欢迎访问机床与液压官方网站!

咨询热线:020-32385312 32385313 RSS EMAIL-ALERT
改进型快速分群算法在偏光板瑕疵检测中的应用
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:


The Application of Improved Rapid Clustering Algorithm in Polarizer Defect Detection
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    为了缩短LCD生产中偏光板瑕疵检测时间,以满足后续生产中动态切割的需求,开发即时偏光板瑕疵区域检测系统。该系统使用线性马达平台移送偏光板成品,以固定的扫描频率触发8k线型扫描摄影机拾取影像,同步处理部分已经拾取到的影像进行瑕疵区域检测。首先使用适应性门槛值进行瑕疵影像分割,再利用以密度为基础的改进型快速分群算法对瑕疵影像进行分群处理与位置范围标记,最后将影像处理与瑕疵标记程序以并行方式交由双核心CPU处理,实现即时瑕疵检测的功能。通过实验可知,检测平台在70 mm/s的移动速度下,单个CCD每秒可连续处理宽约57 mm偏光板影像的瑕疵检测与分群,满足了生产需求。

    Abstract:

    In order to shorten the test time of polarizer in LCD production, to meet the demand of dynamic cutting in subsequent production, the realtime polarizer defective region detection system was developed. In the system, a linear motor platform was used to transfer the finished polarizer product, it triggered an 8k linear scan camera to pick up images at a fixed scan frequency, the defective region detection of the pickedup image was done by the synchronization processing section. The defective images were cut with the adaptive thresholds, then the defective images were grouped and marked the position range, by using the improved rapid clustering algorithm based on density. Finally, the image process and the markup programs were processed in parallel by the dualcore CPU, to realize the function of immediate defect detection. The experiments show that, at the speed of 70 mm/s of the detection platform, a single CCD can continuously process the defect detection and clustering of polarizer images with a width of approximately 57 mm per second, meeting the production demand.

    参考文献
    相似文献
    引证文献
引用本文

任文杰,李磊.改进型快速分群算法在偏光板瑕疵检测中的应用[J].机床与液压,2019,47(16):93-98.
. The Application of Improved Rapid Clustering Algorithm in Polarizer Defect Detection[J]. Machine Tool & Hydraulics,2019,47(16):93-98

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2020-03-12
  • 出版日期: