Optical design with deep learning

Studien-/ Bachelor-/ Diplom- oder Masterarbeit,Hiwi-Stelle


M.Sc. Chia-Wei Chen


zu vergeben

Möglicher Beginn:

ab sofort

The Fraunhofer Institute of Optronics, System Technologies and Image Exploitation IOSB is the largest European research institution in the field of image acquisition, processing and analysis. In the Visual Inspection Systems (SPR) department, new approaches to automatic image evaluation and machine vision are pursued and implemented in systems for solving industrial visual inspection tasks.

Optical design is an optimization problem. The goal is to find the parameters which can achieve the best performance, including optical aberrations and mechanical tolerances. The design process is usually very time-consuming and challenging due to the non-linear properties and physical constraints. Nowadays, deep learning is widely used in many different applications, e.g., natural language processing, autonomous driving and computer vision. This technique can also be applied to optical design [1,2].

[1] https://doi.org/10.1364/OE.27.028279

[2] https://doi.org/10.1117/12.2313144


  • Literature review on state of the art for optical design based on deep learning 
  • Extend existing models and evaluate procedures and performance on suitable data sets 
  • Develop data processing and visualization tools

Students with a background in physics, computer science, mathematics, or other engineering majors. Prior knowledge in optical design and deep learning is welcome. Strong analytical and problem-solving skills, excellent communication, organizational and interpersonal skills are expected.

M. Sc. Chia-Wei Chen
Fraunhofer IOSB Visual Inspection Systems SPR
Fraunhoferstraße 1, 76131 Karlsruhe
Email: chia-wei.chen@iosb.fraunhofer.de
Tel.: 0721 6091-590