Automated Annotation of Shearographic Measurements Enabling Weakly Supervised Defect Detection

TL;DR

This paper introduces an automated pipeline that generates annotated datasets for shearographic defect detection, reducing manual effort and enabling scalable, weakly supervised learning in industrial settings.

Contribution

The authors develop a novel automated labeling method combining Grounded DINO, SAM masks, and YOLO labels for shearographic defect detection datasets.

Findings

Generated bounding boxes are suitable for weakly supervised learning.

High-resolution masks enable qualitative defect visualization.

The pipeline significantly reduces manual annotation effort.

Abstract

Shearography is an interferometric technique sensitive to surface displacement gradients, providing high sensitivity for detecting subsurface defects in safety-critical components. A key limitation to industrial adoption is the lack of high-quality annotated datasets, since manual labeling remains labor-intensive, subjective, and difficult to standardize. We present an automated labeling pipeline that generates candidate defect bounding boxes with Grounded DINO, refines them using SAM masks, and exports YOLO-format labels for downstream detector training. Quantitative evaluation shows the generated boxes are suitable for weakly supervised learning, while high-resolution masks provide qualitative visualization. This approach reduces manual effort and supports scalable dataset creation for robust industrial defect detection.