A picture of ripeness: investigating image-based techniques for oil palm fruit grading

Munirah Rosbi, . and Zaid Omar, . and Marsyita Hanafi, . (2025) A picture of ripeness: investigating image-based techniques for oil palm fruit grading. Journal of Oil Palm Research (Malaysia), 37 (1). pp. 1-15. ISSN 2811-4701

Oil palm is a highly efficient crop that can produce more oil per unit of land than any other type of oil seed. Palm oil is in high demand, and its production can significantly contribute to a country’s economic growth. However, the traditional method of grading palm fruit is still prevalent in Malaysia, which requires skilled workers to classify the harvested fruit according to its ripeness. This approach can be costly and labourintensive. Therefore, several studies have investigated automated palm fruit classification techniques that could reduce costs and labour in the industry. This article provides a review of these studies, with a specific focus on vision-based classification techniques. The article discusses approaches based on image processing encompassing pre-processing, feature extraction and classification steps. The survey’s results indicate that there is a lack of technique to effectively address outdoor images, such as colour correction methods. Therefore, further research is necessary to develop a better segmentation and colour correction procedures. Overall, the f indings of this study could help improve the efficiency and sustainability of palm oil production, thereby contributing to economic growth and environmental conservation.

Full text available from: Official URL: https://jopr.mpob.gov.my/wp-content/uploads/2024/0...

Abstract

Oil palm is a highly efficient crop that can produce more oil per unit of land than any other type of oil seed. Palm oil is in high demand, and its production can significantly contribute to a country’s economic growth. However, the traditional method of grading palm fruit is still prevalent in Malaysia, which requires skilled workers to classify the harvested fruit according to its ripeness. This approach can be costly and labourintensive. Therefore, several studies have investigated automated palm fruit classification techniques that could reduce costs and labour in the industry. This article provides a review of these studies, with a specific focus on vision-based classification techniques. The article discusses approaches based on image processing encompassing pre-processing, feature extraction and classification steps. The survey’s results indicate that there is a lack of technique to effectively address outdoor images, such as colour correction methods. Therefore, further research is necessary to develop a better segmentation and colour correction procedures. Overall, the f indings of this study could help improve the efficiency and sustainability of palm oil production, thereby contributing to economic growth and environmental conservation.

[error in script]

Item Type:	Article
AGROVOC Term:	oil palms
AGROVOC Term:	palm oils
AGROVOC Term:	fruits
AGROVOC Term:	grading
AGROVOC Term:	classification
AGROVOC Term:	image processing
AGROVOC Term:	efficiency
AGROVOC Term:	sustainability
AGROVOC Term:	economic growth
AGROVOC Term:	automation
Geographical Term:	Malaysia
Depositing User:	Nor Hasnita Abdul Samat
Date Deposited:	02 Apr 2026 07:20
Last Modified:	02 Apr 2026 07:20
URI:	http://webagris.upm.edu.my/id/eprint/25224

Actions (login required)

View Item