facebook tracking

Master Thesis - Experimenting with image signal processing for machine vision

Scroll to content


Producing images from digital cameras starts with capturing light from a sensor, which is turned into human or machine readable images by various processing techniques. These techniques greatly impact the performance of algorithms using this image as input. Therefore, a thorough analysis is required to define a pipeline that would produce optimal image output from raw image sensor data. This pipeline is called Image Signal Processor. 

Image Signal Processing, aimed for human vision, is over-represented in research and often development in machine vision is motivated by these studies. However, the challenge is not the same and it is important to measure and understand the difference.

When developing such an application, one of the major challenges is to benchmark the image quality. This is defined as a metric which is used as a base for comparing different versions, covering as many use cases as possible

Project Description 

 In this thesis project, you will focus on:

  • Defining a quality metric using either some traditional image processing technique (e.g. image matching based on extracted features) or a neural network. The accuracy would would act as the base of comparison.
  • Developing a good understanding of cameras and image processing. This includes techniques such as debayering, color correction, tone mapping, gamma correction, etc. 
  • Experiment with building different Image Signal Processors, measuring traditional Image Quality parameters impact on machine vision accuracy. Examples of Image Quality parameters include resolution, color accuracy, bit depth and dynamic range.
  • Drawing a conclusion on what affects the overall performance and how these effects can be explained.


We are looking for a student with an interest in image processing for computer vision. The following skills would be highly valuable:

  • Python/C++/Matlab programming
  • Image Processing
  • Reading scientific papers
  • Handling large datasets
  • Machine learning is a plus

Further information

Final application date: 20 November 2021.

Planned start: January 2022, with some flexibility.

Duration: 30 ECTS

Please send in individual applications with CV, motivational letter and grade transcripts. If you wish to partner with someone, simply note that in your application. 

For questions regarding the project, please contact:

  • Attila Door: attila.door@zenseact.com
  • Sebastian Franzén: sebastian.franzen@zenseact.com
  • Valter Hesselmark: valter.hesselmark@zenseact.com


Additional information

  • Remote status

    Flexible remote

Or, know someone who would be a perfect fit? Let them know!

Gothenburg, Sweden

Lindholmspiren 2
417 56 Gothenburg, Sweden Directions View page


Career site by Teamtailor