AUTOMATED OIL PALM FRIDT GRADING USING ARTIFICIAL INTELLIGENCE

This project deals with the grading of oil palm fruit based on ripeness of oil palm
fruit. The current procedure in the palm oil mills is graded manual by human
graders. The result from manual grading are very subjective and inconsistent as it
varies and depends on techniques and experience of each human graders. Hence,
it affects the quality and quantity of the oil that can be extracted. In this project, a
new model of automated grading system for oil palm fruit is developed using the
RGB color model and artificial fuzzy logic. The purpose of this grading system is
to distinguish between the three different classes of oil palm fruit which are
underripe, ripe and overripe. The ripeness or color ripening index was based on
different color intensity. The grading system uses a computer and a CCD camera
to analyze and interpret images correspondent to human eye and mind. The
computer program is developed for the image processing part like the
segmentation of colors, the calculation of the mean color intensity based on RGB
color model and the decision making process using fuzzy logic to train the data
and make the classification for the oil palm fruit. The program developed has been
able to distinguish the three different classes of oil palm fruit automatically with
86.67% of overall efficiency. This project provides a very good technique to
standardize the oil palm fruit grading system over a large area and the research
will continue to normalize the system to be able to use under different source of
lighting.