Parallel Processing of Morphological Pattern Spectrum for a Massive-Parallel Memory-Embedded SIMD Matrix Processor MX-1

The use of digital images has become quite widespread in legal, medical, and private contexts. However, anyone can easily edit or manipulate any digital image on a computer. In addition, the digital image are familiar to us with the population smartphone or mobile device. The morphological pattern spectrum is a mathematical method to describe size distributions of structuring element in an image. This algorithm is possible to execute effective image processing and has a lot of repeated calculation. Thus, conventional embedded processors low-throughput per power consumption for morphological pattern spectrum processing. A Massive-Parallel Memory-Embedded SIMD (Single Instruction Multiple Data) Matrix Processor (MX-1), which is based on C language processing, has been developed for effective parallel processing to satisfy with required specification on the digital convergence. The MX-1 consists of two SRAMs and 2-bit processing elements. The MX-1 can execute fast morphological pattern spectrum processing by parallel operation. Thus, throughput per power consumption of MX-1 is up to about 22 times than conventional parallel processor. And throughput per CMOS process-based power consumption of MX-1 is up to about 22 times than conventional parallel processor. The MX-1 performance for morphological pattern spectrum processing becomes effective solution for mobile devices on the digital convergence era.