| Colour indexing is a method to index an image database by its colour content. The standard approach for colour indexing is based on colour histograms. For each colour comparison, the histograms of the colour distributions are calculated and compared by a histogram distance metric such as intersection to measure the similarity between the colour content of the images. A method for colour indexing is proposed that is based upon Nonparametric statistics. Nonparametrics compare the ordinal rankings of sample populations, and unlike Parametric statistics, Nonparametrics maintain their significance when the underlying populations are not Normally distributed.; Principal component analysis is performed to extract the three orthogonal axes of maximum dispersion for a given colour signature. These axes are then used to select Lipschitz embeddings to generate sets of scalars that combine all colour channel information. These scalar sets are compared against a ranked database of such scalars using the Moses test for variance. On the resulting top matches, the Wilcoxon test of central tendency is applied to yield the best overall match.; The method has been tested extensively on a number of image databases, and has been compared against eight standard histogram methods using four colour space transformations. The tests have shown the recognition performance of the Nonparametric method to be competitive with, and in certain cases superior to, the best histogram methods. The technique also shows a greater robustness to noise than all histogram methods, with a noise robustness comparable to that of the more expensive Variable Kernel Density method. |
