The agricultural and plantation product processing industry is growing
rapidly, agriculture is one of the most important sectors in Indonesia. Indonesia is
one of the largest mango producing countries in the world. Unfortunately, even
though the amount of production is abundant and has a large market, mango
production is still done manually. This is due to the lack of technological
breakthroughs for mango farmers. Based on the above problems, research on
mangoes will be carried out using the Convolutional Neural Network (CNN)
method, as a solution in classifying ripe honey mangoes, raw honey mangoes, and
non-mangoes. The purpose of this study was to build a mango classification
system using the CNN method in order to facilitate farmers in classifying ripe
mangoes, unripe mangoes and non-mangoes. The search results obtained from
research on honey mango fruit were 81.64%.