A distributed memory architecture for morphological image processing

With the popularity of object based coding methods, mathematical morphology has become a popular technique for image segmentation. Morphological filters are used in a pre-processing step to improve the chances of obtaining meaningful segmentation results. However, the computational cost of these filters limits their practical use. With the increasing demand for real time implementation, there is need for reducing the processing time of the operations. In this paper, an efficient architecture is proposed for performing morphological operations by utilizing the inherent pipelining that exists in these operations and the inherent parallelism in the incoming data. It is a distributed memory architecture with four processing elements operating under the control of a master processor. The details of the architecture along with the simulation and synthesis results are provided.