SyncAI.news, a Varaisys broadcasting
Defect-Aware Masked Autoencoders for Unsupervised Clustering of Steel Surface Defects
SP

Shrey Patel

· 1 min read

ResearcharXiv cs.CV

Defect-Aware Masked Autoencoders for Unsupervised Clustering of Steel Surface Defects

arXiv:2611.20645v1 Announce Type: new Abstract: Reliable recognition of steel surface defects is essential for automated quality inspection, but supervised methods require manually annotated images that are costly to produce. This creates a need for methods that can learn useful defect representations directly from unlabelled surface images. This work presents Defect aware MAE, a label free masked autoencoder for clustering steel surface defects. Each image is divided into 196 patches, with 147 masked and 49 processed by a Vision Transformer encoder. Texture energy based masking preferentially hides structurally informative patches, while a shallow decoder reconstructs HOG descriptors that capture local texture and edge orientation. Strong augmentation and an intermediate layer readout are used to form the final image features. Pretraining uses no class labels, bounding boxes, detection branch, or contrastive loss. On the held out NEU DET split, UMAP followed by Ward clustering achieves 98.6% Hungarian matched accuracy, 0.968 NMI, 0.967 ARI, and 0.838 silhouette score. Under an identical encoder and training schedule, it exceeds a SimCLR contrastive baseline in accuracy, NMI, ARI and silhouette scores across 5 seeds.

Original source

This story was published by arXiv cs.CV and written by Shrey Patel. SyncAI.news shows a preview; the complete article is on the publisher's site.

Read the full story on arxiv.org

Similar News