基于GLOH算子和局部特征融合的人脸识别

FACE RECOGNITION BASED GLOH DESCRIPTOR AND INTEGRATION OF LOCAL FEATURES

为了减小高维特征算子的计算复杂度、提高识别算法的准确率,提出一种基于GLOH(Gradient Location and OrientationHistogram)算子的人脸识别算法。首先将人脸图像划分为4个独立的子区域并对提取的特征点进行聚类。为了更有效地描述人脸特征以及特征匹配,为不同的区域赋予不同的权重值,并采取整体结合局部聚类子区域的方法进行人脸识别。通过在ORL人脸图像库上的实验,验证了算法的有效性,特别是在不同表情、不同姿态等干扰因素的条件下,表现出了较好的稳定性和鲁棒性。

In order to reduce the computational complexity of high-dimensional feature descriptor and to improve the accuracy of recognition algorithm,we propose a face recognition algorithm which is based on GLOH descriptor.First,face image is divided into four separate sub-regions and the feature points extracted are clustered.In order to describe the face feature and feature matching more effectively,different regions are given different weight values.The method of the whole combining with local clustering sub-region is employed for face recognition.The effectiveness of the algorithm is verified by experiments on the ORL face image database,which demonstrates good stability and robustness especially under the conditions of some confounding factors such as different facial expressions,postures and so on.