Color rolling suppression algorithm considering luminance and color constancy

摘要

.When a digital image sequence is sampled with a periodi-ncally emitting light source, the color rolling (CR) phenomenon occurs,nwhich is shown by periodical variations of color and luminance values.nIn conventional CR suppression (CRS) methods, color variation hasnbeen reduced by using auto white balancemethods. However, the CRnphenomenon still appears in the resulting image sequences due toninterfield illuminant intensity variation. In the proposed CRS method,nthe interfield luminance and color variations are simultaneously sup-npressed by estimating the illuminant change between the currentnand the target fields. In order to consider the object motions, a mo-ntion detection technique is used to estimate the luminance changesnthat occurred due to the CR phenomenon. Moreover, the illuminantncolor is estimated using the CR achromatic color distribution in thenchromaticity space which is founded on the periodicity of the CR phe-nnomenon. Based on the motion detection and the achromatic colorndetection techniques, the illuminant is estimated using the obtainedncolor components in a common area of both static and achromaticnregions. The experimental results demonstrate that our strategy ef-nficiently suppresses the CR phenomenon without being affected bynmoving objects and produces luminance and color constant imagensequences. © 2011 SPIE and IS&T. [DOI: 10.1117/1.3582130]