A computational feedback model for generalized homogeneity evaluations of luminance distributions

Automated objective evaluations of the homogeneity of light distributions have proven to be truly challenging in the past. Although multiple promising procedures have been proposed in recent years, these have the drawback that they were designed for a specific lighting function or product and do not offer a generic approach. Recently we developed a functional model that mimics the first contrast processing stages of the human visual system. It allows a more generalized approach to the evaluation of homogeneity by calculating the perceived contrast of an arbitrary luminance distribution. While showing good correspondence to psychophysically measured results for the ModelFest data set, the model could not reproduce the results that are dependent on contour integration effects. Based on these findings we propose an enhancement of the model structure by implementing two additional filter stages in addition to feedback. The filter stages are designed to simulate the behavior of the bipole cells of the secondary visual cortex. This cortex region has been shown to contribute to contour detection mechanisms as well as textural segmentation. The implemented filters lead to an interaction of signals in adjacent areas within each stage. The feedback connections transport the processed information back to the preceding stages. This allows to reproduce behavior of the visual system which is known to enhance the detection of closed contours as well as the perception of illusionary contours.