アラカワ カオル   Arakawa Kaoru
  荒川 薫
   所属   明治大学  総合数理学部
   職種   専任教授
言語種別 英語
発行・発表の年月 2003/12
形態種別 国際会議議事録
標題 Extended Component-Separating Nonlinear Digital Filter with Optimization
and Its Application to Face Image Beautification
執筆形態 共著(筆頭者)
掲載誌名 Proc. IEEE ISPACS2003
巻・号・頁 pp.139-144
著者・共著者 Kaoru Arakawa and Hiroo Naito
概要 Authors proposed before a nonlinear digital filter named as a component-separating nonlinear digital filter (CS filter for short) in order to reduce smallamplitude random noise on signals which contain both continuous changes and abrupt ones. However, this filter has a drawback that it cannot suppress low-frequency components in noise adequately.
In this paper, an extended version of the CS filter is proposed. This filter can suppress the low-frequency
noise adequately as well as the high-frequency one.
Moreover, a method for its optimum design is also
proposed. As to the application, this filter is shown to be
quite effective for face image processing to make the skin
look smooth and beautified.