摘要
人脸识别是对从视频图像中检测到的人脸区域进行身份的认证,是将待识别人脸与数据库中的人脸进行匹配的过程。将EHMM应用于人脸识别,提取人脸的DCT系数特征作为观察向量,用EHMM算法进行人脸模型训练和识别,并使用OpenCV对人脸识别算法进行功能仿真验证和相关探究,达到较好的人脸识别效果。实验结果表明,正常光照下,该算法的识别率在95%以上。
Face recognition extracts the target face from the input image and then calculates the distances between the detected face and the faces saved in the face lib. The one that gives out the minimum distance is selected as the identified person. Adopts an EHMM to implement the face recognition, where DCT coefficients of the face image is selected as the observation vectors. Executes simulation experiments based on OpenCV platform. The experiments show that under normal illumination, the recognition accuracy can be above 95%.
基金
国家级大学生创新训练项目(No.201210285031)
苏州大学大学生创新创业训练计划
