Identification and Classification of Oil Palm Maturity Using Machine Learning Techniques

Murinto, Murinto and Rosyda, Miftahurrahma (2022) Identification and Classification of Oil Palm Maturity Using Machine Learning Techniques. [Artikel Dosen]

[thumbnail of HASIL CEK_Murinto_Identification and Classification of Oil Palm Maturity Using Machine Learning Techniques.pdf] Text
HASIL CEK_Murinto_Identification and Classification of Oil Palm Maturity Using Machine Learning Techniques.pdf

Download (2MB)

Abstract

Oil palm is the main plantation crop in Indonesia, oil palm is the most efficient producer of vegetable oil. Oil palm fruit is one of the fruits that has a certain level of maturity in a relatively fast time. The distribution of oil palm fruit in various regions makes it important to identify and classify the maturity of oil palm fruit based on its maturity level. The degree of ripeness of the bunches at harvest is closely related to the oil content contained in the fruit. Accuracy problems are often encountered in research related to image classification. One challenge that arises is finding an appropriate representation of the data so that important structures of the data can be seen easily. One of the processes carried out to get better accuracy is the segmentation process. Through the use of proper segmentation techniques, the desired accuracy will be obtained. One of the techniques used in the segmentation method is to use the swarm optimization technique and its derivatives. In this study, identification and classification will be implemented using particle swarm optimization (PSO) at thresholding image segmentation in order to obtain better segmentation results when compared to the previous method. The classification is based on existing machine learning techniques, namely support vector machine (SVM). the accuracy rate for the classification of palm fruit maturity based on texture using the Support Vector Machine (SVM) method is obtained, which reaches 92.5%. From the accuracy obtained, it can be concluded that the method used to identify and classify in this study is good.

Item Type: Artikel Dosen
Keyword: Classification, Particle Swarm Optimization, Support Vector Machine
Subjects: Q Science > Q Science (General)
Divisi / Prodi: Faculty of Industrial Technology (Fakultas Teknologi Industri) > S1-Informatics Engineering (S1-Teknik Informatika)
Depositing User: murinto murinto
Date Deposited: 18 Apr 2023 09:48
Last Modified: 23 Sep 2023 05:56
URI: http://eprints.uad.ac.id/id/eprint/43010

Actions (login required)

View Item View Item