ESPI Fringe Denoising Using a Multi-Stage Progressive Image Restoration Network

Fringe pattern denoising is a key issue in electronic speckle pattern interferometry, especially in the processing of increased fringe pattern density. This paper reports an electronic speckle pattern interferometry fringe pattern denoising method based on multi-stage progressive image restoration network. The method decomposes the complex denoising task into multiple subtasks. A three-stage progressive processing architecture is adopted. Through multi-stage collaboration and feature fusion, the method effectively balances noise suppression and fringe pattern detail preservation. The mean squared error, peak signal-to-noise ratio, and structural similarity index measure metrics are used to quantitatively analyze the denoising effect. The denoising results on measured fringe pattern verify the effectiveness and robustness of the proposed method.