Abstract: A method for the automatic detection of colorimetry of CMYK image files includes the calculation of differentiation metrics based on colorant relationships, saturation, and luminance. Summary statistics, such as the means and standard deviations of the differentiation metrics are used to describe a set of known image colorimetry classes. At least one classifying algorithm is used to assign a known colorimetry class to an image of unknown colorimetry. Preferably at least one of the classifying algorithms is a Bayesian Quadratic Classifier algorithm. Mahanalobis distances are used to find the most appropriate colorimetry class to assign to an image. A color image processor operative to automatically detect image colorimetry comprises a training image analyzer, a colorimetry detector and a class associator.