Psychophysical experiments on the PicHunter image retrieval system

Psychophysical experiments were conducted on PicHunter, a content-based ima ge retrieval (CBIR) experimental prototype with the following properties: ( 1) Based on a model of how users respond, it uses Bayes's rule to predict w hat target users want, given their actions. (2) It possesses an extremely s imple user interface. (3) It employs an entropy-based scheme to improve con vergence. (4) It introduces a paradigm for assessing the performance of CBI R systems. Experiments 1-3 studied human judgment of image similarity to ob tain data for the model. Experiment 4 studied the importance of using: (a) semantic information, (b) memory of earlier input and (c) relative and abso lute judgments of similarity. Experiment 5 tested an approach that we propo se for comparing performances of CBIR systems objectively. Finally, experim ent 6 evaluated the most informative display-updating scheme that is based on entropy minimization, and confirmed earlier simulation results. These ex periments represent one of the first attempts to quantify CBIR performance based on psychophysical studies, and they provide valuable data for improvi ng CBIR algorithms. Even though they were designed with PicHunter in mind, their results can be applied to any CBIR system and, more generally, to any system that involves judgment of image similarity by humans. (C) 2001 SPIE and IS&T.