Winda Andrayani Ahmad, Andi Akram Nur Risal, Dewi Fatmarani Surianto, Abdul Wahid
Papaya, a tropical plant commonly found in regions such as the South Pacific and other tropical areas, thrives in low-temperature environments, making it ideally suited for tropical climates. In addition to its popularity as a fresh fruit, papaya is extensively used in various preparations. However, one of crucial post-harvest processes to sort its quality is still conventionally carried out by manual human labor, presenting inherent limitations and shortcomings. This study employs the artificial neural network method to classify the quality of papaya fruit based on three features which are shape, texture and sugar content levels to address this issue. The research encompasses distinct stages, including image acquisition, preprocessing, segmentation, morphological operations, feature extraction, and classification. The dataset used comprises 600 data points, divided into 480 for training and 120 for testing. The outcomes reveal that employing the artificial neural network algorithm based on these features can achieve a commendable accuracy of 90%, with an ME (Mean Error) of 10. The results indicating that by using these features, the papaya quality classification system accomplished within a remarkably swift time of only 1032.705593 seconds. © 2024 Author(s).
Departement of Computer Engineering, State University of Makassar, Makassar, Indonesia
