Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/73098
Full metadata record
DC FieldValueLanguage
dc.contributor.authorYang, Ziwei
dc.date.accessioned2018-01-03T05:37:54Z
dc.date.available2018-01-03T05:37:54Z
dc.date.issued2018
dc.identifier.urihttp://hdl.handle.net/10356/73098
dc.description.abstractIn order to improve the speed and robustness of multi-view face detection, this dissertation proposes a face detection method called multi-view face detection based on asymmetric principal component analysis (APCA) and support vector machine (SVM) classifier. APCA is proposed to remove the unreliable dimensions more effectively than the conventional PCA. Targeted at the two-class problem, an asymmetric discriminant analysis in the APCA subspace is proposed to regularize the eigenvalue which is a biased estimate of the variance in the corresponding dimension. In the training phase, five SVM linear classifiers are trained by using the positive database with different angles. Then APCA is applied on each group after classification. In the test phase, the linear classifier is used to quickly determine the different angles of faces, and then the Bhattacharyya distance is applied on asymmetric DA to further verify the face area of each group, thus to detect the face. The experimental results show the effectiveness and correctness of the proposed method.en_US
dc.format.extent65 p.en_US
dc.language.isoenen_US
dc.subjectDRNTU::Engineering::Electrical and electronic engineeringen_US
dc.titleMulti-view face detectionen_US
dc.typeThesis
dc.contributor.supervisorJiang Xudongen_US
dc.contributor.schoolSchool of Electrical and Electronic Engineeringen_US
dc.description.degreeMaster of Science (Signal Processing)en_US
item.fulltextWith Fulltext-
item.grantfulltextrestricted-
Appears in Collections:EEE Theses
Files in This Item:
File Description SizeFormat 
YANG ZIWEI_2017.pdf
  Restricted Access
2.16 MBAdobe PDFView/Open

Page view(s)

102
Updated on Jun 17, 2021

Download(s)

8
Updated on Jun 17, 2021

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.