A Study on the Analysis of Multiple Illuminants in a Natural Image

[[abstract]] The impact of environmental light source leads into the object show a different color in the process of capture. That can not be faithful performance of the real color. So we can not know the object of image and the real object whether the same. Therefore, if we want to present true color, we must control environmental light source. That can spend substantial costs to construct standards environment space of capture before shooting, or through the color technology after shooting to estimate environment light source to further restore color of objects. This study aimed to develop a method of analysis of multiple illuminants. First of all, we simulated spectral reflectance of objects that used by the proportion of color. Secondly, two methods of single illuminant estimation were derived, including gamut extreme light estimation and invariant color light estimation. We used the natural image via color space conversion to estimate single illuminant by calculating distance form the reference color. Finally, in the analysis of multiple illuminants, we used image segmentation to understand it changing circumstances then making out the main illuminant in the image. We hope through experimental results improve the accuracy of image display. These results will be applied to color amendments of synthesis image, digital archives, medical imaging and other digital image post-production process to make it be more natural and more coordinative.