Research on point cloud segmentation method of mutually adhered cherry tomatoes based on three-dimensional data

In order to improve the accuracy of recognition and segmentation of occluded and stuck cherry tomatoes in three-dimensional infrared images, an improved watershed segmentation method based on point cloud spatial density cluster analysis is proposed. First, a binocular infrared camera is used to collect a three-dimensional point cloud image, and then the point cloud holes are filled with geometric diffusion. The completed point cloud model is passed through a color region growing segmentation algorithm to extract the fruit bunch target. The phenotypic characteristics of the target are analyzed through point cloud density clustering, and then the concave-convex relationship is used to enhance the edge characteristics of the fruit. Finally, secondary clustering is performed based on the watershed idea to separate the blocked and adherent fruits. Experimental results show that the algorithm has a recall rate of 88.9% in target recognition of stuck cherry tomatoes, which improves the three-dimensional segmentation performance of stuck cherry tomatoes.