| 一种基于光流场的复杂背景下人脸定位方法 |
| |
| 引用本文： | 王凌,冯华君,徐之海,李奇. 一种基于光流场的复杂背景下人脸定位方法[J]. 计算机工程与应用, 2003, 39(8): 68-70,73 |
| |
| 作者姓名： | 王凌 冯华君 徐之海 李奇 |
| |
| 作者单位： | 浙江大学现代光学仪器国家重点实验室,杭州,310027 |
| |
| 基金项目： | 国家教委高等学校骨干教师资助计划资助项目 |
| |
| 摘 要： | 图像中人脸的自动定位和提取是进行人脸识别的首要条件。文章提出了一种基于光流场的复杂背景下人脸定位方法。通过改进的光流场约束方程与帧间差法的结合,确定运动人脸的所在区域。并对所在区域进行图像的操作和处理,提取出人脸图像,最后采用基于Karhunen-Loeve变换的人脸识别技术进行识别。该方法具有快速和准确的特点。
|
| 关 键 词： | 光流场 人脸定位 人脸识别 |
| 文章编号： | 1002-8331-(2003)08-0068-03 |
A Method for Human Face Location in a Complex Background Based on the Optical Flow |
| |
| Wang Ling Feng Huajun Xu Zhihai Li Qi. A Method for Human Face Location in a Complex Background Based on the Optical Flow[J]. Computer Engineering and Applications, 2003, 39(8): 68-70,73 |
| |
| Authors: | Wang Ling Feng Huajun Xu Zhihai Li Qi |
| |
| Abstract: | It is very important for human face recognition to find the face automatically in a scene.In this paper,a method to locate human face in a complex background is proposed.The moving face's position can be pointed out bases on the combination of revised definition of optical flow and inter-frame difference.Then the face can be extracted by image processing.In the end,the face is recognized by the method based on Karhunen-Loeve Variation.The method has the feature of accuracy and high speed. |
| |
| Keywords: | Optical flow Face detection Face recognition |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
