...
【24h】

FACE RECOGNITION USING SHEARLETS EDGES FUSION

机译:使用Shearlet边缘融合的人脸识别

获取原文
获取原文并翻译 | 示例
           

摘要

In this paper, to save significant edges, we propose an algorithm of face recognition using shearlets edges fusion (Shearlets_EF). Firstly, the original images are decomposed into subimages by applying the shearlet transform. Then the directional subim-ages are fused by the maximum modulus principle. Thirdly, dividing fused images into patches, facial features of each patch are extracted by using uniform local binary pattern (ULBP) only with 2 times variations (ULBPO2V). Finally, face images are classified by collaborative representation based classification (CRC). Testing in the databases of AR, CMUPIE, Extended Yale B, and LFW, experimental results illustrate that Shearlets_EF has better performance compared to relative algorithms.
机译:在本文中,为了节省显着的边缘,我们使用沉焦融合(Shearlet_ef)提出了一种面部识别算法。首先,通过应用Shearlet变换,原始图像被分解为子像象。然后,定向子年龄被最大模量原理融合。第三,将融合图像分成贴片,通过仅使用2次变化(ULBPO2V)来提取每个贴片的面部特征,通过使用均匀的局部二进制模式(ULBP)来提取。最后,通过基于协作的分类(CRC)来分类面部图像。在AR的数据库中测试,CMUPIE,扩展耶鲁B和LFW,实验结果表明,与相对算法相比,Shearlets_EF具有更好的性能。

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