Image processing algorithm for flow analysis through colour change

Bachelor thesis or Master thesis

Continuous fiber-reinforced profiles are being used more and more in the construction of solar panels, wind turbines and other infrastructure projects. Manufactured in a pultrusion process, inexpensive lightweight profiles are produced in a continuous process. The fibers are impregnated with the plastic in the form of rovings (fiber bundles) during the process and formed directly into the finished profile, similar to extrusion.

© IKV

In order to further optimise the analytical design of injection boxes, a new methodology is being developed to determine the resin residence time distribution. For this purpose, plastics are dyed and the color is changed impulsively during the process. The resulting color image in the cross-section of the profile and its change is evaluated. The aim of the work is the development of a Python (or Matlab) script. This has the task of calculating the elapsed time until the color change for individual pixels/areas from a series of images (profile cross-sections, see image on the left) and drawing conclusions about currents. The complete change of the colour in a profile area represents the dwell time of the color and is relevant for the further optimisation of the impregnation and thus the productivity of the pultrusion process.

Your task

Your profile

• Technical or scientific degree (mechanical engineering, materials science, CES)
• Interest in practical work in the technical center
• Affinity for technology, creativity and willingness to learn
• Independent, conscientious and team-oriented way of working