Aiming at the need of the table tennis picking robot to obtain the position information of the table tennis during operation

a table tennis recognition and positioning system based on monocular vision is designed.Firstly

the image information of the working scene is collected by a monocular camera.After obtaining the image information

the control unit uses the YOLOv5s recognition algorithm to identify the table tennis and finds the pixel coordinates of the center point of the table tennis.Then

Zhang Zhengyou’s camera calibration method is used to obtain the parameter information of the monocular camera

combined with the monocular camera ranging procedure to complete the calculation of the three-dimensional coordinates of the table tennis ball.Finally

the planar coordinates of the table tennis ball relative to the camera are output.The experimental scene is set up

and the table tennis in the scene are identified and located in the experiment.The experimental results show that the designed table tennis recognition and positioning system has a high target recognition accuracy

and the positioning error of the table tennis is within the acceptable range of the robot

which proves that the system can meet the needs of the robot for table tennis ball recognition and positioning.