A Flexible Architecture for Image Processing by

¥ image processing ¥ flexible architecture ¥ pattern matching, recognition ¥ segmentation, shrink,expand ¥ VLSI layout processing ¥ routing- VLSI impact on architectures- Tools and methods for architecture design and description- computer architecture- Map oriented data processing The paper describes an innovative computation resource concept which for a class of data processing problems is an alternative to the von Neumann ma-chine. The 'processor ' , called 'Map Oriented Machine ' (MOM), used for this concept is faster than a von Neumann-type computer, however, it is substan-tially less expensive than a fully parallel hardwarized implementation using full custom or semi custom circuits. Instead of a program store with a pro-gram sequencer a personalized hardware is used, and, to 'program ' this ma-chine CAD tools are used instead of conventional compilers. The MOM concept is a compromise between the purely sequential von Neumann con-cept (sequential control part and sequential data part) and fully parallel solu-tions (parallelized control part and parallelized data manipulation side) insofar, as the control part has been parallelized, the data manipulation side, however, still uses a universal sequential access organisation. b) State of the Art in this Field: There are several approaches on speeding up image processing using VLSI implementations. All of them use a certain number of uniform processing el-ements (PEs) which are connected for intercommu-nication purposes. The arrangements of these PEs can be linear arrays (pipelines) or square arrays. The number of PEs determines the degree in paralleliza-tion achieved. The image data is pulsed through the PEs, if the implementation is based on a systolic ar-ray concept, or the complete image is stored in a PE square array and the PEs perform the same compu-tations simultaneously based on a cellular array concept. The PEs are limited to a rather small range of operations or even they are able to perform just one single operation