Study And Development On Pattern Classification System For Empty Fruit Bunch Based On The Age Profile Of Oil Palm (Elaeis Guineeniss) Trees

Kamarudin, Ain Nurkamilia (2017) Study And Development On Pattern Classification System For Empty Fruit Bunch Based On The Age Profile Of Oil Palm (Elaeis Guineeniss) Trees. Project Report. UTeM, Melaka, Malaysia. (Submitted)

Text (24 Pages) Study And Development On Pattern Classification System For Empty Fruit Bunch Based On The Age Profile Of Oil Palm (Elaeis Guineeniss) Trees.pdf - Submitted Version Restricted to Registered users only until 18 January 2023. Download (1MB)

Abstract

A preliminary research work of collaborative project with Malaysian Palm Oil Board (MPOB) to segregate the empty fruit bunches (EFB) into similar age groups of uniform fibre quality. In general, EFB-fibre obtained from mature (8 year-old and above) oil palm trees are stronger than immature (7 year-old and below) trees. At present, the EFB are manually segregated without an aid of machine vision, and thereby, the process is prone to human error. The purpose of this study is to develop a pattern classification system for EFB, which was obtained from different age profile of oil palm trees. This project represent the development of empty oil palm fruit bunches (EFB) segregation prototype using MATLAB for different age group. Most of the manufactured industry does not utilize the production by using the EFB according to suitable age. This problem occurs because the production do not has the good system to be implemented to segregate the EFB according to their age specification such as to improve the parameters for the machine to be operated. This is very crucial in the industry because different age of EFB for making commercializes purposes and generate additional income to the palm oil company. The approach of this control system starts with building the high percentage of identification for each bunches belongs to which age group. Neural network is use using the MATLAB software to train the parameter given and it automatically adjust to the network's weights and biases. The system starts with after the drying process of the bunches after the milling process, the camera integrated with MATLAB based of Neural Network as the threshold data to identify each bunches group and Arduino act as a segregation component will determine the sample of EFB to segregate according to their age specification. Besides, the EFB will be sorting and identify using LED to recognize their age group, segregate to their classes and ship to the suitable industry.

Actions (login required)

View Item