A video ball-beam system balance position detection system is designed in this thesis. A CMOS camera used to catch the ball-beam video streaming. Using the color image processing algorithms, the equilibrium position of ball-beam is captured and it marked as a cue control the position for equilibrium. Ball-beam system designed from self-tuning PID control fuzzy control scheme. The location of equilibrium position from image is tracked by ball-beam automatically.The following work is completed in this thesis:1. Lagrange equation is used for set up a model for the ball-beam system, and stability of the system is analyses. The fuzzy self-tuning PID control scheme is designed to achieve a balanced position on this system. Firstly, through the MATLAB SIMULINK, and then applied to the ball-beam system to achieve equilibrium position of the actual control system.2. Video stream processing algorithm design:The background model set up, and then designs algorithms to extract the location marker.3. Real-time location of the marker delivery shaft system, system controls the ball-beam to achieve a balance in the specified location.All the algorithms are simulated in the MATLAB, and then programming downloaded from the PC to the ball-beam machine control system implemented in real time. Experimental results show the correctness and effectiveness. |