Image registration is a very important operation in multichannel image applications to correct for the frame-to-frame displacement that occurs during the image acquisition process. In order to register a pair of images, an image registration algorithm often employs some similarity criterion between the image functions, testing different displacement vectors to find an extremum of this similarity measure. A good similarity criterion should be robust with respect to the diverse noise environments encountered in multichannel image applications. Misregistration often leads to invalid results in later image processing stages. In this paper, a new similarity measure based on the number of coincident bits in multichannel images is presented. The similarity criterion incorporated in the image registration algorithm uses a coincident bit counting (CBC) method to obtain the number of matching bits between the frames of interest. The CBC method not only performs favorably compared with traditional techniques, but also renders simpler implementation in conventional computing machines. An image registration algorithm which incorporates the CBC criterion is proposed to determine the translational motion among sequences of images. The analysis of the errors caused by noise, misregistration, and a combination of these two is also included. Some experimental studies using low-contrast coronary images from a digital angiographic sequence have been performed. The results compare favorably with those obtained by using other nonparametric methods. Applications for this algorithm includes digital angiography and mammography.
