• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Signal Processing(ICSP'06); 20061116-20; Guilin(CN) >A Novel Method for Palmprint Recognition Based on Wavelet Transform
【24h】

A Novel Method for Palmprint Recognition Based on Wavelet Transform

机译：基于小波变换的掌纹识别新方法

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

A novel method for palmprint recognition based on wavelet transform is presented in this paper. Because palmprint's principal lines, wrinkles and ridges have characteristics of various resolutions, a method for multi-scale wavelet transform analysis is utilized in palmprint recognition. There are three steps in the scheme, which are image preprocessing, feature extraction and recognition. Original palmprint image is aligned, segmented and enhanced firstly. Then it is decomposed into multi-scale wavelet sub-images. After the wavelet sub-images are segmented into variable blocks, the mean of each block is calculated to form a normalized vector, which corresponds to wavelet sub-image. Finally, the feature vectors of each wavelet sub-image are combined into a vector, called the palmprint feature. Feature matching is based on Euclidean distance between feature vectors and Nearest Neighbor Distance (NND) rule. Experimental results based on PolyU Palmprint Database illustrate that the approach is valid in palmprint recognition.
机译：提出了一种基于小波变换的掌纹识别新方法。由于掌纹的主线，皱纹和山脊具有各种分辨率的特征，因此在掌纹识别中采用了一种多尺度小波变换分析方法。该方案包括图像预处理，特征提取和识别三个步骤。首先对原始掌纹图像进行对齐，分割和增强。然后将其分解为多尺度小波子图像。在将小波子图像分割成可变块之后，计算每个块的平均值以形成归一化矢量，该矢量对应于小波子图像。最后，每个小波子图像的特征向量被组合成一个向量，称为掌纹特征。特征匹配基于特征向量之间的欧几里得距离和最近邻距离（NND）规则。基于理大掌纹数据库的实验结果表明，该方法在掌纹识别中是有效的。

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