Non-Iterative ICA for Detecting Motion in Image Sequences

Abstract

The ICA (Independent Component Analysis) technique can be used to detect difference in a sequence of images. However, most implementations of this algorithm are highly computation intensive. In this chapter, non-iterative algorithms for ICA are introduced and a non-iterative algebraic ICA is selected to detect motion in image sequences. Simulation results in MATLAB are compared with direct image subtraction and FPGA (Field Programmable Gate Array) implementation to speed up the computation of this noniterative algebraic ICA. A system level approach to the design of an FPGA prototype is explained. FPGA allows short development time and enables verification of algorithms in hardware at low cost.