• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>The mediterranean journal of electronics and communications >CENTROID-SHAPE-SIGNATURE-WAVELET DESCRIPTOR BASED HAND BIOMETRIC SYSTEM
【24h】

CENTROID-SHAPE-SIGNATURE-WAVELET DESCRIPTOR BASED HAND BIOMETRIC SYSTEM

机译：基于质心形状签名小波描述符的人体生物测量系统

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

Biometrics is widely used for authorized access and information security. This work focuses on developing the features for hand geometry based biometric system which is robust, distinct, and computationally efficient with respect to time and complexity both. The previous methods used geometry based features like finger lengths, finger widths, palm width. These geometrical features suffer due to high dependency on landmark points. The proposed system uses the hand contour based features. The newly developed feature set is called Centroid Shape-Signature- Wavelet based Descriptors (CSSWD), as it is derived from the shape-signature of the central distance from the centroid of hand contour which is a global shape feature. Equal Error Rate (EER) values of 1.3 % and 0.95 % respectively are obtained with Mean absolute distance and Euclidean distance. In both the cases, identification accuracy of 99.5% is obtained. This performance is found to be better in comparison to other existing algorithms.
机译：生物识别技术广泛用于授权访问和信息安全。这项工作着重于开发基于手部几何的生物特征识别系统的功能，该功能在时间和复杂度方面都非常健壮，独特且计算效率高。先前的方法使用基于几何的特征，例如手指长度，手指宽度，手掌宽度。这些几何特征由于对地标点的高度依赖而受到影响。所提出的系统使用基于手轮廓的特征。新开发的特征集称为“基于质心形特征小波的描述符”（CSSWD），因为它是从到手轮廓质心的中心距离的形状特征中得出的，该特征是全局形状特征。使用平均绝对距离和欧几里得距离分别获得1.3％和0.95％的均等错误率（EER）值。在这两种情况下，识别准确率均达到99.5％。发现与其他现有算法相比，该性能更好。

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