• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Communication and Computing Systems >Feature extraction of hyperspectral face images using PCA in NIR
【24h】

Feature extraction of hyperspectral face images using PCA in NIR

机译：在NIR中使用PCA的高光谱面图像特征提取

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Face identification or Face Recognition (FR) is emergent research area because of wide-ranging of applications in the fields of business and regulation enforcement. Conventional FR methods are facing different challenges of types like object lighting, position dissimilarity, appearance variations, and lead to decrease in performance of object identification and verification. To prevail over all these challenges, Hyperspectral Image Set (HIS) may be used in human FR. The HIS minimize the several limitations because the skin spectra derived with these cubic dataset depicts the unique features for an individual. In this paper a modest and effective technique is discussed to take out the set of Attributes Vectors (AVs) with HIS and correspondingly to decrease the size of HIS. PCA is applied as FR method to extract the AVs in these large data sets. PCA has been proved as a capable tool in Hyperspectral Image Processing (HIP) as well as to minimize the dimensions of data set. Research is conducted by means of CMU hyperspectral dataset by considering the image wavelength in NIR region of Infrared Spectrum (IRS). A successful attributes extraction technique using PCA is studied intended for HIS and investigational outcomes are presented with AVs.
机译：面部识别或面部识别（FR）是紧急研究领域，因为业务领域的应用范围广泛。常规FR方法面临不同物体照明，位置异形，外观变化等类型的不同挑战，并导致对象识别和验证性能降低。为了以所有这些挑战为准，高光谱图像集（他）可以用于人FR。他最大限度地减少了几个限制，因为与这些立方数据集导出的皮肤光谱描绘了个体的独特特征。在本文中，讨论了一种适度和有效的技术，用他的且相应地将一组属性向量（AVS）取出，以减小他的大小。 PCA应用于FR方法以提取这些大数据集中的AV。 PCA已被证明是高光谱图像处理（HIP）中的能力工具，以及最小化数据集的尺寸。通过考虑红外光谱（IRS）的NIR区域的图像波长，通过CMU高光谱数据集进行研究。使用PCA的成功属性提取技术用于他的和调查结果与AVS一起介绍。

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