A Quantitative Assessment Method of Image Matching Methods Based on Cumulative Prospect Theory

Le résumé fourni par la source

Recent years, various image matching methods spring up so that it becomes more and more difficult for decision makers to choose. In this paper, a quantitative assessment method of image matching methods based on cumulative prospect theory is proposed. In the method, the cumulative prospect theory is introduced to deal with the multi outcomes of each measure that assesses matching results, and there are two novelties. First, before the outcomes of each measure are taken into value function, they are normalized depending on the accepted threshold as well as the desired direction of decision makers. Second, in the weighting function the probability distribution function of the outcomes of each measure is calculated using the histogram without any assumption of functional form. Experiment using real matching results obtained by three different methods demonstrated that the assessments by the proposed method are all consistence to the early researches, presenting great effectiveness.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.