Other versions: Acrobat PostScript
Many models and metrics for image quality predict image discriminability, the visibility of the difference between a pair of images. We compare three such methods for their ability to predict the detectability of objects in natural backgrounds: a Cortex transform model with within-channel masking, a Contrast Sensitivity filter model, and digital image difference metrics. Each method was implemented with three different summation rules: the root mean square difference, Minkowski summation with a power of 4, and maximum difference. The Cortex model with a summation exponent of 4 performed best.
Keywords: vision models, image quality metrics, object recognition, target detection, contrast sensitivity, cortex transform.