9th International Congress on Biotechnology and Food Sciences
  • Follow

Accepted Abstracts

Digital Image Processing for Predicting the Maturity Stage of Oil Palm (Elaeis guineensis Jacq.) Fresh Fruit Bunch

Mohd. Hudzari Bin Haji Razali*
Universiti Teknologi MARA, Malaysia

Citation: Razali MHBH (2020) Digital Image Processing for Predicting the Maturity Stage of Oil Palm (Elaeis guineensis Jacq.) Fresh Fruit Bunch. SciTech BioTech-Food Sciences 2020. Thailand

Received: January 16, 2020         Accepted: January 20, 2020         Published: January 20, 2020


Harvesting of the oil palm fresh fruit bunches (FFB) at correct stage of ripening is important in maximizing crop yield. The quality of oil produced depends to a great extent on the correct date of harvesting. To satisfy the concept of machine intelligence for the smart technology, non-destructive and real time simulation method are necessary to predict the FFB maturity stage. Digital value of image known as Hue was used to measure the oil palm fruit colour skin in actual plantation condition. Using Analysis of Variance (ANOVA) method, the Hue was found the best digital value for determine the color surface of FFB among another digital value component of Red, Green and Blue. The relationship of the oil content for Mesocarp oil palm fruits with the digital value of Hue was analysed. The procedure starts from image capturing of the FFB during unripe (black color surface until overripe stage (orange color surface). The ages of oil palm trees chosen in this experiment were of 5, 16 and 20 years old located at Malaysian Palm Oil Board (MPOB) and Universiti Kebangsaan Malaysia (UKM) Research Station, Bangi Lama, Malaysia. The variety of oil palm is Tenera: Elaeis Guineensis. Nikon Coolpix 4500 digital camera with tele-converter zooming and the Keyence machine vision were used to capture the FFB images in actual oil palm plantation. The images from the Nikon digital camera were analysed for optical properties and then compared with the value obtained from Keyence machine vision. The images of oil palm FFB in plantation were captured with setting cameras parameter namely shutterspeed which set to 0.125 seconds, image sensor’s sensitivity (ISO) was set to Normal and white balance were calibrated using the standard white calibration CR-A74. The lighting intensity under oil palm canopy was simultaneously recorded and monitored using Extech Light Meter Datalogger. On the same day, the fruitlets were plucked from FFB and analysed for its oil Mesocarp content using the Soxhlet Extractor apparatus. The calculations to determine the Mesocarp oil content was developed based on the ratio of oil to dry Mesocarp. The MPOB colorimeter was used to validate and compare for the ripeness level. Regression analysis of polynomial 2nd order model shows that the optical property of oil palm fruit was significant in determining the oil from the Mesocarp fruit with respect to the degree of maturity. The formulated predicted equation was Y = -0.0116X2 + 5.2376X – 514.88   (R2 = 0.884) with Y as the Mesocarp oil content and X as the Hue optical property. High correlation of Hue digital value was found between the developed systems using the Nikon digital camera and the Keyence machine vision with correlation close to 0.929 and accuracy of 97%. Simulation model was developed for estimation the harvesting days of FFB on the basis of its oil content. The verification on calculating the harvesting day of FFB was based on previous research of destructive method. The graph to determine the day of harvesting the FFB was contributed in this research. The oil was found to start developing in Mesocarp fruit at 65 days before fruit at ripe maturity stage of 75% oil to dry Mesocarp weight. The audience aims is to investigate innovation applications and last researches in the areas of applied signal processing and digital image for agriculture application.
Keywords: Image processing, Palm oil, Maturity recognition system, Non-destructive technique