Oil palm fruit ripeness classification using optical-based colour sensor

Harvesting oil palm Fresh Fruit Bunches (FFB) at right stage of ripeness is critical to ensure optimum quality and quantity of oil production. Colour of these fruits can be used as an indicator to predict the level of fruits maturity. The manual grading system based on the colour of the fruits, w...

Full description

Saved in:
Bibliographic Details
Main Author: Saadat, Kimia
Format: Thesis
Language:English
Published: 2016
Online Access:http://psasir.upm.edu.my/id/eprint/67099/1/FK%202016%20125%20IR.pdf
http://psasir.upm.edu.my/id/eprint/67099/
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Harvesting oil palm Fresh Fruit Bunches (FFB) at right stage of ripeness is critical to ensure optimum quality and quantity of oil production. Colour of these fruits can be used as an indicator to predict the level of fruits maturity. The manual grading system based on the colour of the fruits, which is currently used in oil palm mills may cause conflict and misconduct in fruits classification. Therefore, developing an automated fruit classification system is essential. A number of automated fruit sorting methods have been offered and tested for practical usage during the past few years but most of them are costly, bulky and time consuming. In this study, a reflective colour sensor is used to determine the ripeness of the palm fruit. A portable optical-based sensor system to classify oil palm fruits based on fruits ripening stages has been designed. Several coloured papers were used to verify the effectiveness of the measurement. Analyses setup on distance, angle and position of the transmitter and receiver of the sensor were carried out. The best wavelength of the light source was chosen for the fruit classification system. All the measurements carried out are nondestructive and the oil palm fruit is possible to be measured on-site. Results from the measurements set-up analyses show that 180º is the best angle between transmitter and receiver. The gap between transmitter and receiver and the gap between fruit and receiver are 1cm and 0.5cm-0.75cm respectively for high accuracy detection. The palm fruit is positioned vertically to the sensor for best detection. 1.5cm-2.3cm of fruit diameter range up to a week old fruit plugged ` the bunch in a dark room (LUX=0) is the best condition for oil palm fruit ripeness classification for this system. The system is able to categorise the colour of different ripeness of oil palm fruit based on its colour reflection. The voltage ranges for unripe, ripe and overripe fruits are 0.478V to 0.511V, 0.674V to 0.714V and 1.22V to 1.24V respectively. The system also indicates the category of the fruit ripeness by displaying green, orange and red light for ripe unripe and overripe fruits respectively.