ABSTRACT
This chapter begins with a review of correlation theory. Locations of the large peaks in the correlation signal indicate relative shifts of the signals for potential alignment. The correlation of binary valued, solid shapes tends to produce a correlation signal with a flat-top response. The peak of the correlation is directly related to the location of the center of the dock in the original image. As long as the dock is centered in the mask, then the amount of shift required to align the mask to the dock in the image is the distance between the center of the frame and the location of the correlation peak. The minimum average correlation energy must be computed with the data in Fourier space and this is designated by the carets. The minimum average correlation energy filter builds filters that produce a very sharp correlation peak when correlated with a training input.
