Image Processing Techniques For Post-Fusion Plasma Wall Monitoring In ITER Tokamak

Monitoring the structural integrity of the ITER tokamak first walls at regular intervals after fusion plasma runs is essential for ensuring the safety and operational efficiency of subsequent plasma operations. The ITER tokamak’s walls undergo significant wear and potential damage due to the extreme conditions of the plasma environment [1], e.g., during plasma disruptions. To monitor the surface state of the said walls, light could be produced by generating plasma glow through Ion Cyclotron Resonant Heating (ICRH) and capturing images of the walls under these conditions with high-resolution visible cameras. However, the intense plasma glow causes hazy images that hinder accurate analysis. Existing dehazing methods like Dark Channel Prior (DCP) are ineffective for plasma images, as they are designed for atmospheric haze [2,3]. This study focuses on developing a comprehensive image dehazing methodology to assess the condition of these walls post-fusion plasma. Our group analyzes images captured during Ion Cyclotron Wall Conditioning (ICWC) campaigns in current tokamaks, demonstrating the effectiveness of dehazing techniques in improving visibility and diagnostics. The enhanced images can then be carefully analyzed to identify potential signs of damage, such as erosion, cracks, or other structural degradation. This work contributes to the ongoing maintenance and optimization of fusion reactors and supports the advancement of fusion energy research.

Keywords: ITER Tokamak; ICRH; Image processing; Dehazing algorithm

References:

[1] G. Federici, “Plasma wall interactions in ITER,” Phys Scr, vol. T124, pp. 1–8, May 2006.

[2] J.-L. Lisani and C. Hessel, “Dehazing with Dark Channel Prior: Analysis and Implementation,” Image Processing On Line, vol. 14, pp. 173–193, Jun. 2024,

[3] Kaiming He, Jian Sun, and Xiaoou Tang, “Single image haze removal using dark channel prior,” in 2009 IEEE Conference on Computer Vision and Pattern Recognition, IEEE, Jun. 2009, pp. 1956–1963.