole Rice, Broken Rice, Deep Learning, YOLOv5, Citra

Item Type: Thesis (Skripsi, Tugas Akhir or Kerja Praktek) (Tugas Akhir)
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions: Fakultas Sains Dan Teknologi > S1 Teknik Elektro
Depositing User: Kaprodi Teknik Elektro
Date Deposited: 22 Jul 2023 02:19
Last Modified: 22 Jul 2023 02:19
URI: http://eprints.uty.ac.id/id/eprint/13109

Actions (login required)

View Item View Item
UTY Open Access is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.