Thesis: Synthetic images for machine vision and image processing* in Waldkirch bei SICK AG

Winter semester 2024/25 – limited to 3-6 months

Synthetic images for machine vision and image processing:

In the past decades, camera systems with increasingly sophisticated capabilities have become ubiquitous. Ranging from cameras embedded in our smartphones to camera arrays in self-driving cars and camera systems controlling industrial manufacturing process, these systems are revolutionizing industries and our everyday lives. At the heart of all these imaging systems lies the image acquisition process with the aim to achieve a “good” input image. Therefore, machine vision engineers often spend a lot of time and go through some empirical best practices to try and evaluate different geometrical and optical configurations of the imaging system. To speed up this tedious task, realistic synthetic images allow to evaluate, analyze, and optimize imaging setups without employing physical parts and sensors. To this end sensor-realistic synthetic images are needed, including noise and defects.

The goal in this thesis will be to setup a simulation to produce realistic synthetic images, starting with a virtual environment, simulating the imaging process, and then processing the images computationally.

YOUR TASKS:

YOUR PROFILE:

YOUR APPLICATION:

We put great emphasis on diversity, reject discrimination and do not think in categories such as gender, ethnicity, religion, disability, age or sexual identity.