Subahan, Muhammad Nur
(2021)
Pengolahan Citra Klasifikasi Tingkat Kematangan Tandan Buah Segar (TBS) Kelapa Sawit.
Other thesis, Universitas Islam Riau.
Abstract
Indonesia is the country that owns the largest oil palm plantation area in the world in 2010-2014, covering an area of 110 million hectares and is the world's largest producer of Crude Palm Oil (CPO) with a productivity level of up to 48.44%. However, the production level of Indonesian palm fresh fruit bunches (FFB) is only at a position with a production of 16.99 ton / ha. The low level of FFB productivity is influenced by two main factors, namely the success of the harvest in the oil palm plantation area and the sorting of FFB produced by the Palm Oil Mill (POM) is still classified as conventional so that it affects the speed of FFB production. There is a need for a system that makes the FFB sorting process fast and accurate. This study uses digital image processing by applying the Radial Basis Function (RBF) artificial neural network algorithm and taking color images of Hue, Saturation and Value (HSV) as a method for classifying between raw and ripe palm FFB. The amount of data used in this study amounted to 21 image testing data of oil palm fresh fruit bunches (FFB). Based on the results of testing the accuracy of the system has a relatively high level of accuracy, which is 95,2%, indicating that the classification system for the classification of the maturity level of oil palm fresh fruit bunches (FFB) can be applied.
Actions (login required)
 |
View Item |
