BerryScope: AR Strawberry Picking Aid

Strawberry picking farms have the problem that visitors do not choose fully ripe straw berries. This is because fully ripe strawberries can easily get damaged during transportation, meaning customers usually see strawberries that are not fully ripe in stores. Therefore, they are likely to pick the strawberries similar to what they see in stores. As a result, visitors may miss the opportunity to taste the most delicious strawberries, and for the farmers, it leads to waste of the ripe strawberries. To solve the problem, this study proposes BerryScope, an augmented reality (AR) system for supporting strawberry picking. Visitors wear an optical see-through head-mounted display (OSTHMD) to pick strawberries. Based on the images captured by the OSTHMD's camera, the system uses deep neural network (DNN) technology to analyze the ripeness of the strawberries and overlays harvest instructions on fully ripe strawberries. We conducted experiments at an actual strawberry farm involving 15 people and confirmed that using the proposed system enables users to choose more ripe strawberries.