Discriminating Tomatoes Based on Computer Vision Technology in Natural Outdoor Scenes

Abstract

This paper investigated the use of chrominance and intensity information natural outdoor scenes as a means of guidance for a robotic manipulator in the harvest of fruit. A Bayesian classification model was developed which could discriminate tomatoes from the natural background using only color features in a digital color Image.The classification was evaluated by centroid calculation technique and it could correctly classify over 80% of the tomatoes pixels.