基于小波分解和多分类支持向量机的脸谱识别

小波分解提取脸谱特征具有对表情变化不敏感的特点，支持向量机竹=为分类器具有很高的推广性能，无需先验知识，针对小波分解和支持向量机所具有的优点，提出了一种新的脸谱识别算法，在该算法中无需对洲练图像进行预处理，直接使用小波分解方法对脸谱图像进行特征提取，用所提取的脸谱特征向量组合成新的脸谱特征向链洲练多分类支持向量机模型，最后用训练好的支持向量机进行脸谱识别，在训练中分别采用了三种不同的核函数；使用ORL脸谱图像库对该算法进行了测试和评估，测试结果表明了该算法在识别性能方面的优越性。

Face Recognition Based on Wavelet Decomposition and MuitiSVM

The extracted features from human face images by wavelet decomposition are less sensitive to the facial expression variations.As a classifier,the SVM is provided with high generalization performance without need to add a transcendental knowledge.For the merit of wavelet decomposition and SVM,a novel algorithm for face recognition is presented.The process of the proposed method needs not preprocess the face images,directly extracts the appropriate features of human faces by wavelet decomposition,and composes the extracted features vectors to new features vectors,then trains the multi-class SVM model by the new face feature vectors,and uses the trained SVM model to classify the human faces at last.During the training process,three different kernel functions are adopted,The ORL faces database is selected to test and evaluate the proposed algorithm.The results of the test indicate the good performance of proposed algorithm in face recognition.