PCA-Guided Autoencoding for Structured Dimensionality Reduction in Active Infrared Thermography

2025-08-11

AI Summary

Nonlinear autoencoders (AEs) for dimensionality reduction in active infrared thermography (AIRT) often yield unstructured latent spaces, hindering effective defect representation. Method: This paper proposes a PCA-guided structured autoencoding framework, introducing a PCA distillation loss that explicitly aligns AE latent representations with the linear subspace identified by principal component analysis—thereby endowing the latent space with interpretable geometric and semantic structure while preserving nonlinear modeling capacity. The approach is fully unsupervised and architecture-agnostic. Results: Evaluated on AIRT datasets of PVC, CFRP, and PLA materials, the method significantly outperforms state-of-the-art dimensionality reduction techniques: it improves defect region contrast by +23.6%, signal-to-noise ratio by +18.4%, and downstream neural network classification accuracy by +5.2–9.7 percentage points. It establishes an interpretable, robust low-dimensional representation paradigm for quantitative AIRT defect characterization.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Representation Learning for Vision
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Active Infrared thermography (AIRT) is a widely adopted non-destructive testing (NDT) technique for detecting subsurface anomalies in industrial components. Due to the high dimensionality of AIRT data, current approaches employ non-linear autoencoders (AEs) for dimensionality reduction. However, the latent space learned by AIRT AEs lacks structure, limiting their effectiveness in downstream defect characterization tasks. To address this limitation, this paper proposes a principal component analysis guided (PCA-guided) autoencoding framework for structured dimensionality reduction to capture intricate, non-linear features in thermographic signals while enforcing a structured latent space. A novel loss function, PCA distillation loss, is introduced to guide AIRT AEs to align the latent representation with structured PCA components while capturing the intricate, non-linear patterns in thermographic signals. To evaluate the utility of the learned, structured latent space, we propose a neural network-based evaluation metric that assesses its suitability for defect characterization. Experimental results show that the proposed PCA-guided AE outperforms state-of-the-art dimensionality reduction methods on PVC, CFRP, and PLA samples in terms of contrast, signal-to-noise ratio (SNR), and neural network-based metrics.

Problem

• Structured dimensionality reduction for AIRT data
• Aligning latent space with PCA components
• Improving defect characterization in non-destructive testing

Innovation

• PCA-guided autoencoder for structured dimensionality reduction
• PCA distillation loss aligns latent space with PCA components
• Neural network metric evaluates defect characterization suitability