著者

上武 英朗 柳本 豪一 吉岡 理文 大松 繁

出版者

一般社団法人 システム制御情報学会

雑誌

システム制御情報学会 研究発表講演会講演論文集

巻号頁・発行日

vol.10, pp.119, 2010

---

現在,ソーシャルブックマークサービスが普及している.ソーシャルブックマークとは,ネット上で公開・共有されるWebページのブックマークであり,ユーザが有益な情報が得られる情報源として期待されている.本研究では,ソーシャルブックマークデータ(Webページ,ユーザ,タグ)をテンソルとして表現し,このテンソルを分解することでクラスタリングを行う.分解手法としてCP分解,Tucker分解を用いる.またクラスタリング結果から他手法との分類精度の比較を行う.

著者

姜 東植 大松 繁 吉岡 理文 小坂 利寿

出版者

一般社団法人 電気学会

雑誌

電気学会論文誌. C (ISSN:03854221)

巻号頁・発行日

vol.118, no.12, pp.1706-1711, 1998

---

In this paper, we propose a neuro-classification method of the new and used bills using time-series acoustic data. The technique used here is based on an extension of an adaptive digital filter (ADF) by Widrow and the error back-propagation method. Two-stage ADFs are used to detect the desired acoustic data of bill from noisy input data. In the first stage, superfluous signals are eliminated from input signals and in the next stage, only the desired acoustic data is detected from output signal of the two-stage ADFs. The output signal of two-stage ADFs is transformed into spectral data to produce an input pattern to a neural network (NN). The NN is used to discriminate the new and used bills. It is shown that the experimental result using two-stage ADFs is better than that obtained by using original observation data.

著者

田中 隆治 吉岡 理文

出版者

一般社団法人 電気学会

雑誌

電気学会論文誌C(電子・情報・システム部門誌) (ISSN:03854221)

巻号頁・発行日

vol.131, no.11, pp.1895-1900, 2011-11-01 (Released:2011-11-01)

参考文献数

7

---

In recent years, Augmented Reality (AR) becomes the focus of attention as a technology for obtaining some information. Most AR systems have used some markers to display any information. However, since markerless AR systems can be used intuitively, they are researched actively.In this paper, we study about the AR system based on the hand recognition as a markerless system. The reason why we used a hand is that we don't need to prepare or carry on any tool, and can easy to watch any information on a hand. Taehee Lee et al. introduced HandyAR which employ the hand recognition. In this method, there are some problems. At first, when the method learns skin model, it needs many learning data. This is because it uses RGB color model. Secondly, it is difficult to get hand because hand images which are extracted by the skin model have noise. Thirdly, the method uses finger positions to estimate the hand coordinate. Therefore it is sensitive to finger state. We improved HandyAR to solve these problems. Experimental result showed that our method had higher performance than conventional methods.