Influence of the Luminance L* during Color Segmentation in the L*a*b* Color Space

In this paper a study of the influence of luminance L* at the L* a* b* color space during color segmentation is presented. The behavior of different color image segmentation algorithms is studied, in particular, the techniques based on only the Euclidean metric of a* and b* and an adaptive color similarity function defined as a product of Gaussian functions in a modified HSI color space. For the evaluation purposes, synthetic images that accurately assess the performance of the color segmentation were particularly designed. The testing system can be used either to explore the behavior of a similarity function (or metric) in different color spaces or to explore different metrics (or similarity functions) in the same color space. From the obtained results it follows that the color parameters a* and b* are not independent on the luminance parameter L* as one may initially assume.