Neural network-based generative optimization for athermal lens design

SeongHyun Jhun; YunYong Lee; Seung-Han Park

doi:10.1117/12.3037232

4 October 2024 Neural network-based generative optimization for athermal lens design

SeongHyun Jhun, YunYong Lee, Seung-Han Park

Author Affiliations +

Proceedings Volume PC13131, Current Developments in Lens Design and Optical Engineering XXV; PC131310D (2024) https://doi.org/10.1117/12.3037232
Event: Optical Engineering + Applications, 2024, San Diego, California, United States

Conference Poster

Abstract

A neural network based generative optimization algorithm was investigated for designing athermalized lens design. In particular, deep learning framework was developed by employing PyTorch and incorporated lens variable conversion techniques along with a differentiable ray tracing module. The framework, combining supervised optimization with unsupervised optimization, could generate diversified lens designs starting from reference lens system including aspheric surfaces. Our generative optimization algorithm could also be applied to the design of athermal lens systems that minimize thermal focus shift with temperature changes. In addition, using the developed algorithm and considering the first order thermal expansion coefficient of each lens, we were able to design an all-plastic athermal lens system composed of polycarbonate and polymethyl methacrylate materials. The RMS spot size averaged over all fields and Seidel aberration were minimized for thermally expanded lens systems at various temperatures. The developed framework is expected to help lens designers create optimal designs.

Citation Download Citation

SeongHyun Jhun, YunYong Lee, and Seung-Han Park "Neural network-based generative optimization for athermal lens design", Proc. SPIE PC13131, Current Developments in Lens Design and Optical Engineering XXV, PC131310D (4 October 2024); https://doi.org/10.1117/12.3037232

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