Thesis - Surface defect detection on 3D data

Inspection of surfaces for defects, such as scratches, fingerprints on an aluminum sheet and other cosmetic blemishes is a common request in the industry. As these features often are small and sometimes quite similar to the object inspected, detection of these can be quite difficult.

Inspection tasks can often be solved using classical image processing algorithms, but it can be challenging when the features are subtle and highly variable as surface defects often are. Deep Learning methods, however, have been shown to work well for 2D data for these types of inspections. It's quite common for our customers to use a camera that produces 3D data at the same time as some 2D data.

Deep learning methods can however be tricky in practice, as it is traditionally required to have large amounts of data. One option to get around this would be to use deep learning-based anomaly detection, which requires only a few defect-free samples.

How would one design an algorithm or deep learning algorithm for surface defect detection when the data available is 3D and 2D?

The master thesis work focuses on the following/example of research questions

• Can reliable surface defect detection be implemented with classical image processing algorithms?
• How can deep learning be used for surface defect detection?
• How well can a surface defect detection system be generalized to solve different types of tasks?

Prerequisites

This thesis will require knowledge in image processing, machine learning and basic programming (Python recommended). The thesis will provide an opportunity to solve challenging problems faced by the machine vision industry while working with real state-of-the-art hardware.

Contact

For more information about the position, contact:

Malcolm Vigren, Application Developer, malcolm.vigren@sick.se

or

Charlotte Axelsson, HR Manager, charlotte.axelsson@sick.se

Welcome with your application 14^th of October at the latest!

SICK is a world-leading supplier of sensors and sensor solutions for industrial applications. We are 11 000 employees in 50 countries and our headquarter is located in Freiburg, Germany. SICK in Linköping is an innovation center for Machine Vision and we are 80 committed employees with a big interest in image processing and visualization. For more than 35 years, our team at SICK Linköping has successfully developed and delivered software for technically leading products within the field of 2D and 3D vision, as well as system solutions for i.e. robot guidance and quality control.

At SICK in Linköping, we are very proud of being a healthy and attractive workplace. For many years, we have been elected as one of the best workplaces in Sweden according to the survey Great Place to Work, the latest award is from 2021. We work actively to reduce our climate footprint and we are active in various ways to contribute to the society and to increase diversity at our workplace.