Oil palm fruit bunches maturity prediction using standard deviation of colour
Determination of correct maturity stage of the oil palm fruit bunches is crucial in maximizing oil extraction rate (OER) of the fresh fruit bunch (FFB). A right determination of the maturity stage will lead to the right time of fruit bunch harvesting day which is only ripe fruit bunch will be harves...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2015
|
| Subjects: | |
| Online Access: | http://psasir.upm.edu.my/id/eprint/56665/1/FK%202015%2036.pdf http://psasir.upm.edu.my/id/eprint/56665/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Determination of correct maturity stage of the oil palm fruit bunches is crucial in maximizing oil extraction rate (OER) of the fresh fruit bunch (FFB). A right determination of the maturity stage will lead to the right time of fruit bunch harvesting day which is only ripe fruit bunch will be harvested. Traditional inspections of oil palm fruits to determine its degree of maturity was inaccurate and varies among different inspectors. While current practices of oil palm maturity stage determination such thru observation of number of loose fruits felled on the ground was destructive and time consuming. In this research, non-destructive and real time image capturing was applied and digital value of hue was used as a color space for image analysis. The research is conducted to monitor the color deviation of the oil palm bunch during the ripening process. The procedure in this study is started with real time image acquisition using Keyence camera which will give real time reading of HSB (Hue, Saturation and Brightness) digital number. Nine different fruit bunch from 5 years old of the oil palm trees selected for this study and the variety of oil palm is Tenera: Elaeis Guineensis. The hue digital value obtained from the Keyence camera real time reading will then analyzed and correlated with mesocarp oil content and number of loose fruits to develop a model for oil palm fruit bunch maturity stage. During image capturing, light intensity surrounding the fruit bunch was monitored using Extech Lightmeter. Regression analysis of linear model shows that the color deviation derived from hue standard deviation was significant in estimating the days to harvest the fruit bunch. The equation model obtained was y = 0.109x + 0.89 and R2 = 0.726 with y, x and R2 respectively represent hue standard deviation i.e. color deviation, estimated days to harvest and regression squared. It was physically observed that color distribution on surface of ripe fruit bunch is more uniform while color distribution for under ripe is less uniform. Delayed days to harvest affected the fruit bunch to become overripe. Overripe means the number of loose fruits and content of mesocarp’s free fatty acid (FFA) would increase. In determination of oil content, it was found that there is a different pattern rate of oil accumulation in different part of the fruit bunch. Overall relationship of the average oil content with days to harvest shows a high correlation with equation model y = 0.779x + 69.79 and R2 = 0.774 with y, x and R2 respectively represent percentage of mesocarp oil content, estimated days to harvest and regression squared. From the model developed will hopefully helpful in predicting the optimum day to harvest the oil palm fruit bunch. |
|---|