Detecting Ripeness of Strawberry and Coordinates of Strawberry Stalk using Deep Learning

Abstract

The ripeness of strawberries is an important criterion for determining whether strawberries can be picked or not, and the strawberry stalk coordinates of strawberries are important to automate strawberry harvesting. We propose a model that simultaneously learning the ripeness of strawberry and the coordinates of the stalk through two-path model with semantic segmentation. Two-path model can be seen as a model that performs two tasks at the same time. This model detecting the ripeness of strawberries with 90.33% accuracy and recognizing the coordinates of the stalk with 71.15%

Topics & Concepts