计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (25): 189-192.DOI: 10.3778/j.issn.1002-8331.2008.25.057
• 图形、图像、模式识别 • 上一篇 下一篇
刘丽华
收稿日期:
修回日期:
出版日期:
发布日期:
通讯作者:
LIU Li-hua
Received:
Revised:
Online:
Published:
Contact:
摘要： 光照是影响人脸识别率的主要因素，它已成为人脸识别技术发展的瓶颈。根据商图像理论，在所构造的低维训练集上对待识别的图像进行光照条件估计，通过加光和去光两种方法，实现了光照补偿的目的，并通过识别实验验证了其补偿效果。
关键词: 人脸识别, 商图像, 光照补偿, 线性子空间
Abstract: Illumination is a pivotal factor in the ratio of face recognition.It is still one bottleneck preventing development of face recognition technique.Based on quotient image theory，the illumination condition is estimated on facial de-dimension illumination training set.The aim of facial illumination compensation is implemented by two basic strategies of adding light and reducing light. And the results show that this improved facial illumination compensation method is effective.
Key words: face recognition, quotient image, illumination compensation, linear subspaces
刘丽华. 基于线性子空间和商图像理论的人脸光照补偿[J]. 计算机工程与应用, 2008, 44(25): 189-192.
LIU Li-hua. Facial illumination compensation based on linear subspaces and quotient image theory[J]. Computer Engineering and Applications, 2008, 44(25): 189-192.
0 / 推荐
导出引用管理器 EndNote|Ris|BibTeX
链接本文: http://cea.ceaj.org/CN/10.3778/j.issn.1002-8331.2008.25.057
http://cea.ceaj.org/CN/Y2008/V44/I25/189
