A Machine Vision System for Tomato Cluster Harvesting Robot

Abstract

Dutch style greenhouse for tomato production has become popular recently in many countries while cluster tomatoes have gained popularity among consumers. To improve harvest efficiency of the cluster tomatoes in large scale Dutch production systems, it is desirable to replace manual labor with automated machines. In this paper, a machine vision system developed for autonomous tomato fruit cluster harvesting is described. Since the difficulty of recognizing the grasping point depended on exposure of plant parts and on robot access angle, acquired images were classified into three groups. The research results show a 73% success rate in automatically locating grasping points for the robotic end-effector on main stems of the cluster tomatoes that can be visually identified by human eyes.