计算机与现代化 ›› 2013, Vol. 1 ›› Issue (5): 70-72.doi: 10.3969/j.issn.1006-2475.2013.05.016
• 图像处理 • 上一篇 下一篇
林景亮，李祖裕
收稿日期:
修回日期:
出版日期:
发布日期:
LIN Jing-liang， LI Zu-yu
Received:
Revised:
Online:
Published:
摘要： 离散余弦变换(DCT)具有图像信号处理后对图像显示效果影响不明显的优点。本文针对离散余弦变换(DCT)系数在人脸识别中的应用，研究不同的DCT块进行编码后其系数对图像质量和识别结果的影响，提出利用快速2DDCT（二维离散余弦变换）代替普通DCT的人脸特征提取方法。该方法可以使得在对人脸进行平均分割后的图像块处理时间大大减少，从而能够在同样的时间内提取更多的DCT系数用于人脸识别。实验结果表明，该方法快速而且是有效的。
关键词: 离散余弦变换, 快速算法, 人脸特征提取
Abstract: Discrete cosine transform (DCT) has an advantage that is the image signal processing without affecting image quality. The paper proposes a new algorithm for face feature extraction. The face image is the first segmented in average, and the fast 2D-DCT algorithm instead of common DCT algorithm is used in the stage of feature vector extraction. The new algorithm has been tested, and the result proves that 2D-DCT can reduce most of time for the stage of face recognition or can extract more DCT coefficient than common DCT algorithm. Experimental results show that the proposed algorithm is fast and effective.
Key words: DCT, fast algorithm, face feature extraction
中图分类号:
TP391.41
林景亮;李祖裕. 基于快速二维离散余弦变换算法的人脸特征提取[J]. 计算机与现代化, 2013, 1(5): 70-72.
LIN Jing-liang;LI Zu-yu. Face Feature Extraction Based on Fast 2D-DCT Algorithm[J]. Computer and Modernization, 2013, 1(5): 70-72.
0 / / 推荐
导出引用管理器 EndNote|Ris|BibTeX
链接本文: http://www.c-a-m.org.cn/CN/10.3969/j.issn.1006-2475.2013.05.016
http://www.c-a-m.org.cn/CN/Y2013/V1/I5/70
