Color indexing using wavelet-based salient points

Color is an important attribute for image matching and retrieval. Most of the attention from the research literature has been focused on color indexing techniques based on global color distributions. However, these global distributions have limited discriminating power because they are unable to capture local color information. We present a wavelet-based salient point extraction algorithm. We show that extracting the color information in the locations given by these points provides significantly improved retrieval results as compared to the global color feature approaches.