Underpinning the development of modern machine vision systems is a new generation of DSPs designed to support parallel processing. This paper describes the general features of the Texas Instruments TMS320C40 and how these are useful for image processing systems. The paper shows how some of the features of the C40 processor can be used to easily provide a high performance engine when executing image processing algorithms. Moreover it shows how parallel processing systems for image processing can be built with the C40, which can handle the high bandwidth data transfer and processing speeds required. It also describes how an application was built to make use of these features in a parallel image processing system based on the VME bus, and how the power, flexibility and ease of use of the C40 made it a natural candidate for the implementation of the algorithm. (5 pages)