PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
March 10, 2026Expert Systems with Applications2 citationsOpen Access

Interpretable semi-supervised 3D deep anomaly detection for surface defect localization in wire-laser directed energy deposition using point clouds

View Full Paper
ZWZhiyuan WangYGYang GaoLYLong Ye

Key Points

  • This research aims to develop a semi-supervised framework for detecting and localizing surface defects in wire-laser directed energy deposition components.
  • Utilized high-density 3D point clouds for surface evaluation
  • Employed patch-based segmentation and voxelization
  • Trained convolutional autoencoder and variational autoencoder on normal patches
  • Mapped reconstruction-error heatmaps to localize defects
  • Conducted hyperparameter study on voxel resolution and patch size
  • Optimized convolutional autoencoder achieved 86.09% precision
  • Variational autoencoder reached 86.43% precision and improved defect localization with mIoU of 0.7234
  • Activation-map analysis provided interpretability by highlighting key geometric areas
  • Lower voxel resolutions and smaller patch sizes reduced false positives effectively

Abstract

Wire-laser directed energy deposition (WL-DED) enables the fabrication of large-scale metallic components but frequently suffers from process-induced surface defects that hinder part quality and increase post-processing costs. Automated inspection is challenging because defects are diverse and rare, making large labelled datasets impractical. This paper proposes an interpretable semi-supervised framework for surface detection and localization on WL-DED components using high-density 3D point clouds acquired by laser scanning. The workflow includes point-cloud preprocessing, patch-based segmentation, voxelization, and semi-supervised representation learning of defect-free surface morphology. Two 3D deep autoencoder models, i.e., a convolutional autoencoder (CAE) and a variational autoencoder (VAE), are trained exclusively on normal patches and detect anomalies through voxel-wise reconstruction errors. Defects are localized by mapping reconstruction-error heatmaps back onto the original surface, enabling quantitative visualization of defect severity. Experimental results on WL-DED thin-wall samples show that the optimized CAE achieves 86.09% precision, while the VAE reaches 86.43% precision with improved defect localization (mIoU up to 0.7234). Activation-map analysis provides interpretability by highlighting geometric regions that drive anomaly responses. A hyperparameter study demonstrates that lower voxel resolutions and smaller patch sizes improve robustness and reduce false positives. The proposed framework generalizes to more complex multi-bead, multi-layer structures with minimal retraining, supporting practical deployment for intelligent inspection and decision-making in additive manufacturing quality assurance.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Wang et al. (2026) studied this question.

synapsesocial.com/papers/69af949670916d39fea4b996https://doi.org/10.1016/j.eswa.2026.131963
Ask AI
Helpful
Bookmark
Share
View Full Paper