LEDTD: Local edge direction and texture descriptor for face recognition
详细信息 查看全文
- 作者：Jing Lia ; b ; hustlijing@hust.edu.cn" class="auth_mail" title="E-mail the corresponding author ; Nong Sanga ; b ; nsang@hust.edu.cn" class="auth_mail" title="E-mail the corresponding author ; Changxin Gaoa ; b ; cgao@hust.edu.cn" class="auth_mail" title="E-mail the corresponding author
- 关键词：Face recognition ; Local binary pattern ; Kirsch mask ; Local edge direction ; Image texture feature
- 刊名：Signal Processing: Image Communication
- 出版年：2016
- 出版时间：February 2016
- 年：2016
- 卷：41
- 期：Complete
- 页码：40-45
- 全文大小：561 K
文摘
- •
A new image descriptor LEDTD is presented for face recognition.
- •
LEDTD is the occurrence of local image center pixel edge direction and the texture information.
- •
WPCA is introduced to reduce the feature dimension and improve recognition performance.
- •
Experimental results on four databases demonstrate the superiority of our LEDTD compared with other image representation approaches.