| 基于小波分析和DCT的人脸特征提取 |
| |
| 引用本文： | 王克奇,朱金魁,白雪冰.基于小波分析和DCT的人脸特征提取[J].自动化技术与应用,2009,28(4):65-68. |
| |
| 作者姓名： | 王克奇 朱金魁 白雪冰 |
| |
| 作者单位： | 东北林业大学,机电学院,黑龙江,哈尔滨,150040 |
| |
| 摘 要： | 人脸特征提取是人脸识别中重要的一个环节。本文提出利用小波分析,对人脸图像进行压缩,然后对压缩后的图像进行离散余弦变换,将提取到的系数作为特征向量。把特征向量用支持向量机进行分类。本文以matlab7.0为开发平台在ORL和YALE人脸图像库中对该方法的可行性进行了测试。实验表明,该方法与传统的KLT方法相比可以减少运算时间,并提高识别率5%-10%左右。它在准确率和计算速度上都取得了良好效果。
|
| 关 键 词： | 人脸特征提取 小波分析 离散余弦变换 支持向量机 |
Face Feature Extraction Based On the Wavelet Analysis And Discrete Cosine Transform(DCT) |
| |
| WANG Ke-qi,ZHU Jin-kui,BAI Xue-bing.Face Feature Extraction Based On the Wavelet Analysis And Discrete Cosine Transform(DCT)[J].Techniques of Automation and Applications,2009,28(4):65-68. |
| |
| Authors: | WANG Ke-qi ZHU Jin-kui BAI Xue-bing |
| |
| Affiliation: | ( College of Machinery and Electrcal Engineering, Northeast Forestry University, Harbin 150040 China) |
| |
| Abstract: | This paper presents a new method for face feature extraction. The face image is first compressed by using the wavelet analysis, and the Discrete Cosine Transform is then used to extract the feature vectors for classification by the Support Vector Machine （SVM）. It is tested on the ORL and Yale face database with matlab 7.0. Experiment shows that the new method can reduce the calculation time and increase the identification rate by about 5%-10% compared with the traditional KLT method. |
| |
| Keywords: | face feature extraction wavelet analysis discrete cosine transform support vector machine |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
