【正文】 人臉識(shí)別研究得到了前所未有的重視，國(guó)際上發(fā)表有關(guān)人臉識(shí)別等方面的論文數(shù)量大幅度增加，僅從 1990 年到 2022 年之間，SCI 及 EI 可檢索到的相關(guān)文獻(xiàn)多達(dá)數(shù)千篇，這期間關(guān)于人臉識(shí)別的綜述也屢屢可見。 interest. In the 1960 s, Bledsoe proposed face recognition system of semiautomatic mode and feature extraction method. In the 70 s, the developed countries such as America, Britain began to attach great importance to the research of face recognition and making progress. In 1972, Harmon with 佳木斯大學(xué)學(xué)士學(xué)位論文佳木斯大學(xué)信息電子技術(shù)學(xué)院32interaction is carried out on the face recognition methods in theory and practice. That same year, Sakai face image automatic recognition system is designed. In the early 80 s t. Minami developed better than that of Sakai face image automatic identification system. But early face recognition typically require some prior knowledge of the people, can39。% 計(jì)算原圖像中人臉模板的坐標(biāo)startx = cxmodx。 end。end。g=mean(ggg)。是你們毫無保留的指導(dǎo)和細(xì)致耐心的幫助，我才能順利完成這次畢業(yè)設(shè)計(jì)?；?Matlab 的人臉識(shí)別系統(tǒng)的仿真的研究很有意義。global im。T = CreateDatabase(TrainDatabasePath)。測(cè)試庫(kù)路徑選擇...39。else gui_mainf(gui_State, varargin{:})。gui_Name39。Position39。 x2=x2+r。s=r*c。借助人臉描述對(duì)提取出來的人臉就可以進(jìn)行人臉識(shí)別，即通過提取人臉特征來確定其身份。j2=imfilter(i,h)。figure,subplot(1,2,1),imhist(i)。j=rgb2gray(i)。所謂人臉特征提取即針對(duì)人臉的某些特征進(jìn)行的提取。由于人類有非常復(fù)雜的細(xì)節(jié)變化，例如眼鏡、胡須、發(fā)型等附屬物的干擾，這就給該項(xiàng)技術(shù)帶來了巨大挑戰(zhàn)。國(guó)家“十一五”科技發(fā)展規(guī)劃就將人臉識(shí)別技術(shù)的研究與發(fā)展列入其中，并明確指出“要在生物特征識(shí)別技術(shù)領(lǐng)域縮小與世界先進(jìn)水平的差距，開展生物特征識(shí)別應(yīng)用技術(shù)研究，開發(fā)具有高安全性、低誤報(bào)率的出入口控制新產(chǎn)品。佳木斯大學(xué)畢業(yè)論文基于 Matlab 的人臉識(shí)別系統(tǒng)設(shè)計(jì)與仿真學(xué) 院 信息電子技術(shù)專 業(yè) 電子信息工程班 級(jí) 11 級(jí) 1 班學(xué) 籍 號(hào) 11100540116姓 名 楊 雷指導(dǎo)教師 周經(jīng)國(guó)佳 木 斯 大 學(xué)2022 年 6 月 10 日佳木斯大學(xué)學(xué)士學(xué)位論文佳木斯大學(xué)信息電子技術(shù)學(xué)院1摘要人臉識(shí)別即指利用分析比對(duì)人臉視覺特征信息從而達(dá)到身份鑒別效果的計(jì)算機(jī)技術(shù)。 ”在這種形勢(shì)下，國(guó)內(nèi)一些科研院所在人臉識(shí)別技術(shù)上有了重大發(fā)展和突破。成功構(gòu)造出人臉識(shí)別系統(tǒng)將為解決其他與之類似的復(fù)雜問題提供重要的啟示。人臉特征提取，也被稱為人臉表征，是對(duì)人臉特征進(jìn)行建模的過程。imshow(j)。subplot(1,2,2),imhist(j)執(zhí)行后得到的圖像如下所示：佳木斯大學(xué)學(xué)士學(xué)位論文佳木斯大學(xué)信息電子技術(shù)學(xué)院9 圖 均衡化后的灰度圖像 圖 均衡化前后的直方圖對(duì)比圖 灰度圖像平滑與銳化處理平滑濾波器的作用是模糊圖像或者消除噪聲, 圖像處理工具箱為我們提供了 wiener2()來實(shí)現(xiàn)對(duì)圖像噪聲的自適應(yīng)濾波，medfilter2()函數(shù)用來實(shí)現(xiàn)中值濾波。figure,subplot(1,2,1),imshow(i)subplot(1,2,2),imshow(j2)佳木斯大學(xué)學(xué)士學(xué)位論文佳木斯大學(xué)信息電子技術(shù)學(xué)院10執(zhí)行上述代碼后得到的圖像如下所示：圖 平滑濾波效果圖圖 銳化濾波效果圖圖 中，第一個(gè)為加入噪聲的圖像，第二個(gè)為濾波后的圖像；圖 中，第一個(gè)為原灰度圖像，第二個(gè)為銳化后的圖像。開 始圖像采集 檢測(cè)定位是否定位成功是否處理成功圖像預(yù)處理人臉識(shí)別是否識(shí)別成功識(shí)別結(jié)果結(jié) 束是是是圖 基本框架圖 人臉檢測(cè)定位算法人臉檢測(cè)定位算法可分為兩大類：一類是基于隱式特征的方法；另一類是基于顯式特征的方法所謂的顯式特征，即指人類肉眼可以直觀看到的特征，如膚色、臉部結(jié)構(gòu)、臉部輪廓等。 for i=1:10 y1=1。end figure,imshow(BW)%%%%%%%%%%%%%%%%%%人臉定位%%%%%%%%%%%%%%%%%%%%%%% L = bwlabel(BW,8)。,[BB2(1,j2),BB2(1,j1),BB2(1,j),BB2(1,j)],39。, mfilename, ... 39。佳木斯大學(xué)學(xué)士學(xué)位論文佳木斯大學(xué)信息電子技術(shù)學(xué)院19end% End initialization code DO NOT EDIT% Executes just before faceCore is made visible.function faceCore_OpeningF(hObject, eventdata, handles, varargin)% This function has no output args, see OutputF.% hObject handle to figure% eventdata reserved to be defined in a future version of MATLAB% handles structure with handles and user data (see GUIDATA)% varargin mand line arguments to faceCore (see VARARGIN)% Choose default mand line output for faceCore = hObject。)。%[m V_PCA V_Fisher ProjectedImages_Fisher] = FisherfaceCore(T)。global TrainDatabasePath 。目前到處可見攝像頭，監(jiān)控錄像，這些的普及，使人臉識(shí)別具有重大商業(yè)價(jià)值。最后，我還要感謝身邊的朋友和同學(xué)，在大學(xué)生活的四年里，我在你們的陪伴中成長(zhǎng)，謝謝你們?cè)谧霎厴I(yè)設(shè)計(jì)這段日子里給予我的幫助。（2）找區(qū)域邊界function [left, right, up, down] = bianjie(A)[m n] = size(A)。if (right ~= 1) break。end。starty = cymody。t get rid of the intervention of people. Into the niy s, as a result of the parties face the pressing needs of the face recognition system, the research of face recognition is very popular. Face recognition method is a major breakthrough, entered the stage of the real machine automatic identification such as Karhunen Loeve transform or a new neural work technology. Face recognition research obtained the unprecedented attention, the number of papers published on face recognition and so on increased dramatically, from 1990 to 1990 alone, between SCI and EI can be retrieved as many as thousands of articles, related literature about face recognition during the period of this review is also visible. Abroad, there are many schools in facial recognition technology research, research field is very wide. These studies are military, police and big pany attaches great importance to and support the domestic some wellknown colleges and universities are engaged in the research of face recognition.Face recognition is a frontier topic in the field of pattern recognition, but the current face recognition was still in the stage of research topic, it is not active topic in the field of practical application. While humans can tell a person had no difficulty (in) the human face, but the use of puters for fully automatic face recognition has many difficulties, displays in: the face is a rigid body, the face changes。國(guó)外有許多學(xué)校在研究人臉識(shí)別技術(shù)，研究涉及的領(lǐng)域很廣。ve 變換等或新的神經(jīng)網(wǎng)絡(luò)技術(shù)。 % 產(chǎn)生矩形坐標(biāo)佳木斯大學(xué)學(xué)士學(xué)位論文佳木斯大學(xué)信息電子技術(shù)學(xué)院31附錄 2 外文參考文獻(xiàn)及翻譯Research development of face recognition at home and abroadFace recognition is one of the most outstanding ability in human visual. Its research involves pattern recognition, image processing, biology, psychology, cognitive science, with other biometric identifica method based on human perception and puter interaction field has a