Design of assembly line material sorting system based on machine vision

In order to solve the problems of low recognition accuracy and inaccurate automatic grasping effect in industrial production lines, a vision-based robotic arm grasping system was designed. openmv module was used to collect image data of objects in motion, and then the image data of objects were processed to determine the shape, color and position of the objects to be measured. Then the motion control module controls the six-axis manipulator to realize the sorting operation of the object. The device uses STM32 single chip microcomputer as the core controller to control the grasp of the robot arm and the rotation of the conveyor belt. Through image processing to complete the positioning, through coordinate transformation, inverse kinematics to complete the motion control. Finally, the materials of different colors were sorted, and the average grasping success rate reached 98%. The results show that the system has a strong practical significance for industrial production line.