Abstract: A new kind of hybrid model based on Support Vector Machine（SVM） and Hidden Markov Model（HMM） is proposed and has been used in face recognition.The faces in the photos are first located.The independent features of the faces are then picked up and the hybrid model is used to recognize the faces.The experimental results based on ORL face database show that compares with other methods,such as pure SVM and pure HMM,the hybrid model has the best performance.

摘要： 采用支持向量机（SVM）和隐马尔可夫模型（HMM）相结合的方法进行人脸识别。首先对照片中的人脸进行定位,从定位区域提取人脸各个器官的独立基特征,然后使用支持向量机和隐马尔可夫混合模型对定位区域进行人脸识别。利用SVM和HMM结合的优点,取得较高的识别率。