KLASIFIKASI CITRA JENIS BUAH DAN SAYURAN MENGGUNAKAN ALGORITMA YOLOV11

AZANTHA, IRVAN MALIK and Arsalan, Osvari (2025) KLASIFIKASI CITRA JENIS BUAH DAN SAYURAN MENGGUNAKAN ALGORITMA YOLOV11. Undergraduate thesis, Sriwijaya University.

	Image RAMA_55201_09021282025060_cover.jpg - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (769kB)
	Text RAMA_55201_09021282025060.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy
	Text RAMA_55201_09021282025060_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (11MB) | Request a copy
	Text RAMA_55201_09021282025060_0028068806_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (740kB)
	Text RAMA_55201_09021282025060_0028068806_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (834kB) | Request a copy
	Text RAMA_55201_09021282025060_0028068806_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (582kB) | Request a copy
	Text RAMA_55201_09021282025060_0028068806_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (284kB) | Request a copy
	Text RAMA_55201_09021282025060_0028068806_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (664kB) | Request a copy
	Text RAMA_55201_09021282025060_0028068806_06.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (225kB) | Request a copy
	Text RAMA_55201_09021282025060_0028068806_07_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (197kB) | Request a copy
	Text RAMA_55201_09021282025060_0028068806_08_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (137kB) | Request a copy

Abstract

Automatic image classification of fruits and vegetables plays a crucial role in enhancing efficiency in the agricultural and retail sectors, yet it faces challenges due to visual complexities such as intra-class variation and inter-class similarity. This research aims to implement and evaluate the effectiveness of the You Only Look Once version 11 (YOLOv11) algorithm, specifically the YOLOv11s-cls variant, for the image classification task of 34 types of fruits and vegetables. The dataset used is "Fruit and Vegetable Classification" from Kaggle, which has undergone a cleaning process, resulting in approximately 2913 images. Pre-processing methods include resizing images to 640x640 pixels and data augmentation through stretching with a 1:1 aspect ratio. The YOLOv11s-cls model, previously trained on ImageNet, was fine-tuned using a transfer learning approach. Training was conducted for 20 epochs with monitoring of loss and accuracy curves. Model performance evaluation utilized metrics such as accuracy, precision, recall, F1-score, and a confusion matrix. The research results indicate that the YOLOv11s-cls model achieved an overall accuracy of 90,8% on the test set. Analysis of the confusion matrix and per-class F1-scores identified classes with excellent performance (e.g., cauliflower, corn, cucumber with an F1-score of 1,00) as well as more challenging classes (e.g., chilli pepper with an F1-score of 0,58, potato with 0,67, apple and paprika with 0,71), generally attributed to visual similarities between classes. This study demonstrates that YOLOv11s-cls is a promising algorithm for fruit and vegetable image classification, contributing to the development of automatic identification systems in related fields.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Augmentasi Data, Buah dan Sayuran, Deep Learning, Klasifikasi Citra, Transfer Learning, YOLOv11
Subjects:	T Technology > T Technology (General) > T1-995 Technology (General)
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Irvan Malik Azantha
Date Deposited:	20 Aug 2025 07:44
Last Modified:	20 Aug 2025 07:44
URI:	http://repository.unsri.ac.id/id/eprint/183045

Actions (login required)

View Item