Scholarlink

<p>Additive Manufacturing (AM), particularly Laser Powder Bed Fusion (L-PBF), faces critical challenges in defect detection due to the scarcity of high-quality training data and severe class imbalance, which significantly degrade the accuracy of deep learning models. To address these issues, this study proposes a novel data augmentation framework combining geometric transformations with an enhanced Variational Autoencoder-Generative Adversarial Network (VAE-GAN). Traditional augmentation techniques (rotation, scaling, flipping) are first applied to alleviate sample imbalance, followed by the improved VAE-GAN to synthesize high-fidelity defect images, thereby enriching dataset diversity. Experimental results on an L-PBF defect dataset demonstrate significant improvements in detection performance: Specifically, comparative experiments were conducted to evaluate the performance of YOLOv4, YOLOv7, YOLOv8, SSD, and Faster R-CNN on defect detection tasks before and after data augmentation. The results demonstrated significant mAP improvements across all models, with YOLOv4 achieving the most substantial enhancement (+ 19.94%, from 65.18% to 85.12%) despite its lower baseline performance. Faster R-CNN attained the highest post-augmentation mAP (87.69%), representing the best overall performance. YOLOv8 exhibited an optimal balance between real-time processing and accuracy (67.07%→84.62%, + 17.57%), approaching Faster R-CNN’s performance level. While SSD showed the smallest improvement (+ 12.40%), it maintained a relatively high baseline mAP (86.39%). These results validate the effectiveness of the proposed method in overcoming data scarcity and improving defect detection accuracy in AM.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Research on L-PBF defect detection based on VAE-GAN data augmentation with multi-head attention mechanism

  • 期刊论文

  • 发表日期：

    2026年2月19日

查看全文
  • Yanhui Ma,
  • Lusheng Li,
  • Chengke Wang,
  • Shuo Li,
  • Bin Fu,
  • Zhiqiong Wang

摘要

Additive Manufacturing (AM), particularly Laser Powder Bed Fusion (L-PBF), faces critical challenges in defect detection due to the scarcity of high-quality training data and severe class imbalance, which significantly degrade the accuracy of deep learning models. To address these issues, this study proposes a novel data augmentation framework combining geometric transformations with an enhanced Variational Autoencoder-Generative Adversarial Network (VAE-GAN). Traditional augmentation techniques (rotation, scaling, flipping) are first applied to alleviate sample imbalance, followed by the improved VAE-GAN to synthesize high-fidelity defect images, thereby enriching dataset diversity. Experimental results on an L-PBF defect dataset demonstrate significant improvements in detection performance: Specifically, comparative experiments were conducted to evaluate the performance of YOLOv4, YOLOv7, YOLOv8, SSD, and Faster R-CNN on defect detection tasks before and after data augmentation. The results demonstrated significant mAP improvements across all models, with YOLOv4 achieving the most substantial enhancement (+ 19.94%, from 65.18% to 85.12%) despite its lower baseline performance. Faster R-CNN attained the highest post-augmentation mAP (87.69%), representing the best overall performance. YOLOv8 exhibited an optimal balance between real-time processing and accuracy (67.07%→84.62%, + 17.57%), approaching Faster R-CNN’s performance level. While SSD showed the smallest improvement (+ 12.40%), it maintained a relatively high baseline mAP (86.39%). These results validate the effectiveness of the proposed method in overcoming data scarcity and improving defect detection accuracy in AM.