Ripeness level classification of pineapple using image processing with neural network

Pineapples are an economic harvest that generates income for Thailand, especially fresh pineapples because they are one of the growth export fresh fruit rates in 2021 according to the report from National News Bureau of Thailand. Then the proper classification for the quality of the fresh pineapple is much more important for the export market in Thailand. Therefore, the purpose of this study is to develop an automatic tool for ripeness level classification of pineapple using image processing with neural network. The ripeness of pineapple is divided into three levels: unripe, ripe, and fully ripe. The method used image processing that image converted from RGB color space to HSV color space and morphological image processing. For the classification, we applied the neural network consisting of the input layer with one node, the hidden layer with one node, and the output layer with three nodes. The results achieve high classification accuracy around 92. 31% . The study demonstrates that the proposed technique is appropriate for an automated classification system.