FACE CLASSIFICATION USING OPTIMUM FEATURES OF LPWT FACE IMAGES

机译：使用LPWT人脸图像的最佳特征进行人脸分类

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

In this paper, we propose a robust face classificationrnmethod using optimum feature vector of LPWT (Logpolarrnwavelet transform) face images. Face feature vectorrnis extracted from LPWT face image using HLAC (Higherrnorder local autocorrelation). The extracted highrndimensional face feature vector is transformed tornoptimum low dimensional feature space by MNLMrn(Membership-based nonlinear least-square minimization)rnmethod to improve classification rate.rnIn order to show the effectiveness of our method,rncomputer simulations were executed. Actual scale andrnillumination degraded face images were used to classifyrnface image. The classification accuracy was evaluatedrncompared with several combined methods such as PCA,rnLDA and spectroface. It is shown that the proposedrnmethod using the optimum face feature vectors achievernhigh classification rate for degraded face imagesrnclassification compared with the conventional standardrnapproaches.

机译：在本文中，我们提出了一种使用LPWT（对数小波变换）人脸图像的最佳特征向量的鲁棒人脸分类方法。使用HLAC（高阶局部自相关）从LPWT人脸图像中提取人脸特征向量。通过基于成员最小二乘最小二乘法（MNLMrn）的方法将提取的高维人脸特征向量转化为低维低维特征空间，以提高分类率。为了证明我们方法的有效性，进行了计算机仿真。使用实际比例和照度退化的人脸图像对人脸图像进行分类。与几种组合方法如PCA，rnLDA和Spectroface相比，对分类精度进行了评估。结果表明，与传统的标准方法相比，该方法利用最优的人脸特征向量对分类后的人脸图像实现了较高的分类率。

著录项

来源
《Proceedings of the IASTED international conferences on informatics 2010》|2010年|p.556-561|共6页
会议地点 Marina del Rey(US);Marina del Rey(US);Marina del Rey(US);Marina del Rey(US);Marina del Rey(US)
作者
Naoki Kato; Shigeo Wada;
展开▼
作者单位

Graduate School of Engineering, Tokyo Denki University 2-2 Kanda-Nishiki-cho, Chiyoda-ku, Tokyo 101-8457, Japan 09kme14@ms.dendai.ac.jp;

Graduate School of Engineering, Tokyo Denki University 2-2 Kanda-Nishiki-cho, Chiyoda-ku, Tokyo 101-8457, Japan wada@cck.dendai.ac.jp;

展开▼
会议组织
原文格式 PDF
正文语种 eng
中图分类信息处理技术;计算技术、计算机技术;
关键词
face classification; log-polar transform; wavelet transform; higher order local autocorrelation; dimensional reduction;

机译：人脸分类;对数极坐标变换;小波变换;高阶局部自相关;降维;

相似文献

外文文献
中文文献
专利

1. Optimal Combination of Classification Algorithms and Feature Ranking Methods for Object-Based Classification of Submeter Resolution Z/I-Imaging DMC Imagery [J] . Francisco Alonso-Sarr#x000ED, a, Fulgencio C#x000E1, Remote Sensing . 2015,第4期

机译：亚目标分辨率Z / I成像DMC图像的基于对象分类的分类算法和特征排序方法的最佳组合
2. Ischemic stroke lesion detection, characterization and classification in CT images with optimal features selection [J] . R. Kanchana, R. Menaka Biomedical Engineering Letters . 2020,第3期

机译：具有最佳特征选择的CT图像中的缺血性卒中病变检测，表征和分类
3. Optimal feature and classifier selection for text region classification in natural scene images using Weka tool [J] . Soni Rituraj, Kumar Bijendra, Chand Satish Multimedia Tools and Applications . 2019,第22期

机译：使用Weka工具在自然场景图像中进行文本区域分类的最佳特征和分类器选择
4. FACE CLASSIFICATION USING OPTIMUM FEATURES OF LPWT FACE IMAGES [C] . Naoki Kato, Shigeo Wada IASTED International Conferences on Informatics . 2010

机译：使用LPWT面部图像的最佳特征进行脸部分类
5. Data mining to identify optimal spatial aggregation scales and input features: Digital image classification with topographic LIDAR and LIDAR intensity returns. [D] . Tullis, Jason Alan. 2003

机译：数据挖掘以识别最佳的空间聚集尺度和输入特征：具有地形LIDAR和LIDAR强度返回的数字图像分类。
6. Ischemic stroke lesion detection characterization and classification in CT images with optimal features selection [O] . R. Kanchana, R. Menaka 2020

机译：具有最佳特征选择的CT图像中的缺血性卒中病变检测表征和分类
7. Optimal Combination of Classification Algorithms and Feature Ranking Methods for Object-Based Classification of Submeter Resolution Z/I-Imaging DMC Imagery [O] . Fulgencio Cánovas-García, Francisco Alonso-Sarría 2015

机译：基于对象分类的子分辨率Z / I成像DmC图像的分类算法和特征排序方法的优化组合
8. Selection of optimal textural features for maximum likelihood image classification. [R] . Rosenblum, W. I., Salvaggio, C., Schott, J. R. 1990

机译：选择最佳可能性图像分类的最佳纹理特征。

FACE CLASSIFICATION USING OPTIMUM FEATURES OF LPWT FACE IMAGES

摘要

著录项

相似文献

相关主题

期刊订阅