|Event Type: Lecture and Computer Exercise|
|Module Representative: Prof. Dr.-Ing. Georg Schmitz|
|Lecturer: Dr.-Ing. Stefanie Dencks|
|APPOINTMENTS IN SUMMER SEMESTER|
|Start: Friday, 04/08/2022|
|Lecture: Friday, 08:15 AM - 09.45 AM, ID 04/445|
|Exercise: Tuesday, 10:15 - 11.45 AM, CIP-Pool 1|
|Date by arrangement with the lecturer.|
|Type of Exam:||oral|
After successful completion of the module, students have knowledge of multidimensional digital signal processing. They know and understand the acquisition of multidimensional image data of the most important diagnostic imaging methods, can model these and derive consequences for their processing. Students can classify the different steps of image processing into abstract task categories (e.g. filtering, segmentation, classification), know selected procedures in detail, and can explain and apply them. The students are able to analyse a given image processing task and to develop and algorithmically implement a suitable solution. The methods are taught using medical image data as an example, but students can also transfer the methods to other application areas. Exercises in small groups on computers enable the students to apply the acquired knowledge in a small team, and to explain and discuss their solutions.
Basic principles and specific methods of image processing are introduced, which are particularly applied to medical image data. However, several of these methods are also applied in other application areas, e.g. in industrial image processing.
In the first section, both the reception by the human visual system is outlined, and the students are familiarized with the definitions and basic principles of image processing as well (discretization, sampling theorem, global parameters of images). The second section imparts knowledge of the most important operations in the image domain (histogram modulation, filtering, morphological operations, geometric operations, distance transform, ...). The third section comprises methods of information extraction (segmentation algorithms, texture analysis, description of shape). The fourth section focuses on classification and various methods of machine learning (e.g. support vector machines, deep learning). The topic of the fifth section is image restauration. Additionally, an overview of image registration and 3D-visualization is given.
Knowledge of system theory, Fourier transform, and signal processing equivalent to the level of Bachelor in Electrical Engineering and Information Technology are a prerequisite. Basic programming skills in Matlab are advantageous.
This course is organized via Moodle. The necessary information will be given in the first lecture.