ISO 20462-2:2005 defines a standardized psychophysical experimental method for estimating photographic image quality through triplet comparison. It describes an initial category step for selecting test samples, followed by a triplet comparison step in which three samples are assessed simultaneously. The standard covers experimental procedures, sample combination planning, observer assessment categories, Scheffe's statistical method, conversion to just noticeable differences (JNDs), and comparisons with paired comparison techniques.
The method is designed to support large sample sets while providing precise scaling, reduced observer stress, and repeatable results. Informative annexes include sample combination calculations, standard portrait images, experimental performance data, Scheffe's method, and JND conversion guidance.
Application Scope: Suitable for researchers, imaging engineers, camera and display developers, image-processing specialists, and organizations conducting subjective visual assessments of photographic image quality.
