PhD in High-Speed Hyperspectral and Super-Resolution Terahertz Imaging

Technical University of Denmark; Kongens Lyngby; On-site; DKK 350,000 - 420,000; Full time

Job summary

Technical University of Denmark (DTU) Electro invites applications for a PhD project focusing on high-speed hyperspectral terahertz imaging. You will work at the interface of ultrafast THz photonics, nonlinear detection, coherent imaging, and machine learning to push imaging speed, bandwidth, and resolution.

The project includes international collaboration and a potential 3–6 month research stay at Prof. Xi-Cheng Zhang’s group, University of Rochester.

Qualifications

• PhD in a relevant field (physics, electrical engineering, or closely related disciplines).
• Strong publication record and experience with ultrafast photonics or THz sensing.
• Experience applying machine learning to imaging or signal processing.
• Good collaboration skills and ability to work in international teams.

Responsibilities

• Develop and operate a broadband coherent THz imaging platform.
• Implement wide-field imaging based on THz-to-NIR upconversion and CMOS detection.
• Acquire and analyze time-domain hyperspectral THz data (>10 THz bandwidth).
• Apply CNNs and autoencoder architectures for super-resolution reconstruction and material identification.
• Collaborate across Denmark and international partners; publish results and present at conferences.

Skills

Machine learning

THz imaging

CNN

Autoencoder

Education

PhD in physics/electrical engineering or related

Tools

CMOS focal-plane array

Optical rectification

Laser-induced plasma THz sources

Job description

Are you interested in pushing terahertz (THz) imaging beyond today’s limits in speed, bandwidth, and spatial resolution? At DTU Electro, you will work at the interface of ultrafast THz photonics, nonlinear optical detection, coherent imaging, and machine learning to develop a new generation of high-speed hyperspectral THz imaging technology. The PhD project is part of the DFF-funded INSPIRING project (High-speed hyperspectral and super-resolution terahertz imaging). INSPIRING will combine broadband THz-to-near-infrared upconversion with CMOS-array detection to enable high-speed THz imaging. Your main role will be to establish the imaging platform and develop hyperspectral and machine-learning methods for super-resolution imaging and material identification. You will work in an international, interdisciplinary team at DTU Electro. The project includes a planned 3–6-month research stay in Prof. Xi-Cheng Zhang’s group at the University of Rochester, USA.

Responsibilities and qualifications

Your research will focus on high-speed coherent THz imaging with machine learning.

You will:

• Develop and operate a broadband coherent THz imaging platform using optical rectification and/or laser-induced plasma THz sources.
• Implement wide-field imaging based on THz-to-NIR upconversion and CMOS focal-plane-array detection.
• Acquire and analyze time-domain hyperspectral THz data with a target bandwidth above 10 THz.
• Apply machine-learning methods, including CNN and autoencoder-type architectures, for super-resolution reconstruction and material identification.
• Study semiconductors, polymers, multilayer structures, and industrial test samples, and collaborate with project partners in Denmark and abroad.
• Publish your results in peer-reviewed journals and present them at international conferences.

A letter motivating the application (cover letter) Curriculum vitae Grade transcripts and BSc/MSc diploma (in English) including official description of grading scale