• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>International Journal of Innovative Computing Information and Control >ROBUST FACE RECOGNITION FRAMEWORK WITH BLOCK WEIGHTED SPARSE REPRESENTATION BASED CLASSIFICATION
【24h】

ROBUST FACE RECOGNITION FRAMEWORK WITH BLOCK WEIGHTED SPARSE REPRESENTATION BASED CLASSIFICATION

机译：基于加权加权稀疏表示的鲁棒人脸识别框架

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

The sparse representation based classification method can be divided into two categories: holistic approaches and local feature-based approaches. In spite of the significant success in face recognition, improvements on higher robustness or lower computational complexity are still necessary for its real application. Thus, we first propose a novel Block Weighted Sparse Representation based Classification (BW-SRC) method based on the maximum likelihood model. Then, to ensure the accuracy of BW-SRC, we conduct a pre-alignment process by utilizing the locations of local feature points (in this article, we use SIFT keypoints). Combining the pre-alignment process and BW-SRC, we establish a novel framework for robust face recognition, which is more effective and more robust than the state-of-the-art methods in practical scenarios. Finally, by conducting experiments on AR and Yale databases, the performance of our proposed method and framework is demonstrated and compared with global SRC and blocked SRC. The proposed framework is proven as low-computation, alignment-free and robustness to rotation, illumination and disguise, and more appropriate for practical scenarios.
机译：基于稀疏表示的分类方法可以分为两类：整体方法和基于局部特征的方法。尽管人脸识别取得了巨大的成功，但对于其实际应用而言，仍然需要提高鲁棒性或降低计算复杂度。因此，我们首先提出一种基于最大似然模型的新颖的基于块加权稀疏表示的分类（BW-SRC）方法。然后，为了确保BW-SRC的准确性，我们利用局部特征点的位置进行了预对准过程（在本文中，我们使用SIFT关键点）。结合预对准过程和BW-SRC，我们建立了一个新颖的人脸识别框架，该框架比实际方案中的最新技术更有效，更鲁棒。最后，通过在AR和Yale数据库上进行实验，证明了我们提出的方法和框架的性能，并将其与全局SRC和阻止的SRC进行了比较。所提出的框架被证明具有低计算量，免对准和对旋转，照明和伪装的鲁棒性，并且更适合于实际情况。

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