针对当前工业场景焊缝自动化检测效率低和精度不足的问题，提出了一种针对激光焊缝缺陷的基于全卷积单阶段目标检测改进模型（weld detect fully convolutional one-stage object detection, WDFCOS）。该模型以原全卷积单阶段目标检测模型（fully convolutional one-stage object detection, FCOS）为基础，设计了轻量型骨干网络RelementNet，利用元素乘法的特性更加高效地提取图像特征；使用金字塔注意力网络结构加强原有的特征融合网络，组成双向特征融合网络；设计了一种新型的高效多尺度注意力上下文聚集模块，以减少特征融合过程中的信息衰减。通过在自建数据集上进行实验，采用WDFCOS模型平均精度可以达到82.6%，相比FCOS模型提高了6.9%；同时，在更适合工业场景的边缘设备上检测速率可以达到49.1帧／s，与基线模型相比，检测时间降低84.4%，实现了更加快速和精确的焊缝缺陷检测。

Aiming at the problems of low efficiency and insufficient accuracy of automatic welding seam inspection in industrial scenes, WDFCOS model for weld seam defects was proposed. Based on FCOS model, a lightweight backbone network, RelementNet, was designed to extract image features more efficiently. PAN(pyramid attention network) structure was used to strengthen the original feature fusion network and form a bidirectional feature fusion network. A new multi-scale attention CGM module was designed to reduce the information decay in the process of feature fusion. Through experiments on self-built data sets, AP(average precision) can reach 82.6% by WDFCOS model, which is 6.9% higher than FCOS model. At the same time, the detection rate can reach 49.1 frames per second on the edge equipment more suitable for industrial scenarios, and the detection time is reduced by 84.4% compared with the baseline model, achieving a faster and more accurate weld defect detection.