Ripeness detection simulation of oil palm fruit bunches using laser-based imaging system

Ripeness is one of important factors for quality sorting of harvested oil palm fresh fruit bunches (FFB). Traditional ripeness classifications using FFB color and number of fruit loose for harvesting have some disadvantages especially for high oil palm trees. A laser based imaging system is proposed to substitute the traditional method. In this study, ripeness detection simulation of oil palm FFBs was performed. The system composed of two diode lasers with 532 nm and 680 nm in wavelengths and a CMOS camera which was set on a rotating plate for easy adjustment of laser beam hitting FFB. The FFB samples were placed on an aluminum platform with 4 height variations, 1.5 m, 2 m, 2.5 m, and 3 m. The relations of reflectance intensities represented by Red Green Blue (RGB) values of the FFB images to the height variations and ripeness levels of FFBs with and without laser beam were analyzed. The samples were from Tenera variety with 4 ripeness levels called F0, F1, F3, and F4. The results showed that the red component of RGB values were dominant for FFBs without laser and with red laser. The average RGB values are higher for F3 (ripe) level and F4 (overripe). Imaging with green laser showed consistency. Imaging methods using laser was able to differentiate ripeness levels of oil palm fresh fruit bunch, it could be applied for future remote detection of oil palm FFB ripeness.