首页> 外文会议>International Symposium on Neural Networks pt.1; 20040819-20040821; Dalian; CN >Local Face Recognition Based on the Combination of ICA and NFL
【24h】

Local Face Recognition Based on the Combination of ICA and NFL

机译:基于ICA和NFL结合的局部人脸识别

获取原文
获取原文并翻译 | 示例

摘要

This paper presents a local face recognition algorithm that is based on independent component analysis (ICA) and the nearest feature line (NFL). First, we separate a face image into several facial components. Then, we extract feature through combination of principal component analysis (PCA) and ICA; in the step of recognition, we first get each part of distance by NFL, then we calculate the synthetical distance by combining different parts. Compared with holistic image representation, this method has many advantages, such as a much higher recognition rate, more stable and flexible in practice.
机译:本文提出了一种基于独立成分分析(ICA)和最近的特征线(NFL)的局部人脸识别算法。首先,我们将面部图像分为几个面部成分。然后,通过结合主成分分析(PCA)和ICA来提取特征;在识别步骤中,我们首先通过NFL得到距离的每个部分,然后通过组合不同的部分来计算综合距离。与整体图像表示相比,该方法具有很多优点,例如更高的识别率,在实践中更加稳定和灵活。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号