Artificial Neural Network for Classification of Immature and Mature Coffee Beans Using RGB Values

Abstract

© 2020, World Academy of Research in Science and Engineering. All rights reserved. This paper is about differentiating immature coffee beans from mature coffee beans using their red, green and blue color values, and using these features as variables for classification using the feed forward back propagation artificial neural network. This paper is an extension of the previous article wherein the classifier used are the 23 machine learning algorithms of MATLAB’s Classification Learner App. The same dataset was used from the previous study but the classifier was replaced by the feed forward back propagation artificial neural network. In the previous study the highest classification accuracy was achieved by the Quadratic Support Vector Machine with 94 % accuracy and 0.62 seconds training time. In this paper the feed forward backpropagation artificial neural network achieved 95 % classification accuracy and training time of 5 minutes. Comparing the two classifier in differentiating immature coffee beans from mature coffee beans, higher accuracy was achieved by feed forward backpropagation artificial neural network but it also required longer training time.