Pre-harvest oil palm FFB nondestructive evaluation technique using thermal-imaging device

Abstract

Oil Palm is widely cultivated in Indonesia due to its superior economical value. Currently, oil palm plantations have many challanges, particularly the time of harvesting. Many losses occurred during this process and therefore, they need best harvesting practice to obtain good results. While the quality of oil palm fresh fruits bunch (FFB) (i.e. ripeness) can be determined after harvest, a pre-harvest assessment technique is not present at the moment. In this study, we observe FFBs at four different fruit ages, e.g. 120, 140, 160 and 180 days after anthesis (DAA). A non-destructive evaluation was performed by measuring the FFBs surface temperature using a thermal camera. In addition, the influence of ambient temperature and other environment parameters were observed. Afterwards, the FFBs harvested, and its chemical compositions measured. A model developed to correlate the FFB surface temperature with its ripeness. Results showed that, regression model of the ratio of FFB oil content and moisture content with surface temperature has coefficient of determination of 0.6732. Moreover, when FFB surface temperature was correlated with its carotene content using multiple–regression, the model R2 was obtained at 0.8122. In addition, FFB surface temperature can be used to predicted the FFB age with R2 of 0.8344. This suggests a strong correlation between FFB surface temperature and its age. Furthermore, the ratio of FFB oil content and moisture content with its ripeness, and surface temperature, open the opportunity for a more accurate pre-harvest nondestructive evaluation technique to determine optimum harvest window (OHW) prediction of the bunch. The result delivers a solution of better harvesting practice in oil palm industry.

A Review of an Artificial Intelligence Framework for Identifying the Most Effective Palm Oil Prediction

“…The ideal phenotype subfeatures based on various regions should also be investigated. Because of the wide range of existing factors, such as diseases, tree detection, and physical characteristics of palm oil yield, predictions in similar fields should be made over several years [45,105]. Palm oil yield predictions in similar fields should be recorded over several years due to variations in existing factors such as diseases, tree detection, and physical characteristics [45,105].…”

“…Because of the wide range of existing factors, such as diseases, tree detection, and physical characteristics of palm oil yield, predictions in similar fields should be made over several years [45,105]. Palm oil yield predictions in similar fields should be recorded over several years due to variations in existing factors such as diseases, tree detection, and physical characteristics [45,105]. Palm oil forecasting does not emphasise phenotype data similarly to genotype representation; therefore, redundant features must be eliminated to ensure the predictive model accuracy.…”

A Review of AI Techniques in Fruit Detection and Classification: Analyzing Data, Features and AI Models Used in Agricultural Industry

“…This conventional technique presents some problems, namely 1. time-related issues; a slow and tiresome process (Thakur et al, 2020) 2. The general unreliability of the technique includes human errors and subjectivity in the technique (Makky and Cherie, 2021) 3. Labor intensiveness (Fadilah et al, 2012) and 4.…”