基于Boosting RBF神经网络的人脸年龄估计

年龄变化是引起人脸外观变化的主要原因，但每个人的生活方式不同，难以准确地从人脸图像中估计年龄。该文提出了一种基于人脸图像的年龄估计方法，用NMF方法提取人脸特征，通过RBF神经网络确定一个人脸图像及其相符年龄之间的估计函数。在此基础上，为了提高神经网络的泛化能力和故障诊断的准确性，利用Boosting方法构造一个基于神经网络的函数序列，将它们组合成一个加强的估计函数，实验结果表明了该方法的正确性。

Age Estimation of Facial Images Based on Boosting RBF Neural Network

Age change is the mainly factor of causing variations in the appearance of human faces. Due to many lifestyle factors, it is difficult to precisely estimate age from face images. This paper presents a method to estimate age from facial images and uses NMF for extraction of facial features. These facial features are used to build a function for an age estimation a relationship between a facial image and its corresponding age using RBF neural network. In order to improve the precision of the RBF neural network for age estimation, it uses boosting algorithm to build a integration-neural network, a set of experiments of age estimation are presented. Experiment results show good estimation performance of the method.