Padua point interpolation and Lp-norm minimisation in colour-based image indexing and retrieval

Abstract

Colour has proven to be a very powerful feature for image indexing. Many examples of image retrieval systems based on colour or chromaticity histograms have been proposed, following on from the histogram intersection method of Swain and Ballard. Here the authors introduce a compact representation of the chromaticity histogram which is based on the Padua point interpolation technique. Specifically, the histogram is represented as a linear combination of Chebyshev polynomials. This bounds a certain maximum deviation, as opposed to a least-squares criterion used in previous work. With this in mind, the minimisation of different Lp norms and the L∞ norm of the error is compared.After presenting the Padua point image indexing and retrieval method, the authors compare its performance to the histogram intersection, the discrete cosine transform, and a dataset oriented method based on principal component analysis. The experiments show that the Padua points match and, in some cases, improve the performance of these methods. This is significant as the proposed method is not tuned (unlike the PCA approach to any dataset). Finally, the behaviour of the Padua point method is analysed in relation to the minimisation of different norms.