Application of image processing and machine learning for classification of laser-induced damage morphology

Linas Smalakys; Erikas Švažas; Robertas Grigutis; Andrius Melninkaitis

doi:10.1117/12.2500335

16 November 2018 Application of image processing and machine learning for classification of laser-induced damage morphology

Linas Smalakys, Erikas Švažas, Robertas Grigutis, Andrius Melninkaitis

Author Affiliations +

Proceedings Volume 10805, Laser-Induced Damage in Optical Materials 2018: 50th Anniversary Conference; 108052B (2018) https://doi.org/10.1117/12.2500335
Event: SPIE Laser Damage, 2018, Boulder, Colorado, United States

Abstract

Interest in qualitative analysis of damage morphology of laser-induced damage test sites has increased in recent years. Such analysis can potentially provide valuable information about underlying damage mechanisms and can be used for separation of different damage modes. However, morphological analyses are currently performed manually and only on a few test sites at a time. In this work, a novel computational approach to the analysis of damaged test sites is presented. Image processing algorithms were applied to images of test sites in order to identify damaged test sites and extract features of damage morphology. Unsupervised machine learning was performed to automatically cluster damaged test sites. It was shown that ZrO₂ single layer’s laser-induced damage can be separated into well defined clusters. The clusters were grouped to distinct catastrophic and color-change modes. Characteristic damage curves of different damage modes were investigated to reveal different fatigue behavior.

Citation Download Citation

Linas Smalakys, Erikas Švažas, Robertas Grigutis, and Andrius Melninkaitis "Application of image processing and machine learning for classification of laser-induced damage morphology", Proc. SPIE 10805, Laser-Induced Damage in Optical Materials 2018: 50th Anniversary Conference, 108052B (16 November 2018); https://doi.org/10.1117/12.2500335

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available