光学元件表面疵病快速检测分类方法研究

A Quick Method of Debris Detecting and Classifying for Oversize Optical Element

按ICF相关工程标准进行光学元件表面疵病快速检测是ICF系统中急需解决的问题。本文采用基于替换数组的标记方法和边界跟踪算法快速提取疵病的特征信息,然后利用大量疵病检测结果建立疵病特征信息库及疵病分类准则,提出了一种超大图像的自动快速检测分类方法。实验结果表明该方法能满足实时性需求,取得了较好分类检测效果。

The Quick inspection of surface debris which according to ICF engineering standard is an urgent problem expected to be solved . Basing on replacing array, a pixel label method and edge following algorithm are given to extract imperfections feature quickly. Then a debris information base is established and also a classification rule is brought up which using the result of inspection, the system gives a fast inspection and classification method which aims at processing oversize image. Experiment result shows it can meet the real-time needs and arrive at a good classification conclusion.