RANCANG BANGUN SISTEM IDENTIFIKASI HAMA TANAMAN CABAI MENGGUNAKAN RASPBERRY PI DAN YOLOv8

Alif, Fadel Muhammad (2026) RANCANG BANGUN SISTEM IDENTIFIKASI HAMA TANAMAN CABAI MENGGUNAKAN RASPBERRY PI DAN YOLOv8. Bachelor thesis, Institut Teknologi Kalimantan.

	Text 04211024_Abstract_en.pdf Restricted to Repository staff only until 10 April 2028. Download (341kB) | Request a copy
	Text 04211024_Abstract_id.pdf Restricted to Repository staff only until 10 April 2028. Download (356kB) | Request a copy
	Text 04211024_Bibliography.pdf Restricted to Repository staff only until 10 April 2028. Download (339kB) | Request a copy
	Text 04211024_Chapter_1.pdf Restricted to Repository staff only until 10 April 2028. Download (508kB) | Request a copy
	Text 04211024_Chapter_2.pdf Restricted to Repository staff only until 10 April 2028. Download (1MB) | Request a copy
	Text 04211024_Chapter_3.pdf Restricted to Repository staff only until 10 April 2028. Download (1MB) | Request a copy
	Text 04211024_Chapter_4.pdf Restricted to Repository staff only until 10 April 2028. Download (1MB) | Request a copy
	Text 04211024_Conclusions.pdf Restricted to Repository staff only until 10 April 2028. Download (372kB) | Request a copy
	Text 04211024_Enclosure.pdf Restricted to Repository staff only until 10 April 2028. Download (6MB) | Request a copy
	Text 04211024_Illustrations.pdf Restricted to Repository staff only until 10 April 2028. Download (324kB) | Request a copy
	Text 04211024_Notation.pdf Restricted to Repository staff only until 10 April 2028. Download (301kB) | Request a copy
	Text 04211024_Preface.pdf Restricted to Repository staff only until 10 April 2028. Download (341kB) | Request a copy
	Text 04211024_Table_of_Contents.pdf Restricted to Repository staff only until 10 April 2028. Download (359kB) | Request a copy
	Text 04211024_Tables.pdf Restricted to Repository staff only until 10 May 2028. Download (296kB) | Request a copy
	Text 04211072_Cover.pdf Restricted to Repository staff only until 10 April 2028. Download (584kB) | Request a copy
	Text 04211024_Cover.pdf Restricted to Repository staff only until 10 April 2028. Download (584kB) | Request a copy
	Text 04211024_Approval_Sheet.pdf Restricted to Repository staff only until 10 April 2028. Download (125kB) | Request a copy
	Text 04211024_Statement_of_Autthenticity.pdf Restricted to Repository staff only until 10 April 2028. Download (87kB) | Request a copy
	Text 04211024_Publishing_Agreement.pdf Restricted to Repository staff only until 10 April 2028. Download (96kB) | Request a copy
	Text 04211024_Preface.pdf Restricted to Repository staff only until 10 April 2028. Download (98kB) | Request a copy
	Text 04211024_Presentation.pdf Restricted to Repository staff only until 10 April 2028. Download (1MB) | Request a copy

Abstract

Pest infestation on chili plants, specifically Thrips, Silverleaf Whitefly (Bemisia tabaci), Armyworm (Spodoptera litura), and Onion Caterpillar (Spodoptera exigua), is the primary cause of decline in harvest quality and quantity. Current manual monitoring methods are considered ineffective and time-consuming. This study aims to design and build a pest detection system for chili plants targeting Thrips, Silverleaf Whitefly, Armyworm, and Onion Caterpillar using Raspberry Pi and the YOLOv8 algorithm integrated with a monitoring website. The research methodology includes model training, prototype and system design, and system performance testing based on variations in image capture distance (3 cm, 6 cm, 9 cm, 12 cm, and 15 cm) in both simulation and field conditions. The test results indicate that system performance is significantly influenced by detection distance and environmental conditions. In simulation testing, the system achieved optimal performance at close range (3 cm – 6 cm), with the highest accuracy reaching 88% for the Silverleaf Whitefly. However, in field validation testing, there was a significant decrease in accuracy to 5-6%, caused by a high rate of False Positives due to background noise and the microscopic physical size of the pests. The system proved ineffective in detecting objects at distances greater than 9 cm. Nevertheless, the system functionally succeeded in integrating pest detection with a responsive website accessible on mobile devices. The conclusion of this study suggests that the implementation of YOLOv8 on Raspberry Pi is effective as a close-range monitoring tool but requires further development in optical aspects and environmental datasets to improve detection robustness in open fields. Keywords: Raspberry Pi, YOLOv8, Pest Detection, Website

Item Type:	Thesis (Bachelor)
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions:	Jurusan Teknologi Industri dan Proses > Teknik Elektro
Depositing User:	Fadel Muhammad Alif
Date Deposited:	09 Jul 2026 07:12
Last Modified:	09 Jul 2026 07:12
URI:	http://repository.itk.ac.id/id/eprint/25327

Actions (login required)

View Item