浙江电力2024，Vol.43Issue(4)：113-120,8.DOI:10.19585/j.zjdl.202404012
基于改进YOLOv8的电力场景通用缺陷检测模型
A general defect detection model for power scenarios using the improved YOLOv8
摘要
Abstract
The current centralized defect detection system faces challenges such as large data volume and poor real-time performance,underscoring the pressing need for distributed detection systems,with edge computing as a repre-sentative.In response,diverse module algorithms are devised based on the single-stage,lightweight detection model YOLOv8 to boost its accuracy in power scenarios.Firstly,the Mosaic data augmentation algorithm is improved,and a conflict relationship table is introduced to mitigate the damage to original image data information caused by tradi-tional data augmentation algorithms and enhance the diversity of image data.Subsequently,the Res2Net module is employed to replace the original Bottleneck module,reinforcing the model's multiscale perception while retaining its lightweight design.The adoption of the CIoU-NMS algorithm over the existing NMS(non-maximum suppression)algorithm improves the recall rate and precision of the detection model in the clustering and deduplication stage.Fi-nally,experiments on fourteen defects in power scenarios consistently demonstrate the proposed model's superior detection accuracy compared to the original model,accompanied by an accelerated detection speed in defect detec-tion power scenarios.关键词
深度学习/缺陷检测/数据增强/非极大值抑制/YOLOv8Key words
deep learning/defect detection/data augmentation/NMS/YOLOv8引用本文复制引用
韩睿,戴哲仁,蒋鹏,李晨,姜雄伟..基于改进YOLOv8的电力场景通用缺陷检测模型[J].浙江电力,2024,43(4):113-120,8.基金项目
国网浙江省电力有限公司科技项目(5211DS220003) （5211DS220003）
