Face Recognition Based On BP Neural Network System Research And Realization

We develop and improve an algorithm in order to detect the faces and recognizetheses identity in daily life images in the varied background。 We use less-dimensionvectors to reduce images complexity and improving interference with noise in images，increasing ability of face detection and recognition。The system of face detection and recognition is divided into three stages： facedetection， face location， and face recognition。 In the first stage， we use a fuzzyGaussian classifier and a face feature extracting neural network to detecting faces inimage。 In this stage， we hope to divide images to face images and non-face imagesroughly by fuzzy neural network classifier。 We compute the fuzzy characteristic vectorof input images，and then accumulate the square errors of characteristic vector betweentraining patterns to exclude the most part of non-face image。Next，we feed the passedimages to the feature extracting neural network for detecting faces accurately。 In theface location stage， we use Gaussian spread method to remove some fault detections inthe previous detecting stage and locate the faces in images。 In the last stage， we use aPCA and a framework of parallel neural networks to recognize the faces that located inthe previous stage。 The PCA can classify each input image to some clusters andactivate their small-scale parallel neural networks to recognize the input images。Our algorithm can reduce the dimension of images， and eliminate a great deal ofnon-face images by classifier。 Therefore， we can decrease the training time andrecognition efficiently。 Further， we can promote the detection and recognition abilityof complex face images accurately。...