Thai banana quality predictor by computer vision

Abstract

This paper proposed the quality of Thai banana from computer vision. CIE-L * a * b * (Commission International de l'Eclairage) was used for RGB color conversion. Three types of Thai banana were tested including Cavendish, Lady Finger and Pisang Awak. The results show that the L * and b * values ​​of the three types of banana indicated the quality of the banana, because L * represents the brightness of the color and b * represents the color from yellow to blue. The value of a* represents the color from red to green and it changes color from the green of the raw banana to the yellow of ripe banana. Each banana has a different maturation time to know the ripening cycle of banana. It is very useful for exporters to plan for exporting banana considering the ripening cycle of banana to solve the problem of banana ripening before reaching customers.