《Application of Face Recognition Technology in Intelligent Ve.docx》由会员分享，可在线阅读，更多相关《Application of Face Recognition Technology in Intelligent Ve.docx（4页珍藏版）》请在三一文库上搜索。
1、Application of Face Recognition Technology in Intelligent Ve Abstract. One being developed intelligent vehicle security system， need to estimate if anyone on a certain range ahead is authorized users then intelligently open the car door or not， in order to ensure work convenience and anti-theft secu
2、rity. This paper proposed a method using face recognition technology to predict the data of image sensor. The experimental results show that， the proposed algorithm is practical and reliable， and good outcome have been achieved in the application of instruction. Keywords： intelligent vehicle securit
3、y system； face recognition； diagnostic faceIntroductionThe traditional mechanical car key is not only discommodious when a person is holding a bundle of goods， but also poor in anti-theft performance. As for the new-style keyless go system， its signal may be intercepted by wireless decoder of crimin
4、als.So， we develop an intelligent vehicle security system based on face recognition technology successfully. If anyone on a certain range ahead is authorized users， it automatically opens the car door. Or else， it keeps locking. This intelligent vehicle security system enjoys a reputation of high se
5、curity because it is photo-communication which is difficult to grab.We present an approach to the detection and identification of human faces and describe a working， near-real-time face recognition system which tracks a subjects head and then recognizes the person by comparing characteristics of the
6、 face to those of known individuals. Our approach treats face recognition as a two-dimensional recognition problem. Taking advantage of the fact that faces are normally upright and thus may be described by a small set of 2-D characteristic views. Face images are projected onto a feature space （“face
7、 space”） that best encodes the variation among known face images. The face space is defined by the “diagnosticfaces”， which are the diagnostic vectors of the set of races. They do not necessarily correspond to isolated features such as eyes， ears and noses. Automatically learning and later recognizi
8、ng new faces is practical within this framework.Recognition under reasonably varying conditions is achieved by training on a limited number of characteristic views（e g. a “straight on” view， a 40 view， and a profile view）.Diagnostic faces for recognitionMuch of the previous work on automated face re
9、cognition has ignored the issue of just what aspects of the face stimulus are important for identification assuming that predefined measurements were relevant and sufficient. This suggested to us that all information theory approach of coding and decoding lace image may give insight into the informa
10、tion content of face images， emphasizing the significant local and global “features”. Such features may or may not be directly related to our intuitive notion of face features such as the eyes， noses， lips and hair. We want to extract the relevant information in a face image encode it as efficiently
11、 as possible and compare one face encoding with a database of models encoded similarly. A simply approach to extract the information contained in an image of a face is to somehow capture the variation in a collection of images， independent of any judge of features and use this information to encode
12、and compare individual face images.In other words， we wish to find diagnostic vectors of the covariance matrix of the set of face images. These diagnostic vectors can be thought of a set of features which together characterize the variation between face images. Each image location contributes more o
13、r less to each diagnostic vectors. So we can display the diagnostic vector as a sort of ghostly face which we call a diagnostic face.Each face image in the training set can be represented exactly in terms of a linear combination of the diagnostic faces. The number of possible diagnostic faces is equ
14、al to the number of face images in the training set. However the faces can also be approximated used only the “best” diagnostic facesthose that have the largest diagnostic values， and which therefore account for the most variance within the set of face images. The primary reason for using fewer diag
15、nostic faces is computational efficiency. The best M diagnostic faces span an M dimensional subspace“face space”of all possible images. As sinusoids of varying frequency and phase are the basis functions of a fourier decomposition （and are in fact diagnostic functions of linear systems）. The diagnos
16、tic faces are the basis vectors of thediagnostic face decomposition.If a multitude of face images can be reconstructed by weighted sums of a small collection of characteristic images， then an efficient way to learn and recognize faces might be to build the characteristic features from known face ima
17、ges and to recognize particular faces by comparing the feature weights needed to （approximately） reconstruct them with the weights associated with the known individuals.The following steps summarize the recognition process：l. Initialization： Acquire the training set of face images and calculate the
18、diagnostic faces which define the face space.2 When a new face image is encountered， calculate a set of weights based on the input image and the M diagnostic faces by projecting the input image onto each of the diagnostic faces.3. Determine whether the image is a face at all （whether known or unknow
19、n）by checking to see if the image is sufficiently close to “face space”. 4. If it is a face， classify the weight pattern as either a known person or as unknown.Calculating diagnostic facesLet a face image I（x，y）be a two-dimensional N by N array of intensity values， or a vector of dimension N2.A typi
20、cal image of size 256 by 256 describes a vector of dimension 65，536， or， equivalently， a point in 65，536-dimensional space. An ensemble of images， then， maps to a collection of points in this huge space.Images of faces， being similar in overall configuration will not be randomly distributed in this
21、huge image space and thus can be described by a relatively low dimensional subspace. The main idea of us is to find the vectors which best account for the distribution of face images within the entire image space. These vectors define the subspace of face images， which we call “face space”. Each vec
22、tor is of length N2， describes a N by N image and is a linear combination of the original face images. Because these vectors are the diagnostic vector of the covariance matrix corresponding to the original face images. and because they are face like in appearance ，we refer to them as diagnostic face
23、s.With this analysis the calculation are greatly reduced from the order of the number of pixels in the images N2 to the order of the number of images in the training set M.In practice， the training set of images will be relatively small（），and the calculations become quite manageable. The associated
24、diagnostic values allow us to rank the diagnostic vectors according to their usefulness in characterizing the variation among the images Normally the background is removed by cropping training images so that the diagnostic face have zero value outside of the face area.Using diagnostic faces to class
25、ify a face imageOnce the diagnostic faces are created， identification becomes a pattern recognition task. The diagnostic faces span an M-dimensional subspace of the original N2 image space .The M significant diagnostic vectors of the L matrix are chosen as those with the largest associated diagnosti
26、c values. The number of diagnostic faces to be used is chosen heuristically based on the diagnostic values.A new face image L is transformed into its diagnostic face component （projected into “face space”） by a simple operation，k=ukT（L-） for .This describes a set of point-by-point image multiplicati
27、ons and summations.The weights from a vector ?T=1，2，3Mthat describes the contribution of each diagnostic face in representing the input face image， treating the diagnostic faces as a basis set for face images. The vector is used to find which of a number of pre-defined face classes， if any， best des
28、cribes the face. The simplest method for determining which face class provides the best description of an input face image is to find the face class k that minimizes the Euclidian distance k=|?-?k|，where ?k is a vector describing the kth face class. A face is classified as belonging to class k when
29、the minimum k become some chosen threshold .Otherwise the face is classified as “unknown”. Using diagnostic faces to detect facesWe can also use knowledge of the face space to detect and locate faces in single images. This allows us to recognize the presence of faces apart from the task of identifyi
30、ng them.Creating the vector of weights for an image is equivalent to projecting the image onto the low dimensional face space. The distance between the image and its projection onto the face space is simply the distance between mean-adjusted input image =L- and ， its projection onto the face space.I
31、n this paper， we propose a face recognition method which is applied in intelligent vehicle security system. Performance in the appliance shows that the method is of high accuracy and trustworthy. The deficiency is that it can only enable face recognition of single person， which points out the direct
32、ion for the improvement of our future work.References1 Tong Lin. Research on Face Recognition and Tracking Algorithm based on Video J. Computer and Modernization. 2013；02（15）；84-92.2 Yuan Li， Chen Qinghu. Multi-modal Face Recognition based on A Few Feature Points J. Computer Engineering and Applicat
33、ions， 2013；02（01）； 17-25.3 Zhu Wenzhong. Comparative Analysis of Face Recognition Algorithm based on Data Set J. Science Technology and Engineering. 20123；01（08）；22-27.4 Zhang Yousai， Yang Shu. Multi-pose face recognition algorithm based on local weighted average virtual samples J. Journal of Jiangs
34、u University of Science and Technology， 2013；02（15）；65-72.5 Gao Xiaojing， Pan Xin. Face Recognition based on GLOH Operator and Local Feature J. Computer Applications and Software. 2013； 05 （15）： 37-41.6 Han Juan. The Research of Face Recognition Algorithm based on Graph Matching J. Computer Knowledg
35、e and Technology.2013；01（05）：75-82.7 Cui Qi， Du Haishun. Research on Face Recognition Method based on Local Matching J. Information and Computer. 2013； 03（15）：80-83.8 Tong Xiaonian， Wen Weiyu. Face Recognition Method Using MapReduce Model for Training Support Vector Machines J. Journal of South-Central University For Nationalities. 2013；03（13）：17-21.