PENGEMBANGAN SISTEM SORTIR OTOMATIS BUAH MENGGUNAKAN MODEL YOLOV8

CADINE, VALENCYA FEBRIYANTY and Yani, Irsyadi and Prakoso, Akbar Teguh (2024) PENGEMBANGAN SISTEM SORTIR OTOMATIS BUAH MENGGUNAKAN MODEL YOLOV8. Undergraduate thesis, Sriwijaya University.

	Text RAMA_21201_03051282126089.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (5MB) | Request a copy
	Text RAMA_21201_03051282126089_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
	Text RAMA_21201_03051282126089_01_front_ref.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_21201_03051282126089_0025127104_0012049213_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (156kB) | Request a copy
	Text RAMA_21201_03051282126089_0025127104_0012049213_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (141kB) | Request a copy
	Text RAMA_21201_03051282126089_0025127104_0012049213_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (606kB) | Request a copy
	Text RAMA_21201_03051282126089_0025127104_0012049213_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (73kB) | Request a copy
	Text RAMA_21201_03051282126089_0025127104_0012049213_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (92kB) | Request a copy
	Text RAMA_21201_03051282126089_0025127104_0012049213_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (3MB) | Request a copy

Abstract

This research aims to develop an automated sorting system for "achacha" fruits using a combination of Basler cameras, the YOLOv8 model for object detection, and the Epson T3 robot arm integrated through LabVIEW. The system is designed to classify fruits into "good" and "bad" categories based on visual characteristics, while ensuring the picking and transferring processes are automatic and precise. This system is expected to improve the efficiency and accuracy of small-scale fruit sorting.The research stages include image capture using a Basler camera operated via Pylon Viewer, dataset annotation on Roboflow, and training of the YOLOv8 model through Anaconda Prompt and Spyder. The resulting dataset is processed to train the YOLOv8 model, which outputs coordinates and fruit classifications. This data is then used by the robot arm to perform sorting based on the categories.The test results show that the system successfully achieved an accuracy of 88.1%, precision of 91.8%, recall of 91.8%, and F1-score of 91.8%. The Precision-Confidence Curve, Recall-Confidence Curve, and F1-Score Curve indicate reliable model performance, with an optimal confidence threshold of 0.217. Additionally, LabVIEW integration ensures smooth communication between the camera, computer, and robot, allowing real-time detection of fruit locations and precise placement to the appropriate locations.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Buah Achacha, YOLOV8, Kamera Basler, LabVIEW, Robot Epson T3, Roboflow, Sistem Sortir Otomatis
Subjects:	T Technology > TJ Mechanical engineering and machinery > TJ210.2-211.47 Mechanical devices and figures. Automata. Ingenious mechanisms. Robots (General)
Divisions:	03-Faculty of Engineering > 21201-Mechanical Engineering (S1)
Depositing User:	Valencya Febriyanty Cadine
Date Deposited:	18 Jan 2025 05:07
Last Modified:	18 Jan 2025 05:07
URI:	http://repository.unsri.ac.id/id/eprint/164805

Actions (login required)

View Item