Perceptually based metrics for the evaluation of textural image retrieval methods

Texture is widely used in CBIR, and there have been a number of studies over the years to establish which features are perceptually significant. However it is still difficult to retrieve reliably images that the human user would agree are "similar". This paper reviews a range of computational methods, and compares their performance in classifying and retrieving images from the Brodatz set. Their performance is then related to the combined ranking of "similar" images from the same dataset, obtained from experiments where human volunteers were asked to identify which images were most like each of the Brodatz images. The full set of 112 images was used. We conclude that no one method consistently returns retrievals which the human user would agree were similar across the full range of textures, but that statistical methods appear to perform better overall. We propose a subset of the Brodatz images for comparison of retrieval methods, based on the correlation between individual rankings.