Machine Sort of Color and Size for Roasted Coffee Using Digital Image Processing

Abstract

Due to the difficulties of sorting the roasted coffee which is usually sorted by visual method by the operators. The method should be performed on expertise. However, this would be affected such as feeling, emotional and expertise skill that lead to uncontrol the roasted coffee sorting. For those reason, it led to the study of coffee bean classification. The coffee bean classification was group in color (in low, medium and high intensities) and size (in smaller than 10 mm, between 10 to 13.5 mm and bigger than 13.5 mm), which is can be classified in to 9 types. In this study the digital image processing technique on the HSV color space is applied in which to classify the color levels and the sizes of the coffee beans. The proposed method was established and developed, then conduced to compare the performance with the operator. The results showed that the proposed method provide the performance on 70.00 % which batter the operators with 37.50%