An evaluation of the suitability of fpgas for embedded vision systems

Recongurable hardware, in the form of Field Pro-grammable Gate Arrays (FPGAs), is becoming increasingly attractive for digital signal processing problems, including image processing and computer vision tasks. The ability to exploit the parallelism often found in these problems, as well as the ability to support different modes of operation on a single hardware substrate, gives these devices a particu-lar advantage over xed architecture devices such as serial CPUs and DSPs. Further, development times are substan-tially shorter than dedicated hardware in the form of Appli-cation Specic ICs (ASICs), and small changes to a design can be prototyped in a matter of hours. On the other hand, designing with FPGAs still requires expertise beyond that found in many vision labs today. This paper looks at the advantages and disadvantages of FPGA technology, its suitability for image processing and computer vision tasks, and attempts to suggest some direc-tions for the future. 1