Correlation of Binarized Images

Abstract

Correlator technique has been used in pattern recognition applications. We introduce a correlation performance measure, peak to background ratio (PBR), defined as the ratio of the peak intensity to the average background variation in a single correlation output. Using this performance measure, we analyze the binary correlator and show that the binarization of the images can result in better performance. We consider a real-time implementation of the binary correlator, using VLSI technology. The analytical results are supported by computer simulations.