Deep learning-based phase measuring deflectometry for one-shot measurement and inspection of specular free-form surfaces

Manh The Nguyen; Young-Sik Ghim; Hyug-Gyo Rhee

doi:10.1117/12.3025140

30 September 2024 Deep learning-based phase measuring deflectometry for one-shot measurement and inspection of specular free-form surfaces

Manh The Nguyen, Young-Sik Ghim, Hyug-Gyo Rhee

Author Affiliations +

Proceedings Volume 13135, Interferometry and Structured Light 2024; 1313503 (2024) https://doi.org/10.1117/12.3025140
Event: Optical Engineering + Applications, 2024, San Diego, California, United States

Abstract

Deflectometry is a non-contact optical technique used for measuring specular free-form surfaces by projecting structured light using a display screen (LCD screen). While one-shot deflectometry has been extensively researched to develop reliable measurement methods, it is very challenging to measure and inspect complex surfaces using one-shot with low reflectivity because of obtaining phase information from a single, poor-quality complex pattern remains difficult. To address this, we propose a novel single-shot deflectometry approach that utilizes deep learning (DL) to measure complex surfaces accurately in a single shot. We employ DYnet++, a deep learning network model capable of retrieving phase information from single composite patterns. By comparing the results with the 16-step phase-shifting (PS) method, we validate the feasibility and effectiveness of our deep learning-based single-shot deflectometry, which offers potential applications in various industries.

Citation Download Citation

Manh The Nguyen, Young-Sik Ghim, and Hyug-Gyo Rhee "Deep learning-based phase measuring deflectometry for one-shot measurement and inspection of specular free-form surfaces", Proc. SPIE 13135, Interferometry and Structured Light 2024, 1313503 (30 September 2024); https://doi.org/10.1117/12.3025140

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