An automated design flow for image processing filter in embedded systems

An M language is being used for algorithm development and verification in embedded systems design recently. The M language processes the whole array simultaneously and it is expected that the design based on M can be implemented faster and the same or better quality. This paper presents a case study of a filter design for image processing using the M language based design flow. In this design flow, Prewitt filter were easily designed by a little coefficient change in a Sobel filter. The results of the case study demonstrated the high efficiency of the M language based design flow as compared to traditional functional modeling and HDL hand coding that it reduced 36% of the total man-hours.