Fuente: WIPO "processing tomato" The invention relates to the field of machine vision. The invention aims to provide a plant and fruit detection algorithm of a tomato picking robot, and the algorithm needs to realize collision-free picking and has the characteristics of high detection speed and high precision. According to the technical scheme, the plant and fruit detection algorithm of the tomato picking robot comprises the following steps: 1-1) acquiring a tomato image; 1-2) inputting the tomato image into a multi-task convolutional neural network model for processing; 1-3) obtaining a bounding box, a key point and a stem of the tomato on the tomato image; the establishment method of the multi-task convolutional neural network model comprises the following steps: 2-1) constructing a data set; 2-2) preprocessing the data; 2-3) selecting YOLO V5s as a basic network model, and adding a key point prediction parameter and a semantic segmentation module into the basic network model; and 2-4) determining model training hyper-parameters, constructing a model loss function, and inputting the data set into the basic network model for training and testing to obtain a multi-task convolutional neural network model.