TEEP-RCNN：基于改进卷积块注意力机制的 Faster R-CNN 钢材表面缺陷检测中的纹理增强边缘感知方法
TEEP-RCNN: Texture-Enhanced Edge-aware Perception for Steel Surface Defect Detection via Improved Convolutional Block Attention in Faster R-CNN
浏览论文内容
中文总结 AI 辅助
针对钢材表面缺陷检测中纹理差异细微和类别不平衡问题，提出基于改进CBAM的Faster R-CNN两阶段检测器TEEP-RCNN，通过注意力增强和推理优化，在NEU-DET上以10个训练周期达到73.3% mAP@50，媲美百周期训练的YOLOv11m。
中文摘要 AI 辅助
钢材表面缺陷检测对于自动化工业质量控制至关重要，但由于类间细微的纹理差异和显著的类别不平衡，该任务仍具挑战性。我们提出了 TEEP-RCNN（纹理增强边缘感知区域卷积神经网络），这是一种基于 Faster R-CNN 构建的两阶段检测器，采用特征金字塔网络骨干和改良的卷积块注意力模块（CBAM）。我们的 CBAM 在通道注意力多层感知机中加入了 dropout 正则化，并在空间注意力分支上应用批归一化，从而减少协同适应并稳定门控逻辑值。训练采用带余弦退火预热的分层学习率策略，将预训练的 ResNet-101 骨干和检测头分别设置不同的更新率。在推理阶段，预测结果通过测试时增强与加权框融合（WBF）进行细化，提高了对细长及边界邻近缺陷的定位稳定性。在包含六类缺陷的 NEU-DET 基准上，TEEP-RCNN 在单个 GPU 上仅训练 10 个周期即达到 73.3% 的 mAP@50 和 37.9% 的 mAP@50-95，与 YOLOv11m（76.2% mAP@50，100 个周期）相比具有竞争力，并在 COCO 指标下于轧入氧化皮类别上优于后者。逐类分析表明，空间注意力分支对细长纹理缺陷（如斑块和划痕）最为有效，而由于裂纹具有分布式的非局部纹理结构，其检测在两种范式下仍是未解决的挑战。
英文摘要
Steel surface defect detection is critical for automated industrial quality control but remains challenging due to subtle inter-class texture differences and pronounced class imbalance. We introduce TEEP-RCNN (Texture-Enhanced Edge-aware Perception Region-based CNN), a two-stage detector built on Faster R-CNN with a Feature Pyramid Network backbone and an improved Convolutional Block Attention Module (CBAM). Our CBAM adds dropout regularization in the channel attention MLP and batch normalization on the spatial attention branch, reducing co-adaptation and stabilizing gating logits. Training uses a differential learning rate protocol with cosine annealing warm-up, separating update rates for the pre-trained ResNet-101 backbone and the detection head. At inference, predictions are refined via Test-Time Augmentation fused with Weighted Box Fusion (WBF), improving localization stability on elongated and boundary-adjacent defects. On the NEU-DET benchmark across six defect categories, TEEP-RCNN achieves 73.3\% mAP@50 and 37.9\% mAP@50-95 in only 10 training epochs on a single GPU, competitive with YOLOv11m (76.2\% mAP@50, 100 epochs) while outperforming it on the rolled-in-scale category under the COCO metric. Per-class analysis shows the spatial attention branch is most effective on elongated texture defects such as patches and scratches, while crazing remains an open challenge across both paradigms due to its distributed non-local texture structure.
