ICUMSA identification of granulated sugar using discrete wavelet transform and colour moments

Abstract

Classification and identification of granulated sugar in Indonesia were previously done with no quantitative standard. In the production of granulated sugar, several stages and condition produce different kinds of sugar, resulting the need of supervision. Standardisation was designed to follow ICUMSA, a standard based on chemical process. System was designed to identify ICUMSA value of granulated sugar from its image. System was designed as Multi-Level Perceptron Artificial Neural Network with one hidden layer of five neurons using Levenberg-Marquardt algorithm with output trained to follow known ICUMSA values. Colour and textural features were extracted from 180 images of granulated sugar for Artificial Neural Network inputs. Colour moments, Haralick features, and symlet wavelet transform were used as features. After feature reduction, the designed system correctly identified ICUMSA and classified the 6 samples of granulated sugar with 3.623% of error.