CD Skripsi

Deteksi Kematangan Tandan Buah Segar Kelapa Sawit Secara Real-Time Berbasis Video Dan Algoritma Yolo

---

Determining the ripeness level of oil palm Fresh Fruit Bunches (FFBs) using computer vision and machine learning has been extensively studied, especially using the YOLO (You Only Look Once) algorithm. Its applications are typically employed for detecting FFBs on trees or piled on the ground. However, the detection of stacked FFBs moving on a conveyor has yet to be done. The objective of this study is to use a computer vision system for detecting the ripeness level of stacked FFBs using two YOLO algorithms to build a detection model, train and test the model, validate the model using video, and measure the model's performance. The determination of the average RGB intensity value for the two FFB ripeness levels was also conducted. An RGB camera was used to acquire image data for oil palm FFBs. The model was built using the YOLOv8l-Seg and YOLOv8x-Seg algorithms, which can detect FFB in real-time. The samples consist of 30 FFBs, divided into ripe and unripe categories. The results showed that the computer vision system is capable of detecting the ripeness of stacked FFBs. Both algorithms can detect the ripeness levels with higher accuracy and speeds; however, YOLOv8x-Seg obtained a higher accuracy, and YOLOv8l-Seg resulted in a higher speed. The results also showed that the average RGB intensity for ripe FFBs is higher than that of unripe FFBs.

Keywords: Oil palm FFB, ripeness detection, computer vision, deep learning, YOLOv8, RGB intensity

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

2003112800

Penerbit

Pekanbaru : Universitas Riau FMIPA Fisika., 2024

Deskripsi Fisik

-

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

2003112800

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subjek

FISIKA

Info Detail Spesifik

-

Pernyataan Tanggungjawab

Mutia

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar