Design to prediction tools for banana maturity based on image processing

Abstract

Image processing is a method to process image become information consisted of phisycal condition of product. Image processing applications in agriculture are currently developing very rapidly, especially for the prediction of fruit quality and maturity. The purpose of this study is to design a tool to predict the maturity of bananas. The type of banana used in this study is barlin banana (Musa paradisiaca. L) with maturity levels 1, 3, 5 and 7. While the image processing characteristics use data R, G and B which are captured by the TCS3200 sensor. From a number of experiments, a distinguishing factor is obtained for the RGB value for each level of maturity, maturity 1 with the value of G> B; maturity 3 with Value R> G, Value (R-G) <9 and G Value 142; maturity 5 with a value of R> G, a value of 9 (R-G) and a value of G 142; maturity 7 with a value of R> G and a value of 142 G. The results of the prediction of the banana maturity using this tool for maturity 1 accuracy prediction 100%, maturity 3 accuracy predictions 80%, maturity 5n accuracy prediction 100% and maturity 7 accuracy predictions 60%.