Cherry Tomato Harvesting Robot Using 3-D Vision Sensor (Part 1)

Recognition of Cherry Tomato

Abstract

A cherry tomato harvesting robot equipped with a 3-D vision sensor was made for experiment. The 3-D sensor emitted three laser beams which contained two components of wavelength; a red (wavelength: 685nm) and an infrared (830nm). Three PSDs were used to detect the lights which were reflected from the crop. The shape of the crop was measured by scanning the laser beams. Red ripe tomatoes were detected by the ratio of the red to the infrared reflected signals. For the purpose of collision avoidance, the locations of the stems and the leaves were recognized by the image processing software.