Institute of Measurement and Automatic Control Studies
Student Research Projects and Theses

Student Research Projects and Theses

© Foto: Bodo Kremmin / LUH
Foto: Bodo Kremmin / LUH

Here you will find the current topics that are offered for student work (study, bachelor, master and / or diploma theses) at the IMR. For more detailed information on the individual topics, please speak to the respective contact person.

Production Measurement and Testing Technology


Industrial and Medical Imaging

  • Machine Learning Based Image Registration with Fully Convolution Networks

     

    Motivation & Aim of the Project

    The aim of this work is the further development of machine learning-based image registration. For this, the existing implementation z. B. can be expanded and improved by a fully convolutional implementation. Additional network architectures such as Pyramid Networks or Spatial Transformer Networks are to be implemented and examined. The background is that, as with a panorama photo, several microscope images are combined to form an overall image in order to be able to evaluate and characterize a larger area. It is essential that the assembly errors are as small as possible.

    In this project, aspects of image processing, surface measurement technology and machine learning are important. Basic knowledge in these areas and programming (with Python) in general is recommended.


    Contact Person


Control Technology


Acoustics

Supervision of External Project

Requirements

The topic of the student thesis must match the current research focus of the institute (control engineering, acoustics, robotics, optical measurement technology, image processing) and meet sufficient scientific requirements.


For the supervision of a master’s thesis, we expect that you have previously written a very good student thesis at the IMR or that you have successfully worked as a research assistant at the IMR.


For the supervision of a bachelor thesis / student research thesis, we expect that you can show good to very good results in the IMR lectures and / or have successfully worked as a research assistant at the IMR.