The object recognition of the vision robot was studied in order to address the problem of how an industrial robot can rapidly and autonomously identify and grasp a specified target workpiece under multi-target conditions, and to apply the vision robot object recognition technology to the design of an intelligent grasping system for industrial robots. The research method is based on the pattern recognition of the image, the detection and positioning research, the relationship between the visual image and the workpiece positioning and grasping, and the proposed monocular vision grasping system based on the contour Hu-invariant moment fast template matching algorithm. The image acquired by the camera is pre-processed first, then the target workpiece is identified by the contour Hu-invariant moment template matching algorithm, the position of the identified target workpiece is obtained by using the principle of minimization of contour moment and second order moment of inertia, and finally the obtained position is sent to the robotic arm control system to guide the robotic arm grasping by establishing SO CKET communication. Based on DobotS tudio, DobotVisionStudio 1.4.2, MVS software development platform and DobotStudio robot, the intelligent gripping system was built and tested.