並列タイトル等Quantification of Blurred Images and its Application to Pattern Recognition
一般注記Human visual sense has two aspects in our feeling for blurred image, that is, one is the amount of blur depending on object size,t he other is the amount of blur independenot f the object size. In the formerf or example,w hen the image size becomesl arger, we feel smaller amount blur. The quantitative evaluation based on entropy for blurred images is proposed in this paper. The author calls this metric "variation entropy". This metric has two kinds of aspects that coincide with the human visual sense. The first is the absolute evaluation of blur, and the second is the relative evaluation of blur. The former can be quantified by "variation entropy for a unit boundary length (or L-type variation entropy:H L)", which is dependento n resolution,a nd the latterc an be quantifiedb y "variatione ntropy for a unit area(or A-type variatione ntropy:H A)", which is independento f resolution.T hese two metricsh ave complementaryp roperties.L ast, two variation entropies are applied to the standard kanji character database, and then the strong relation between variation entropy and accuracy of recognitionis discussed
一次資料へのリンクURLhttps://u-fukui.repo.nii.ac.jp/?action=repository_action_common_download&item_id=23050&item_no=1&attribute_id=22&file_no=1
連携機関・データベース国立情報学研究所 : 学術機関リポジトリデータベース(IRDB)(機関リポジトリ)