Robot grabbing system and method based on small sample learning

The invention discloses a robot grabbing system and method based on small sample learning, and relates to the technical field of robot learning, the system comprises an image acquisition module, an image processing module and an action processing module; the image acquisition module comprises a depth-of-field camera and is used for acquiring images; the image processing module comprises a U-net3 + network structure; an image processor for processing images; and the action processing module comprises an ROS system and a corresponding package, and is used for converting the image information into control information for controlling the motor. The invention provides a segmentation network grabbing algorithm optimized aiming at the shielding condition, a U-net3 + network is enhanced by utilizing the characteristics of dropblock and a Mish activation function, a grabbing system aiming at small sample learning and shielding conditions is designed, the network accuracy is improved under different networks and differ