Comparison between colour models in automatic identification of cane sugar

Abstract

In automation and standardization of quality of cane sugar in sugar factory, quantized identification process needs to be done. Identification of cane sugar was done based on image of cane sugar. In classification and identification based on image, colour models used could influence success rate of identification. This paper presents comparative study among RGB, HSV, HSI, YCbCr, and L*a*b colour models in automatic identification of cane sugar. System designed could identify 8 kinds of cane sugar based on their image with success rate of 85%. System was designed with Artificial Neural Network classifier with one hidden layer using Levenberg-Marquardt algorithm. Colour and textural features were extracted from 120 images of cane sugar for Artificial Neural Network inputs. HSV was the best colour model for identification, with highest result of 87.5%, followed by YCbCr, L*a*b, and RGB.