Robust and automatic measurement of grinding-induced subsurface damage in optical glass K9 based on digital image processing

Yong Jie Zhao, Yun Hui Yan, Ke Chen Song, Hao Nan Li

Research output: Journal PublicationArticlepeer-review

4 Citations (Scopus)

Abstract

Optical glass K9 is a critical kind of optical materials, however experiments have indicated that the mechanical grinding of K9 easily led to subsurface damage (SSD). Although substantial SSD measurement methods have been suggested, the problems including the prior knowledge of SSD and slow measurement speed still impede the reported method applications. To this end, this paper has presented an image-process-based method that can identify and measure the grinding-induced SSD in K9 specimens. By performing grinding trials, the method has been found to be able to accurately (with biggest relative error of 3.13% in comparison with the manually measured results) and quickly (with the measurement speed of 1.68 s per image) measure SSD depths. Without any parameter presetting, the method enables automatic SSD measurements, allowing the users without SSD knowledge to be able to use the method. Moreover, the method has shown the good robustness to the input image size, illumination, tilted specimen placement, and material flaws. The method is therefore anticipated to be meaningful for the industrial manufacturing, design and application of optical glass.

Original languageEnglish
Pages (from-to)320-330
Number of pages11
JournalArchives of Civil and Mechanical Engineering
Volume18
Issue number1
DOIs
Publication statusPublished - Jan 2018
Externally publishedYes

Keywords

  • Grinding
  • Image processing
  • Measurement
  • Optical glass
  • Subsurface damage

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Mechanical Engineering

Fingerprint

Dive into the research topics of 'Robust and automatic measurement of grinding-induced subsurface damage in optical glass K9 based on digital image processing'. Together they form a unique fingerprint.

Cite this