UTHM Institutional Repository

Low level CBIR for palm oil FFB prediction using neural network

Mohamed, Omar Abdi (2018) Low level CBIR for palm oil FFB prediction using neural network. Masters thesis, Universiti Tun Hussein Onn Malaysia.

[img]
Preview
PDF
Omar_Abdi_Mohamed.pdf

Download (7MB)

Abstract

This project presents the low level content based image retrieval for palm oil fresh fruit ripeness prediction using neural network. Content Based Image Retrieval (CBIR) is a system that returns the images taken with the same characteristics. To maintain the quality of oil palm produced the FFB need to be harvest according to the standard that has been set by Malaysia Palm Oil Board (MPOB). Generally, this research discovers the uniqueness of physical and optical characteristics of the oil palm Fresh Fruit Bunches (FFB). The goal is to determine the level of ripeness of the oil palm FFB by creating a neural network and real-time oil palm FFB grading systems, This grading system is a solution to automate the current grading process in order to provide more accurate and reliable results. Apart from that, the costs, labor and time consume will also be reduced. In Malaysia, the grading of oil palm FFB is still performed manually by observing the surface color as the main quality attribute. This project explores the artificial neural network in classification of oil palm and uses its characteristic as a level selector to classify the degree of oil palm FFB ripeness. This classification is chosen according to the MPOB's manual. In this case, the fruit of ripeness are divided into three categories which are unripe, ripe and over-ripe.

Item Type: Thesis (Masters)
Subjects: T Technology > TA Engineering (General). Civil engineering (General) > TA1501-1820 Applied optics. Photonics
Depositing User: Mr. Mohammad Shaifulrip Ithnin
Date Deposited: 13 Aug 2018 03:27
Last Modified: 13 Aug 2018 03:27
URI: http://eprints.uthm.edu.my/id/eprint/10266
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year