Skip to main content

Learning to Sense

The German Research Foundation (DFG) has selected L2S as one out of eight research units in Germany that conduct dedicated fundamental research on artificial intelligence along with an interdisciplinary partner field, in this case sensor system development. The project is situated at ZESS where it can build upon more than 30 years of experience in the field of fundamental and application-oriented, interdisciplinary, research. 

Website
lasers and stuff

Learning to Sense for Advanced Coherent THz Imaging Systems

Coherent imaging in the mm-wave and THz frequency ranges opens up a huge application range, providing innovative imaging and sensing capabilities in optically inaccessible situations, like inter alia remote sensing under arbitrary environmental conditions, autonomous vehicle vision systems resilient towards fog and rain, subsurface imaging for material analysis, quality control, and non-destructive testing, or security related imaging systems for applications such as hidden explosives detection at airport checkpoints. This project plans to use machine-learning based approaches to learn to cope with fundamental limitations of coherent imaging systems, and to train and validate their adequacy in the mm-wave and THz frequency ranges. The experimental realization will concentrate on sparse multiple-input multiple-output (MIMO) synthetic aperture radar (SAR) mm-wave and THz imaging approaches (from 300 GHz - THz), given their wide-ranging application relevance and intrinsic advantages.

 

This objective is guided by following interrelated goals:

  • Learning how physical knowledge containing network architectures can be used to develop adaptive synthetic image generation approaches that enhance the image quality and correct scene dependent interference artifacts for coherent 3D imaging.

  • Evaluating and understanding the robustness of machine-learning based segmentation of reconstructed 3D THz imaging data originating from sparse illumination and sensor arrangements, including differential imaging modes.

  • Assessing and learning if segmentation from raw sensory data can directly be attained, without an intermediate 3D image generation step by synthetic reconstruction.

  • Learning how a sensory task dependent system optimization can fundamentally maximize imaging and recognition capabilities, and at the same time minimize hardware and data acquisition effort.

Overview

  • Icon Kalender

    Duration
    01/12/2022 - 30/11/2026 (Running)

  • Icon Tag

    Research areas
    Sensorics, simulations, machine learning and computer vision

  • Icon Abzeichen Euro

    Funding
    DFG

All projects

 

Jointly Optimizing Sensor System Designs and Neural Networks

Vision

Sensor System Development

We develop and optimize the next generation of CMOS Sensors, THz imaging systems, and 3D microscopes taylored to specific automatic data analysis applications. 

Software Systems

Sensor System Simulation

In order to know how the recorded data changes as the design parameters of a sensor system are changed, we will develop faithful simulators of our three sensor modalities.

Open Book

Machine Learning

We will develop new approaches to jointly optimize for the sensor system design along with neural networks parameters. New network architectures need to handle the changing type of the sensor system's data and become optimal for specific applications. 

Collaborators

The L2S research unit is the center of an international network of researchers working on different aspects of sensorics, simulations, machine learning and computer vision. Below are our closest collaborators all of whom are DFG Mercator fellows of our projects and will therefore spend longer research visits with our research group.