A flexible, component-based software framework for signal and image processing algorithms

Applications of computer vision and signal processing often share ideas and algorithms. Thus the modules required for implementation show a high degree of similarity, so that reuse plays an important role. Unfortunately, scientists are rarely familiar with the powerful concepts that the software engineering community provides in order to develop reusable software nor do they have the appropriate experience to apply these techniques right. For this reason, we present fundamental design issues of a component-based software framework, which guides and supports the academics in writing reusable software components and therefore, improves the efficiency of their work. Our approach relieves the user of writing auxiliary code such as configuration and help dialog but instead supports them in finding solutions to problems of their specific domain.