Utilization of the Deep Learning Convolutional Neural Network (CNN) Method to Classify Pineapple Ripeness Levels

Pineapple (Ananas Comocus L) thrive in tropical and subtropical regions. Pineapples are characterized by their unique traits, including a firm texture and richness in Vitamin C. Pineapple belong to the Non-Climacteric fruit category, and currently, the maturity classification process for pineapples heavily relies on human visual analysis. Human psychological factors can influence the classification results, potentially leading to errors. The main objective of this research is to classify the ripeness level of pineapple based on color, encompassing ripe, semi-ripe, and unripe categories. The aim to provide a more accurate and reliable automated solution in the classification process, utilizing, with the VGG16 model as a tool. A total of 598 data points were used for the training phase, while 40 data points were used for testing. The results showed and accuracy rate 95%, with precision at around 93%, a recall of 93%, and an F1 score reaching 93%. From these findings, it can be concluded that the Convolutional Neural Network method with the VGG16 model is capable of accurately classifying the ripeness of pineapple based on color.