Decoding of Color Coded Structured Light Using a New Color Feature

In this paper, color coded structured light based on pseudo-random sequence is adopted, and a new decoding method is proposed. It includes three steps. First, the stripe centers are extracted through the use of peak detection. Second, the stripe colors are identified using k-means clustering algorithm on a new color feature with high discriminating power. This feature is related to the spectral sensitivity of red, green and blue sensors, the direction of the illumination source, the normal and albedo of the surface, and the spectral power distribution of the incident light, but it is insensitive to the intensity. Finally, a four-step matching method is presented in sequence matching, which is aim to match the subsequences with the length less than the window size. The experimental results show that our proposed decoding method has advantage of high accuracy.