Mehrdad Moradi

RDDPM: Robust Denoising Diffusion Probabilistic Models for Unsupervised Anomaly Detection and Segmentation in Contaminated Data

Anomaly detection in noisy images is essential for quality control in industries like steel, composites, and textiles. Traditional unsupervised anomaly segmentation models, including Robust Principal Component Analysis and Smooth Sparse Decomposition, rely on restrictive data assumptions including anomaly sparsity and...