Mango Maturity Classification using VGG16

Mangoes are mainly cultivated in tropical and subtropical regions. The characteristics of mangoes are sour and sweet taste, pleasant aroma and rich vitamin content of nutrients, making them popular fruits. The work investigated mango ripeness classification using various machine learning classifiers by collecting the images of different stages of mango ripeness and then used them to train a Convolution Neural Network (CNN), MobileNet, ResNet50 and VGG16 classifier. The experimental results showed that the CNN, MobileNet, ResNet50 and VGG16 classifiers achieved the accuracies of 81.30%, 85.11%, 73.66% and 90.08%, respectively. The VGG16 achieved the highest classification accuracy, of which classification accuracies from class 0 to class 5 were 98.85%, 98.85%, 95.80%, 95.80%, 95.42% and 95.42%, respectively.