Parallel processing of image analysis tasks is an essential method to speed up image processing and helps to exploit the full capacity of distributed systems. However, writing parallel code is a difficult and time-consuming process and often leads to an architecture-dependent program that has to be re-implemented when changing the hardware. Therefore it is highly desirable to do the parallelization automatically.
Within this project we examine different concepts of using a multi-agent system for the automatic parallelization of image analysis tasks. The agents plan and control the parallel processing of an image analysis task specified by the user. At this, they exploit all common methods of parallel processing such as data-, task-parallelism and pipelining. To evaluate our concepts in praxis we implement an appropriate multi-agent system based on the Marmot Multi-Agent Library to automatically parallelizes HALCON image analysis programs and operators.