A Composite Illumination Invariant Color Feature and Its Application to Partial Image Matching

Abstract

SUMMARYIn camera-based object recognition and classification, surface color is one of the most important characteristics. However, apparent object color may differ significantly according to the illumination and surface conditions. Such a variation can be an obstacle in utilizing color features. Geusebroek et al.'s color invariants can be a powerful tool for characterizing the object color regardless of illumination and surface conditions.In this work, we analyze the estimation process of the color invariants from RGB images, and propose a novel invariant feature of color based on the elementary invariants to meet the circular continuity residing in the mapping between colors and their invariants. Experiments show that the use of the proposed invariant in combination with luminance, contributes to improve the retrieval performances of partial object image matching under varying illumination conditions.

Person following control for a mobile robot based on color invariance corresponding to varying illumination

“…On the other hand, C shown in Figure . 3(d) shows a small change in the whole image, which obscures the characteristics of the object from the image converted by H. This suggests that the use of C together may reduce the recognition of the target, so it is effective to use H proactively. However, there is a problem that H cannot distinguish its opposite color for a given color, and Kobayashi et al [14] have made the opposite color correspond to a different value by dividing the value of H based on C. In this paper, the following invariance 𝐻′ is obtained by using the invariant 𝐶 0 = 𝐶 ≈ −0.01 as the standard in the same way as [14].…”