Influences of color spaces on printing defect classification

This paper aims at exploring how spectral variation of illumination influences the classification accuracy of print defects in different color spaces, when the color camera is unable to adapt to varying illumination. Spectral variation is achieved through simulated transformation of correlated color temperature of illumination of the test images. The percentile features are used to detect the possible defective areas on printing matter surfaces. Results are demonstrated that the original RGB color space is the best color space under varying illumination, the XYZ and CIELuv spaces being also very promising. An interesting observation is that features from some of the color spaces perform worse than features obtained from grey-level images.