Multiclass classification of Ethiopian coffee bean using deep learning


  • Getabalew Amtate
  • Dereje Teferi


Ethiopian coffee; Coffee bean classification; Deep learning; CNN


Ethiopia is the homeland of Coffee Arabica. Coffee is the major export commodity and a high-income source of foreign currency for the country. In addition to this, coffee has a great role in social interaction between people and is also a source of income for the coffee-producing farmers. Several types of coffee beans grow in Ethiopia. These beans are distinct from each other in terms of quality, color, shape etc. based on their geographical origins. Classification of these coffee beans are based on growing origin, altitude, bean shape and color, preparation method and others. However, the quality of the coffee beans is determined by visual inspection, which is subjective, laborious, and prone to error. This creates the necessity for the development of an automatic method that is precise, non-destructive and objective. Thus, this research aims to develop a model that classifies coffee beans of six different origins of Ethiopia (Jimma, Limmu, Nekemte, Yirgacheffe, Bebeka, and Sidama) in to nine classes. The dataset for this research is collected from the Ethiopian Coffee Quality Inspection and Auction Center (ecqiac). This research followed design science research (dsr) to investigate the problem. Image processing and the state-of-the-art deep-learning techniques were employed to automatically classify coffee bean images into nine different classes grown in six different regions of Ethiopia. A total of 8646 coffee bean images were collected and 1190 images were added using augmentation to make the total dataset 9836. The model is trained and tested by tuning the hyper-parameters of the cnn algorithm. When 80% of the dataset is used for training, 10% for validation, and the remaining 10% for testing, the proposed model achieved a 99.89% overall classification accuracy with 0.92% generalization log-loss. In conclusion, the result of this research shows that deep learning is an effective technique for classification of Ethiopian coffee beans and can be implemented in the coffee industry.