基于主分量分析的BP神经网络人脸图像识别算法

本文提出了一种基于主分量分析法和反向传播神经网络的图像识别方法,并详细阐述了这种方法的具体实现过程。在整个算法过程中,主分量分析法主要用于图像的预处理,也就是提取有用的特征样本;反向传播神经网络则是作为一个分类器对未知图像进行分类。此方法具有较强的自适应性、较高的识别率以及对某些噪声的鲁棒性。

The Arithmetic of Face Image Recognition via BP Neutral Network based on Principal Component Analysis

In this paper we proposed an algorithm of image recognition which is based on both principal component analysis (PCA) and back propagation (BP) neutral network and illustrated the particular steps by which this algorithm is realized. In the entire algorithm, PCA is mainly used in pretreatment of images while BP neutral network functions as a procedure classifying those unknown but pretreated images. This method is considered to be of relatively strong auto-adaptability, high recognition rate and acceptable robustness towards certain noises.