Hypercomplex Correlation for Defect Detection

The normalized cross correlation (NCC) has been used extensively in machine vision for industrial inspection, but the traditional NCC suffers from false alarms for a complicated image that contains partial uniform regions. In this paper, new algorithms of hypercomplex correlation are proposed. We set up a color object model in hypercomplex domain, which provides integrated color characteristics of the object. The proposed hypercomplex correlation in color image can effectively alleviate false alarms in defect detection applications.