基于 gabor变换和 bp神经网络的 人脸检测算法的 m atlab实现（Face detection algorithm based on Gabor transform and BP neural network m implementation based on atlab）.docVIP

基于 gabor变换和 bp神经网络的 人脸检测算法的 m atlab实现（Face detection algorithm based on Gabor transform and BP neural network m implementation based on atlab） Twenty-fifth. Second April 2010 Journal of Zhengzhou University of Light Industry (NATURAL SCIENCE EDITION) J RNAL O F ZH ENG ZHOU U N I ER S I O F L I OU V TY GHT I NDU STRY (N ATU ral Science) V o 25 N 2 L. O. A PR 2010. article number: 1004- 1478 (2010) 02- 0082- 03 based on the face detection algorithm based on Gabor transform and BP neural network M atlab Wang Lijuan (North Central University School of information and communication engineering, Shanxi Taiyuan 030051) Abstract: This paper presents a face detection algorithm based on Gabor transform and BP neural network. The algorithm based on M atlab platform, using Ga bor achieve wavelet transform to extract image the characteristics, and the extracted feature dimension, through the construction of 2 layer BP neural network, extracted from ORL face database of the samples for training and learning by template matching and morphology of corrosion and expansion. Daniel, the face detection for the test sample. The experimental results show that the algorithm has low computational complexity, fast operation, high accuracy of detection and location of the frontal face. Keywords: face recognition; G abor transform; BP neural network classification number: TP334 document code: A 4 Face detection realization by M atlab using Gabor wavelet and BP WANG L I ju an (C ollege of Infor. and Com. Eng., N orth Un IV of China, T Aiyuan 030051, China Abstract face detect IO) A n algorith based on Gabor transfor and BP neura l netw ork w as proposed It m runs in th: m e platfor of M atlab and, uses the G abor w ave let transfor for I age feature extraction then m m m, m akes reductio n of di ensionality, constructs a tw o BP N eural netw Orks to train the sam ples extracted m fr Om the ORL face database and fin ally do the face detection for th e test samp le usin g TEM plate m atch in G a