基于YOLOv7改进算法的硅橡胶绝缘电缆缺陷识别

打开文本图片集

中图分类号：TQ333.93 文献标志码：A文章编号：1001-5922（2026）1-0056-05

Abstract：Defect detection of silicone rubber insulated cables plays an importantrole in ensuring the qualityof power equipment and improving the stability of power production.Typical defects of insulated cables mainlyinclude cracks，film splits，rubber holes and glue overflow.At present，theresearch on defect recognition of silicone rubber insulated cables based on deep learning has such problems as low small-sample detection accuracy，limited model generalization abilityand excessive computational complexity.YOLOv7 was selected as the basic framework and improved by introducing the SIoU loss function and the non-maximum suppresson algorithm DIoU-NMS.In the experiment，a comparative analysis was carried out on the detectionoffour types of defects，and the results showed that the improved algorithm exhibited good detection performance for different defect types.

Key Words：silicone rubber insulation material;defect identification;machine vision; YOLOv7

硅橡胶绝缘电缆在生产过程中会产生胶洞、溢胶、膜裂等其他缺陷问题，这些缺陷会对绝缘性能产生不良影响，从而降低电力传输的稳定性[1]。（剩余7597字）