Abstract:A measurement of 3 D locations of ripe tomatoes by binocular stereo vision was done for harvesting tomatoes in greenhouse. In this method, a pair of stereo images were obtained by stereo cameras, which were then transformed to gray images. According to the gray correlation, corresponding points of stereo images were searched, and a depth image was obtained by calculating distances between tomatoes and stereo cameras based on triangulation principle. The center of tomato was extracted by distinguishing a tomato from background with image processing. The 3D locations of ripe tomatoes were obtained by comparing coordinate values of the center of tomato ±20 mm when distance is less than 1 000 mm. with the depth image. The error of depth is within