| 基于PCA和神经网络的人脸识别算法研究 |
| |
| 引用本文： | 唐赫.基于PCA和神经网络的人脸识别算法研究[J].人天科学研究,2013(6):33-34. |
| |
| 作者姓名： | 唐赫 |
| |
| 作者单位： | 武汉理工大学理学院统计系,湖北武汉430070 |
| |
| 摘 要： | 在MATLAB环境下,取ORL人脸数据库的部分人脸样本集,基于PCA方法提取人脸特征,形成特征脸空间,然后将每个人脸样本投影到该空间得到一投影系数向量,该投影系数向量在一个低维空间表述了一个人脸样本,这样就得到了训练样本集。同时将另一部分ORL人脸数据库的人脸作同样处理得到测试样本集。然后基于最近邻算法进行分类,得到识别率,接下来使用BP神经网络算法进行人脸识别,最后通过基于神经网络算法和最近邻算法进行综合决策,对待识别的人脸进行分类。
|
| 关 键 词： | 人脸识别 主成分 BP神经网络 最近邻算法 |
Research on Face Recognition Algorithm Based on PCA and Neural Network |
| |
| Abstract: | Based on the PCA method for face feature extraction, we take part of the ORI. face database to set the formation of the eigenface space in the MATLAB environment. That each face samples are projected into the space is a projection coefficient vector expressed in a lower dimensional space a face samples, so that we get the training sample set. While anoth- er part of the ORI. face database of people face make the same treatment to be the test sample set. It is classified based on the nearest neighbor algorithm to get the recognition rate, then use the BP neural network algorithm for face recognition, and finally through integrated decision to treat the human face of recognition based on neural network algorithms and nea rest neighbor algorithm to classify. |
| |
| Keywords: | Face Recognition Principal Component Analysis BP Neural Network Nearest Neighbor Algorithm |
| 本文献已被 维普 等数据库收录！ |
|
