The Development of Automatic Coffee Sorting System Based on Image Processing and Artificial Neural Network

Abstract

The objective of this research was to develop an automatic sorting system which consist of a computer program of image processing and artificial neural network to identify quality of green coffee. Image of the green coffee will be analyzed to get six quality parameters which match the green coffee quality criteria namely length, area, perimeter, defect area, index of red color, and index of green color. Those six quality parameters will be used as training data inputs (75 percent) of the developed artificial neural network (ANN). The weights of the ANN architecture will be used to identify the quality class of testing data (25 percent), then integrating with image processing program so the program can identify green coffee quality class automatically. The outputs of the program will be used as inputs for PPI 8255 as the I/O expansions port for the PC. The output of PPI 8255 will be connected with sorting simulator as the replacement of sorting mechanism.