Principal Component Analysis-Based Terahertz Self-Supervised Denoising and Deblurring Deep Neural Networks

2026-01-17

AI Summary

Terahertz imaging is often simultaneously degraded by low-frequency blur and high-frequency noise, and conventional methods struggle to address both issues effectively while relying heavily on manual parameter tuning. This work proposes a novel framework that integrates principal component analysis (PCA) with self-supervised deep learning, introducing PCA into self-supervised image restoration for the first time. By employing a Recorrupted-to-Recorrupted strategy, the model learns noise invariance and achieves joint denoising and deblurring across the full frequency spectrum. Notably, the method requires no ground-truth labels and operates effectively with only a small amount of unlabeled data, successfully recovering fine image details while preserving the physical characteristics of the original signal. Extensive experiments on diverse samples demonstrate its superior performance.

Technology Category

• Machine Learning: Unsupervised & Self-Supervised Learning
• Computer Vision: Image and Video Retrieval
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Privacy-enhancing technologies
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Terahertz (THz) systems inherently introduce frequency-dependent degradation effects, resulting in low-frequency blurring and high-frequency noise in amplitude images. Conventional image processing techniques cannot simultaneously address both issues, and manual intervention is often required due to the unknown boundary between denoising and deblurring. To tackle this challenge, we propose a principal component analysis (PCA)-based THz self-supervised denoising and deblurring network (THz-SSDD). The network employs a Recorrupted-to-Recorrupted self-supervised learning strategy to capture the intrinsic features of noise by exploiting invariance under repeated corruption. PCA decomposition and reconstruction are then applied to restore images across both low and high frequencies. The performance of the THz-SSDD network was evaluated on four types of samples. Training requires only a small set of unlabeled noisy images, and testing across samples with different material properties and measurement modes demonstrates effective denoising and deblurring. Quantitative analysis further validates the network feasibility, showing improvements in image quality while preserving the physical characteristics of the original signals.

Problem

• Terahertz imaging
• frequency-dependent degradation
• denoising
• deblurring
• image restoration

Innovation

• self-supervised learning
• principal component analysis
• terahertz imaging
• image denoising and deblurring
• recorrupted-to-recorrupted