Research on Facial Feature Points Extraction in Color Images

WU Zheng; ZHOU Yue; DU Chun-hua; YUAN Quan; GE Xin-liang

您当前的位置：

首页 >

文章列表页 >

Research on Facial Feature Points Extraction in Color Images

更新时间：2025-07-16

- Research on Facial Feature Points Extraction in Color Images
- Acta Electronica Sinica Vol. 36, Issue 2, Pages: 309-313(2008)
- 作者机构：
  
  1. 上海交通大学图像处理与模式识别研究所,上海,200240
  2. 华夏银行上海分行,上海,200120
  3. 上海交通大学图像处理与模式识别研究所上海,200240
  4. 华夏银行上海分行上海,200120
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP391.41
- Published：2008
- 稿件说明：
移动端阅览
WU Zheng, ZHOU Yue, DU Chun-hua, et al. Research on Facial Feature Points Extraction in Color Images[J]. Acta Electronica Sinica, 2008, 36(2): 309-313.
DOI：

WU Zheng, ZHOU Yue, DU Chun-hua, et al. Research on Facial Feature Points Extraction in Color Images[J]. Acta Electronica Sinica, 2008, 36(2): 309-313. DOI：

摘要

本文提出了一种基于复合局部信息模型的改进Active Shape Model(ASM)算法

并进一步提出了基于人脸特征点Gabor小波特征降维分类的特征点搜索方法

对改进ASM的结果进行精确校正

达到鲁棒精确定位特征点的目的.本文首先对经过Adaboost检测的彩色图像人脸区域进行光照补偿

然后根据眼睛和唇色的色度特性定位双眼和嘴唇中心位置

从而得到较好的人脸特征点的初始位置.在特征点位置搜索中

本文结合肤色概率信息对ASM方法进行了改进

从而提高了仅基于灰度梯度信息的传统ASM方法的鲁棒性和准确性.最后选取改进ASM搜索后的某些特征点一定领域内梯度值较高的点

提取其Gabor小波特征

通过线性判别式分析法(Linear Discriminant Analysis)和最近邻分类法对其进行分类

搜寻最符合训练样本Gabor特征的点作为最佳位置点

修正了ASM的搜索结果

使得搜寻结果更加精确.

Abstract

We present a method of facial feature point extraction based on improved Active Shape Model (ASM) and Gabor wavelet.Facial feature points can be located robustly and precisely by using the method proposed in this paper.Firstly

light compensation is used on the detected facial region.Secondly

we locate the eyes and mouth to obtain a good initial shape of feature points.And then the improved ASM is used to get a robust rough shape.We improve local gray-level model of the traditional ASM by adding local skin similarity probability

and form local hybrid model.Finally

we search the precise location of a certain feature point by classifying the Gabor feature of points around the rough location.The Gabor feature can be trained and dimensionally reduced by Linear Discriminant Analysis (LDA)

and classified by Nearest Neighborhood method.At the end of the paper

experimental results demonstrate the robustness and accuracy of this method.

关键词

Keywords

references

Views

1286

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Building Segmentation from High-Resolution SAR Images Based on Improved Markov Random Field

Related Author

FU Xing-yu

YOU Hong-jian

FU Kun

Related Institution

Key Laboratory of Technology in Geo-spatial Information Processing and Application System, Institute of Electronics, Chinese Academy of Sciences

Institute of Electronics,Chinese Academy of Sciences

Graduate University of Chinese Academy of Sciences

Key Laboratory of Technology in Geo-spatial Information Processing and Application SystemInstitute of Electronics Chinese Academy of SciencesBeijing 100190China

Institute of ElectronicsChinese Academy of SciencesBeijing 100190China

⁰