Color Image Retrieval Based on Non-Parametric Statistical Tests of Hypothesis

A novel method for color image retrieval, based on statistical non-parametric tests such as twosample\udWald Test for equality of variance and Man-Whitney U test, is proposed in this paper.\udThe proposed method tests the deviation, i.e. distance in terms of variance between the query\udand target images; if the images pass the test, then it is proceeded to test the spectrum of\udenergy, i.e. distance between the mean values of the two images; otherwise, the test is dropped.\udIf the query and target images pass the tests then it is inferred that the two images belong to the\udsame class, i.e. both the images are same; otherwise, it is assumed that the images belong to\uddifferent classes, i.e. both images are different. The proposed method is robust for scaling and\udrotation, since it adjusts itself and treats either the query image or the target image is the\udsample of other. \u