基于Faster R-CNN的精密零件表面缺陷识别
Surface Defect Identification of Precision Parts Based on Faster R-CNN
张鹏飞1
作者信息
- 1. 中国航发贵州黎阳航空动力有限公司,贵阳 550014
- 折叠
摘要
常规的精密零件表面缺陷识别方法易受Conv/Pool压缩作用影响,缺陷识别指标不佳,因此提出一种基于Faster R-CNN的精密零件表面缺陷识别方法.实验结果表明,设计的精密零件表面缺陷识别方法的各项识别指标良好,有一定的应用价值,有利于提高精密零件使用性能和降低精密零件应用风险.
Abstract
The conventional surface defect recognition methods for precision parts are easily affected by Conv/Pool compression,resulting in poor defect recognition indicators.Therefore,a precision part surface defect recognition method based on Faster R-CNN is proposed.The experimental results show that the designed Faster R-CNN defect recognition method for the surface of precision parts has good recognition indicators and certain application value,which is conducive to improving the performance of precision parts and reducing the application risks of precision parts.
关键词
Faster/R-CNN/精密/零件/表面缺陷/识别Key words
faster R-CNN/precision/parts/surface defects/recognize引用本文复制引用
出版年
2024
NETL
NSTL
万方数据