Image systems are systems for processing, generating, and transmitting digital images. Typical examples are medical image processing, video games, and video compression. The heterogeneity of the image systems is twofold: Firstly, the computation is spread over several components within a system. Secondly, there are numerous heterogeneous sets of architecture on which different image applications are executed. Both types of heterogeneity lead to interesting topics that are examined in the program’s research training group: novel hardware architectures for image systems, methods and tools for the development of heterogeneous image systems, and applications and algorithms for heterogeneous image systems.
In area A of the program, novel dedicated hardware for capturing and encoding image and video data is explored and developed. In area B, methods and tools for the efficient development and testing of applications on different platforms, as well as practical techniques for self-configuration depending on the underlying hardware are all examined. In area C, four typical image applications are selected that can include dedicated hardware in very different ways, or that should be adapted for a wide range of target architectures.
The core of the research training group is to capture the entire image system’s spectrum from hardware to software. For this reason, knowledge across the different fields of hardware and software, methodology and tool design, and also development of algorithms is essential.
Cooperation partnerse.solutions
Siemens
| |