Pipeline processor implementation of an object centred optimisation process for robust multiple object tracking

This paper presents the problems encountered when mapping prototype algorithms onto a hardware platform. The application area for the developed hardware is the enhancement of objects within complex scenes to aid object tracking. The aim of this paper is to evaluate the expected performance of the mapped algorithms using a state of the art commercial-off-the-shelf (COTS) pipeline processing board. In addition, the required number of COTS processing boards for real-time operation is calculated. A pipeline processor implementation is described throughout this paper along with the problems of identifying parallel processes, deriving the algorithm implementation structures, and calculating the resources required. Due to the fixed-point limitation of the hardware platform, some simplification was required during the mapping process. The impact of these changes upon the enhancement process is also examined