Abstract
Coffee is an agricultural product that not only functions as a fresh drink, but also comes from an annual plant. Indonesia is known as one of the largest countries in coffee production in the world after Brazil, Vietnam and Colombia in 2017. There are two of the most commonly known varieties of coffee, namely arabica coffee (Coffea arabica) and robusta coffee (Coffea canephora). Due to the similarities between the two types of coffee beans, many people, especially those who are not experienced in the world of coffee, have difficulty telling the difference. Therefore, we need a tool that can help overcome this problem, such as smartengine which can classify arabica and robusta coffee beans. The development of this smartengine follows the Software Developer Life Cycle method using a spiral approach which involves several cycle stages. The first stage involves creating a deep learning model using the Deep Learning Life Cycle method which consists of several steps. In the second stage, deep learning models are provided as a service that can be used by other applications through application programming interfaces (APIs). For smartengine implementation, Google Colab with Keras API and TensorFlow backend is used. This smartengine has the ability to detect coffee beans and also allows the retrain process if needed. Testing is carried out using the blackbox method, where the feature functionality of the smartengine is tested. This research succeeded in developing a smartengine that can detect Arabica and Robusta coffee beans.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call