| 大会名称 |
|---|
| 2009年 情報科学技術フォーラム(FIT) |
| 大会コ－ド |
| F |
| 開催年 |
| 2009 |
| 発行日 |
| 2009/8/20 |
| セッション番号 |
| 6H |
| セッション名 |
| 顔画像解析 |
| 講演日 |
| 2009/09/04 |
| 講演場所(会議室等) |
| H会場（9号館2F 923教室） |
| 講演番号 |
| H-010 |
| タイトル |
| IMPROVING THE POSE INVARIANT FACE RECOGNITION PERFORMANCES USING MULTI-STAGE CLASSIFIER |
| 著者名 |
| Wijaya Gede Pasek Suta, 内村 圭一, 胡 振程, |
| キーワード |
| APCA, DLDA, Classifier, DCT, Compact-features |
| 抄録 |
| This paper presents an improvement of alternative to PCA technique (APCA) based face recognition using multistage classifiers. In this case, we implement both the APCA as the first stage classifier and the DLDA as the second stage classifier. Moreover, the compact features of face image which based on DCT analysis is implemented as dimensional reduction. The function of the first stage is to provide the class candidates, which are closely to the query face, and the second stage will determine the best similarity of the query face to the class candidate’s images. The proposed technique achieves better performance for both recognition rate and accuracy parameters than those of stand alone APCA and DLDA when it was tested using several databases (ITS-LAB., INDIA, and FERET) |
| 本文pdf |
| PDF download (142.2KB) |