Wavelet-optimized pseudo-3D accelerated diffusion model for truncated Computed Laminography

doi:10.1016/j.ndteint.2026.103862

Wavelet-optimized pseudo-3D accelerated diffusion model for truncated Computed Laminography

2026-08-22

0

PRE

AI

DOI：10.1016/j.ndteint.2026.103862

Original

Original request for help

Save

Abstract

En 中文

• We propose a novel truncation reconstruction method that effectively recovers data-incomplete regions, expanding the effective field-of-view (FOV) and scanning efficiency. • We develop a highly efficient pseudo-3D diffusion strategy tailored to resolve the CL truncation problem, which promotes volumetric data consistency without the computational burden of native 3D models. • We combine z -directional wavelet regularization with a translation-invariant (TI) mechanism and a low-frequency preservation strategy to robustly suppress inter-slice discontinuity artifacts and mitigate aliasing. • We introduce a fast-sampling architecture adapted for 3D volumetric data, successfully alleviating the bottleneck of slow inference speeds typically associated with diffusion-based generative models.

Wavelet-optimized pseudo-3D accelerated diffusion model for truncated Computed Laminography

Abstract

Journal

Organization

Cited Papers