CD Skripsi

Pemetaan Potensi Air Bawah Tanah Dengan Menggunakan Metode Geolistrik Aturan Wenner Dan Geokimia Di Kelurahan Rintis, Kec. Lima Puluh

---

Separation of oil palm fruit bunches after the threshing process is important to reduce oil loss that remains in the bunches. This study aims to develop an automation system based on computer vision and an Arduino Uno microcontroller to separate empty bunches and partially stripped bunches in real-time. The method used is the YOLOv8 object detection algorithm, trained using a dataset of partially stripped bunch images captured at PTPN V Sei Pagar, Kampar Regency, with an RGB camera. The test samples consisted of two object classes: empty bunches and partially stripped bunches. Annotation was carried out using Roboflow software, and the model was trained and tested with an 80:20 data split. The training results showed that the YOLOv8s model achieved the best performance, with an mAP of 60%, precision of 73%, recall of 58%, and F1-score of 65%. The automation control system was tested directly by connecting the detection results to an Arduino microcontroller that controlled a pneumatic actuator. The test results showed a total separation accuracy of 90% with an average actuator response time of 1 seconds. This system is potential to be used in the palm oil industry.
Keywords: Computer vision, YOLOv8, Arduino Uno, Automation, Oil palm fruit bunch, Oil loss

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

2103110733

Penerbit

Pekanbaru : Universitas Riau FMIPA Fisika., 2025

Deskripsi Fisik

-

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

2103110733

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