The oil palm, a perennial oil yielding crop, is the richest source of vegetable oils which can produce 4 to 6 MT of palm oil (mesocarp oil) and 0.4 to 0.6 t of palm kernel oil per hectare per annum. The extent of oil available/extractable depends on the ripeness stage of the fruits. The present study was undertaken to evaluate different maturity stages of oil palm fruits in terms of color and oil content, establish their inter relationship and to develop prediction models based on color values so that non destructive ripeness evaluation could be achieved. Models developed with Red, Green and Blue (RGB) values showed that the oil content on fresh fruit bunch (FFB) could be predicted with 57 to 66% efficiency. Models developed using L*a*b* values could predict oil content in fresh fruit bunches up to 79% accuracy. Validations of the models were done with different data sets. The RGB based model showed 64% accuracy in prediction. The L*a*b* model upon validation could predict oil percent up to 89% accuracy. The L*a*b* based model would be ideal for incorporating in gadgets like colorimeters for the purpose of color based grading of FFB and prediction of oil content. Further, it will be useful in automation of harvesting through a machine vision system. This will finally help in harvesting at correct stage of ripeness and objective grading as well as price fixation of oil palm FFB.
Key words: Oil palm, fresh fruit bunch, color grading, L*a*b* color space, mesocarp oil.
Copyright © 2021 Author(s) retain the copyright of this article.
This article is published under the terms of the Creative Commons Attribution License 4.0