Model Image Recognition untuk proses sortir buah manggis dengan YOLOv8

Nugraha, Rizky Rahmat (2025) Model Image Recognition untuk proses sortir buah manggis dengan YOLOv8. Sarjana thesis, UIN Sunan Gunung Djati,Bandung.

Preview	Text (COVER) 1_cover.pdf Download (55kB) | Preview
Preview	Text (ABSTRAK) 2_abstrak.pdf Download (23kB) | Preview
Preview	Text (DAFTAR ISI) 3_daftarisi.pdf Download (448kB) | Preview
Preview	Text (BAB I) 4_bab1.pdf Download (109kB) | Preview
	Text (BAB II) 5_bab2.pdf Restricted to Registered users only Download (824kB) | Request a copy
	Text (BAB III) 6_bab3.pdf Restricted to Registered users only Download (324kB) | Request a copy
	Text (BAB IV) 7_bab4.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text (BAB V) 8_bab5.pdf Restricted to Registered users only Download (36kB) | Request a copy
	Text (DAFTAR PUSTAKA) 9_daftarpustaka.pdf Restricted to Registered users only Download (163kB) | Request a copy
	Text (LAMPIRAN) 10_lampiran.pdf Restricted to Registered users only Download (148kB) | Request a copy

Abstract

This study aims to develop an artificial intelligence model based on image recognition using the YOLOv8 algorithm to enhance the efficiency and accuracy of the mangosteen sorting process, one of Indonesia's key export commodities. Despite its significant economic potential, only about 25% of Indonesian mangosteens meet export standards, primarily due to visual defects such as yellow latex stains and speckled blemishes on the fruit's skin. The manual sorting process commonly performed by farmers and exporters is often time-consuming and inconsistent. Therefore, this research adopts a deep learning approach to automatically identify the quality of mangosteens based on their visual characteristics. The YOLOv8 algorithm was selected for its advantages in providing high accuracy and efficient object detection. This study evaluates the model's performance in detecting visual attributes of mangosteens, such as skin surface, color, and stem condition, which are critical indicators of export quality. The model was tested with various data configurations, achieving its best accuracy rate of 73%. These results indicate that the model can deliver significant performance in classifying fruit quality, supporting more precise decision-making in the sorting process. In conclusion, the use of YOLOv8 for mangosteen quality detection can improve sorting efficiency and accuracy while offering tangible benefits to Indonesia's agricultural industry in meeting stringent export standards. This research is expected to pave the way for implementing similar technology in other commodities.

Actions (login required)

View Item