• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Smart City and Systems Engineering >Mechanical Parts Defect Detection Approach Based on Computer Vision Technology
【24h】

Mechanical Parts Defect Detection Approach Based on Computer Vision Technology

机译：基于计算机视觉技术的机械零件缺陷检测方法

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

The defect detection of mechanical parts in industry is studied in this article. The defects detection process is the key object of our research, to design the whole scheme for detecting system to proposes vision technology-based mechanical parts defects detection algorithm. It first adopts the improved threshold segmentation algorithm to delete noise and enhance the collected part image so that defect information in image is clearer. Then, it extracts defect characteristics to recognize and classify defects. Finally, main component analysis is used to reduce dimension on the extracted characteristics to reduce calculating complexity and establish characteristics parameter library as input parameters of BP classifier. BP neural network principle is applied to classify defects. Experiments show that this system can complete all jobs of defects detection in mechanical parts and satisfy requirement of practical industry detection.
机译：本文研究了工业中机械零件的缺陷检测。缺陷检测过程是我们研究的关键对象，设计用于检测系统的整个方案，提出基于视觉技术的机械部件缺陷检测算法。首先采用改进的阈值分割算法来删除噪声并增强收集的部分图像，以便图像中的缺陷信息更加清晰。然后，提取缺陷特征以识别和分类缺陷。最后，主要成分分析用于减少提取特性的尺寸，以减少计算复杂性并建立特征参数库作为BP分类器的输入参数。 BP神经网络原理应用于分类缺陷。实验表明，该系统可以完成机械零件中的所有缺陷检测的所有作业，并满足实际行业检测的要求。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号