DETEKSI OBJEK PENGENALAN TINGKAT KEMATANGAN TANDAN BUAH SEGAR KELAPA SAWIT MENGGUNAKAN CENTERNET BERBASIS ANDROID

No Thumbnail Available

Date

2023-06

Journal Title

Journal ISSN

Volume Title

Publisher

Elfitra

Abstract

The maturity level of oil palm Fresh Fruit Bunches (FFB) is a determining factor for the quality of Crude Palm Oil (CPO). The sorting method after harvest or before entering the boiling process is generally done manually relying on vision and experience. This method is prone to subjective errors. Imaging methods are developing rapidly due to advances in computers and image processing techniques, particularly for sorting and grading systems. CenterNet is an architecture for object detection that is significantly enhanced with the help of deep learning, CenterNet is a method for exploring the center of a geometric grid. Plus a keypoint. This study used 18,000 FFB image data, the model can detect the maturity level of oil palm from five maturity levels, namely ripe 1 in the black-red color category, ripe 2 in the red color category, ripe 3 in the red-yellow color category, in the yellow-orange category, and tanko with gray color category. The performance of the results of the model has an accuracy rate of 88%.

Description

Keywords

CenterNet, FFB, Object Detection, Palm Oil

Citation

Perpustakaan

Collections