チャーノフの顔形グラフの顔形要素に対する個人誤差の評価

概要

論文の詳細を見る
The purpose of this study is to investigate the following discussions on the face graph that could not be described accurately in the previous paper^<1)>; (1) How would the contribution of variation of features be affected by the visual reaction of different panel members? (2) How would the reaction error within an individual be unavoidable when they were set the same face graphs at random? The experiments of the panel members' visual reaction and the analysis by ANOVA and contingency table were carried out by assigning to L_<16>(2^<15>) orthogonal array table. The results are as follows. (1) The effective features to visual reaction are slant of eyebrows, curvature of mouth, slant of eyes and interaction between slant of eyebrows and curvature of mouth in sequence. However, the contribution of its interaction is very small. In consequence of the experiments, the contribution of the features in the face graph can be reliable, as the results obtained are as same as those in the previous paper. (2) The individual reaction error is smaller than the reaction error between panel members when they were set the same face graphs twice at random. And the error variance of the panel members was nearly as same as that in the previous discussion. Among different panel members, the error variance was stable, though the panel members were restricted only to our university students.
明治大学の論文