CD Skripsi

Rangkaian Otomasi Deteksi Cacat Pada Brondolan Buah Kelapa Sawit Menggunakan Algoritma You Only Look Once

---

Automation is a key element in the concept of Industry 4.0, aims to enhance efficiency and flexibility in production processes. This study built an automated circuit for defect detection of oil palm fruitlets using You Only Look Once (YOLO) algorithm. YOLO was chosen for its ability to detect objects in real-time with high accuracy. This study involves several stages, including the preparation of tools and materials, data acquisition and annotation to build a dataset, building a YOLO detection model, model training and testing, and design an arduino based automation circuit. The dataset consists of two class of oil palm fruitlets, ripe and rotten. The model used YOLOv8-n algorithm, which was trained and tested to detect these two classes using Google Colab. Model performance analysis used was conducted using a confusion matrix with parameters including precision, recall, and F1-score. The analysis results obtained precision of 96.4%, recall of 95.8%, and F1-score of 96.1% for ripe class, and precision of 98.8%, recall of 92.0%, and F1-score of 95.3% for the rotten class. The overall detection system model had the highest accuracy of 94.4% with a mean Average Precision (mAP) of 95.9%. The automation circuit used computer vision integrated with an an arduino board with servo motor to separate the fruitlets based on the detection results. The automation circuit had a high accuracy of 95.4% with an error accuracy of 4.5%. The results show that circuit and the model potential to be used to sorting oil palm fruitlets based on two classes, ripe and rotten.



Keywords : Defect Detection, Oil Palm Fruitlets, Computer Vision, YOLO Algorithm,Automation Circuit, Analysis.

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

2003110320

Penerbit

Pekanbaru : Universitas Riau FMIPA Fisika., 2024

Deskripsi Fisik

-

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

2003110320

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subjek

FISIKA

Info Detail Spesifik

-

Pernyataan Tanggungjawab

mardiah

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar