Ping-pong ball trajectory tracking device and method based on deep learning

Abstract

The invention discloses a ping-pong ball trajectory tracking device and method based on deep learning. The device comprises a posture sensing racket, a six-axis sensor, wireless transmission equipment, a first camera, a second camera, USB 3.1 transmission equipment, an FPGA, a visual processing unit, a central processing unit, a display screen, a memory, a GPRS module and network equipment. Two cameras capture table tennis pictures, the FPGA, the visual processing unit and the central processing unit are combined with the posture sensing racket to obtain a table tennis track, track analysis is conducted, the ball hitting condition of a player is obtained, the memorizer is used for storing data, the data is uploaded to the cloud through the GPRS module, and real-time checking is achieved. The three-dimensional position of the ping-pong ball is obtained in real time, the motion trail of the ping-pong ball is drawn, ball hitting conditions of players are analyzed, equipment is arranged in an embedded mode, utilized resources are small, and operation is easy and convenient.

Description

technical field [0001] The invention belongs to the embedded field, and in particular relates to a table tennis trajectory tracking device and method based on deep learning. Background technique [0002] At present, the common analysis of table tennis events is through the subjective judgment of human eyes and coaches. Because the coach observes the trajectory from a single perspective, there will be visual misalignment of the position of the table tennis ball. The coach cannot accurately obtain the three-dimensional trajectory information of the table tennis ball. Correct game analysis has high requirements on the ability of coaches, which also makes it impossible for players to get better batting plans in most table tennis games. In addition, in some cases, the coach needs to review the previous game video to review the table tennis game. The multiple switching of the camera perspective and a single perspective will also affect the coach’s judgment on the trajectory of the...

AI Technical Summary

Problems solved by technology

[0002] At present, the common analysis of table tennis events is based on the subjective judgment of human eyes and coaches. Because the coach observes the trajectory from a single perspective, there will be visual misalignment of the position of the table tennis ball. The coach cannot accurately obtain the three-dimensional trajectory information of the table tennis ball.

Correct game analysis has high requirements on the ability of coaches, which also makes it impossible for players to get better batting plans in most table tennis games

In addition, in some cases, the coach needs to review the previous game video to review the table tennis game. The multiple switching of the camera perspective and a single perspective will also affect the coach’s judgment on the trajectory of the table tennis ball and affect the coach’s understanding of the game. correct analysis of

[0003] In the daily training of table tennis players, it is also difficult for table tennis players to objectively understand the law of their hitting trajectory, and it is difficult to find the correct method to correct their own hitting shortcomings. The scientific analysis of table tennis players' daily training has also become a major problem

Images