Illumination Normalization of Human Face Based on the Second-Order Polynomial Model

XIE Xiao-hua; LAI Jian-huang; ZHENG Wei-shi

您当前的位置：

首页 >

文章列表页 >

Illumination Normalization of Human Face Based on the Second-Order Polynomial Model

更新时间：2025-07-16

- Illumination Normalization of Human Face Based on the Second-Order Polynomial Model
- Acta Electronica Sinica Vol. 38, Issue 8, Pages: 1791-1797(2010)
- 作者机构：
  
  1. 中山大学数学与计算科学学院,广东,广州,510275
  2. 中山大学信息科学与技术学院,广东,广州,510275
  3. 伦敦大学玛丽皇后学院计算机系,伦敦,英国
  4. 广东省信息安全技术重点实验室,广东,广州,510275
  5. 中山大学数学与计算科学学院广东广州,510275
  6. 中山大学信息科学与技术学院广东广州,510275
  7. 伦敦大学玛丽皇后学院计算机系英国伦敦
  8. 广东省信息安全技术重点实验室广东广州,510275
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP391
- Published：2010
- 稿件说明：
移动端阅览
XIE Xiao-hua, LAI Jian-huang, ZHENG Wei-shi. Illumination Normalization of Human Face Based on the Second-Order Polynomial Model[J]. Acta Electronica Sinica, 2010, 38(8): 1791-1797.
DOI：

XIE Xiao-hua, LAI Jian-huang, ZHENG Wei-shi. Illumination Normalization of Human Face Based on the Second-Order Polynomial Model[J]. Acta Electronica Sinica, 2010, 38(8): 1791-1797. DOI：

摘要

根据人脸光照变化非线性的特性

本文提出用二次多项式模型来描述非正面光照条件下的人脸图像与正面光照条件下的人脸图像在对应象素点灰度之间的关系

进而提出一种人脸光照归一化方法.本方法的一个重要特点是独立于先验的物理模型而通过建立统计回归模型来学习不同光照图像之间的关系.此外

提出用基于PCA的方法对光照归一化后的人脸图像进行加权补偿

进一步改善图像的视觉效果.在Extended YaleB和CMU-PIE人脸库上的实验结果表明

新方法在改善人脸视觉效果同时能大大提高人脸识别准确率.

Abstract

We proposed to describe the relationship between the pixel grey values of the face images under frontal and non-frontal illumination conditions by using a second order polynomial model.Correspondingly

an illumination normalization method based on such nonlinear model was formed.The proposed method learns the illumination variations in a statistical manner by using regression model without any prior physical knowledge.Furthermore

in order to improve the visual quality

a PCA-based weighting compensation for the normalized face image was proposed.The experimental results on the Extended Yale B and CMU-PIE face databases show that the proposed method can attain good visualization for face images

and significantly improve the face recognition performance.

关键词

Keywords

references

Views

1487

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Lightweight New Fesidual Face Recognition Method Based on Feature Fusion

Multi-task Learning and Identity-constrained Generative Adversarial Network for Face Frontalization and Recognition

A Kernel Orthogonal Discriminant Locality Preserving Projections Method

Matrix Pattern Based Minimum Within-Class Scatter Support Vector Machines

A New Approach for Face Recognition Based on Singular Value Features and Neural Networks

Related Author

HUI Kang-hua

YAN Jian-qing

GAO Si-hua

HE Huai-qing

HUANG Xin-yan

LIU Fang

BAO Qian-yue

LI Ren-peng

Related Institution

College of Computer Science and Technology， Civil Aviation University of China

Joint International Research Laboratory of Intelligent Perception and Computation， Xidian University

International Research Center for Intelligent Perception and Computation， Xidian University

Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education， Xidian University

School of Artificial Intelligent， Xidian University

⁰