Bachelor theses / Master theses / Diploma theses

The modelling of disperse two-phase flows in Computational Fluid Dynamics is the subject of a current research project of the Institute of Fluid Dynamics at the HZDR. To validate these numerical methods, image data were obtained from experiments on horizontal droplet flows of air and water (refer to Figure). For the comparison with the simulation the image data is to be automatically evaluated with regard to size and speed distributions. Existing program libraries (ImageJ/Fiji, SciKit-Image, Matlab Image Processing Toolbox) are to be used for this purpose.
To support this work we are searching for student assistant. The candidate will be responsible for the following tasks:

• literature study on noise filtering and edge detection techniques
• preprocessing of image data (background subtraction, normalisation, noise filtering)
• edge detection and determination of drop size / velocity
• binning and averaging to calculate distribution functions
• validation of the method by means of image data of particles of known size
• documentation of the results

Requirements of the respective university can be accounted for after consultation.

• interest in fluid mechanics, especially multiphase flows
• good command of a programming/scripting language (preferably Java, Python or Matlab)
• ideally experienced in the field of image processing

• duration: 4-6 months
• start date: as of now
• compensation for expenses