Plasma Shape Identification and Fitting on the Basis of Image Processing on Tokamak

To perform experiments and data evaluation in tokamak devices, plasma shape identification is needed for plasma equilibrium control and analysis. A new framework for plasma shape identification is currently being developed in EAST. The foundations of the system are high speed CCD, the remote computer and the algorithms for plasma boundary identification and fitting. In this paper, a new quasi-real time method for plasma shape identification, based on the modified Canny algorithm and carried out in EAST, is proposed. Comparison with the result of EFIT and other diagnostic data, our method has not only been verified the stability and dependability of the detection method, but also revealed it being a reliable tool for operators and physicists to analyze the behavior of plasma. The real time plasma shape identification can be introduced to the plasma control system in the coming campaign of EAST.