2値画像エントロピー値と言語表現による感性量との相関性

概要

論文の詳細を見る
The concept of entropy which was introduced as a thermodynamic property has been extended to the economics theory, information theory, and others. Even though "entropy" is used in many fields of science, it is not easy to understand the concept of entropy, as well as describing it by words/terms. We assume that one of the causes of such difficulty is the gap between psychological quantity caused by words/terms used in explaining the concept of entropy and actual entropy value. In this report, we investigated the correlation between the entropy and the psychological quantity obtained by showing binary dot pattern images to participants, and examined the correlation between the entropies calculated by several different algorithms and the psychological quantity by three verbal expressions. We have found that the entropy calculated by means of measuring the length between each black pixel has highest correlation to psychological quantity.