Banana Ripeness Classification using HSV Colour Space and Nearest Centroid Classifier

Banana is a common fruit which found throughout Southeast Asia and are beneficial both as delicacy dessert good for health. Although quite easy to obtain, many people today find it difficult identify the correct ripeness stage of banana, especially when purchasing from traditional vendors, where varying degree available. This research sought possibility classify banana by its peel colour with HSV space feature classified using Nearest Centroid Classifier (NCC). ‘Ambon Lumut’, ‘Kepok’ ‘Raja’ bananas used examples they among most types available use in Indonesia, divided into 4 classes according different usage banana: unripe, almost ripe, overripe. Photographic images training test data. The experiment conducted cleaned have background removed, this also resulting 73.33% recognition. recognition results each class respectively are: Green = 93.33%; Almost Ripe=80%; Ripe=66.67% Overripe=53.33%.