Image-Intrinsic Priors for Integrated Circuit Defect Detection and Novel Class Discovery via Self-Supervised Learning

TL;DR

This paper introduces IC DefectNCD, a self-supervised framework utilizing image intrinsic priors for defect detection and novel defect class discovery in integrated circuit manufacturing, reducing reliance on annotations and improving robustness.

Contribution

The paper presents a novel self-supervised approach that leverages image intrinsic priors for defect detection and classification, enabling detection of unseen defects without extensive annotations.

Findings

Robust defect detection across multiple fabrication stages.

Effective classification of unseen defect types.

Superior performance compared to existing unsupervised methods.

Abstract

Integrated circuit manufacturing is highly complex, comprising hundreds of process steps. Defects can arise at any stage, causing yield loss and ultimately degrading product reliability. Supervised methods require extensive human annotation and struggle with emergent categories and rare, data scarce defects. Clustering-based unsupervised methods often exhibit unstable performance due to missing priors. We propose IC DefectNCD, a support set free framework that leverages Image Intrinsic Priors in IC SEM images for defect detection and novel class discovery. We first develop Self Normal Information Guided IC Defect Detection, aggregating representative normal features via a learnable normal information extractor and using reconstruction residuals to coarsely localize defect regions. To handle saliency variations across defects, we introduce an adaptive binarization strategy that produces stable subimages focused on core defective areas. Finally, we design Self Defect Information Guided IC Defect Classification, which incorporates a soft mask guided attention mechanism to inject spatial defect priors into the teacher student model. This enhances sensitivity to defective regions, suppresses background interference, and enables recognition and classification of unseen defects. We validate the approach on a real world dataset spanning three key fabrication stages and covering 15 defect types. Experiments demonstrate robust performance on both defect detection and unseen defect classification.