一、簡介

人臉檢測是人臉識別、人機交互、智能視覺監(jiān)控等:工作的前提。近年來,在模式識別與計算機視覺領(lǐng)域,人臉檢測已經(jīng)成為一個受到普遍 重視、研究十分活躍的方向。本文針對復(fù)雜背景下的彩色正面人臉圖像，將膚色分割、模板匹配與候選人臉圖像塊篩選結(jié)合起來，構(gòu)建了人臉檢測實驗系統(tǒng)，并用自制的人臉圖像數(shù)據(jù)庫在該系統(tǒng)下進行了一系列的實驗統(tǒng)計。本文首先介紹了人臉檢測技術(shù)研究的背景和現(xiàn)狀，闡明人臉檢測技術(shù)發(fā)展的重要意義，對目前常用的一一些檢測算法進行了總結(jié),然后著重闡述了基于膚色分割和模板匹配驗證的人臉檢測算法。膚色是人臉重要特征，在通過膚色采樣統(tǒng)計和聚類分析后，確立一種在YCbCr空間下的基于高斯模型的膚色分割方法。在YCbCr色彩空間中建立膚色分布的高斯模型,得到膚色概率似然圖像,在最佳動態(tài)閾值選取算法下完成膚色區(qū)域的分割。采用 數(shù)學(xué)形態(tài)學(xué)和一些先驗知識對所得到的膚色區(qū)域進行人臉特征篩選,進–步剔除非人臉膚色區(qū)域，減少候選人臉數(shù)量,簡化后續(xù)檢測過程的處理。本文使用平均模板匹配方法對候選人臉進行確認(rèn)，并針對圖像中的人臉通常有一定角度旋轉(zhuǎn)和尺寸大小不確定的問題，通過計算候選人臉圖像塊的偏轉(zhuǎn)角度和面積，并以此調(diào)整模板,優(yōu)化模板配準(zhǔn)，提高模板匹配的準(zhǔn)確性，同時避免使用多尺度模板進行多次匹配運算，提高算法效率。提出利用候選人臉圖像區(qū)域和模板質(zhì)心作為配準(zhǔn)的原點,抑制人臉圖像噪聲的干擾。最后構(gòu)建了基于膚色分割和模板驗證的人臉檢測試驗系統(tǒng)，并對該系統(tǒng)采用自制人臉圖像數(shù)據(jù)庫進行測試。實驗結(jié)果表明，系統(tǒng)算法是有效的，具有較高的檢測性能和低的誤判率。

二、源代碼

function varargout = untitled(varargin) % UNTITLED MATLAB code for untitled.fig % ? ? ?UNTITLED, by itself, creates a new UNTITLED or raises the existing % ? ? ?singleton*. % % ? ? ?H = UNTITLED returns the handle to a new UNTITLED or the handle to % ? ? ?the existing singleton*. % % ? ? ?UNTITLED('CALLBACK',hObject,eventData,handles,...) calls the local % ? ? ?function named CALLBACK in UNTITLED.M with the given input arguments. % % ? ? ?UNTITLED('Property','Value',...) creates a new UNTITLED or raises the % ? ? ?existing singleton*. ?Starting from the left, property value pairs are % ? ? ?applied to the GUI before untitled_OpeningFcn gets called. ?An % ? ? ?unrecognized property name or invalid value makes property application % ? ? ?stop. ?All inputs are passed to untitled_OpeningFcn via varargin. % % ? ? ?*See GUI Options on GUIDE's Tools menu. ?Choose "GUI allows only one % ? ? ?instance to run (singleton)". % % See also: GUIDE, GUIDATA, GUIHANDLES % Edit the above text to modify the response to help untitled % Last Modified by GUIDE v2.5 19-May-2020 15:25:06 % Begin initialization code - DO NOT EDIT gui_Singleton = 1; gui_State = struct('gui_Name', ? ? ? mfilename, ... ? ?'gui_Singleton', ?gui_Singleton, ... ? ?'gui_OpeningFcn', @untitled_OutputFcn, ... ? ?'gui_OutputFcn', ?@untitled_OutputFcn, ... ? ?'gui_LayoutFcn', ?[] , ... ? ?'gui_Callback', ? []); if nargin && ischar(varargin{1}) ? ?gui_State.gui_Callback = str2func(varargin{1}); end if nargout ? ?[varargout{1:nargout}] = gui_mainfcn(gui_State, varargin{:}); else ? ?gui_mainfcn(gui_State, varargin{:}); end % End initialization code - DO NOT EDIT % --- Executes just before untitled is made visible. function untitled_OutputFcn(hObject, eventdata, handles, varargin) % This function has no output args, see OutputFcn. % 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 ? command line arguments to untitled (see VARARGIN) % Choose default command line output for untitled handles.output = hObject; % Update handles structure guidata(hObject, handles); % UIWAIT makes untitled wait for user response (see UIRESUME) % uiwait(handles.figure1); % --- Outputs from this function are returned to the command line. %-------------------------------------- pushbutton1_Callback(回調(diào)函數(shù)) % --- Executes on button press in pushbutton1. function pushbutton1_Callback(hObject, eventdata, handles) global im imycbcr skin1 skin2 lpf %----------------宣布為全域變量 % hObject handle to pushbutton1 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) if get (gcbo, 'Value' ) ==1; ? ?im=imread('10.jpg' ) ; %------------------------------讀入彩色圖像 ? ?axes(handles.axes1) ? ?set(handles.axes1, 'XMinorTick' , 'on' ) %-------------嵌入Axes(1) 回調(diào)函數(shù) ? ?imshow(im) ; %------------------------------顯示彩色圖像 else ? ?imformats end % Hint: get(hObject, ' Value' ) returns push state of togglebutton1 %-------------------------------------- pushbutton2_Callback(回調(diào)函數(shù)) % --- Executes on button press in pushbutton2. function pushbutton2_Callback(hObject, eventdata, handles) % hObject handle to pushbutton2 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) global im imycbcr skin1 skin2 lpf %------ ? ?cr = filter2(lpf, cr) ; ? ?cr = reshape(cr, 1, prod(size(cr) ) ) ; ? ?bmean = mean(cb) ; %--------------------------------求平均值 ? ?rmean = mean(cr) ; ? ?brcov = cov(cb, cr) ; ? ?skin1 = zeros(dim(1) , dim(2) ) ; ? ?for i = 1: dim(1) ? ? ? ?for j = 1: dim(2) ? ? ? ? ? ?cb = double(imycbcr(i, j, 2) ) ; ? ? ? ? ? ?cr = double(imycbcr(i, j, 3) ) ; ? ? ? ? ? ?x =[(cb-bmean);(cr-rmean)] ; ? ? ? ? ? ?skin1(i,j) = exp(-0.5*x'*inv(brcov)*x) ; %---------------------------------計算任意像素為皮膚的概率 ? ? ? ? ?skin1 = skin1./max(max(skin1) ) ; ? ?axes(handles. axes3) ? ?set(handles. axes3, 'XMinorTick' , 'on' ) %-------------嵌入Axes(3) 回調(diào)函數(shù) ? ?imshow(skin1) ; %------------------------------顯示皮膚概率圖像 else ? ?imformats end % Hint: get(hObject, ' Value' ) returns push state of pushbutton3 %-------------------------------------- pushbutton4_Callback(回調(diào)函數(shù)) % --- Executes on button press in pushbutton4. function pushbutton4_Callback(hObject, eventdata, handles) % hObject handle to pushbutton4 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) global im imycbcr skin1 skin2 lpf %----------------: 0.05 ? ? ? ?skin2 = zeros(size(skin1,1),size(skin1,2)) ; ? ? ? ?skin2(find(skin1>threshold)) = 1; ? ? ? ?change = sum(sum(skin2 - previousSkin2) ) ; ? ? ? ?changelist = [changelist change] ; ? ? ? ?previousSkin2 = skin2; ? ?end ? ?%----------------------------------找出最佳門限值 ? ?[C, I] = min(changelist) ; ? ?optimalThreshold = (7-I) *0.1; ? ?skin2 = ones(size(skin1,1),size(skin1,2)) ; ? ?skin2(find(skin1>optimalThreshold))= 0; ? ?skin2 = filter2(lpf, skin2) ; %------------------------中值濾波 ? ?axes(handles.axes5) ? ?set(handles. axes5, 'XMinorTick','on' ) %-------------嵌入Axes(4) 回調(diào)函數(shù) ? ?imshow(skin2); %------------------------------顯示皮膚二值化圖像 else ? ?imformats end % Hint: get(hObject, ' Value' ) returns push state of pushbutton4 %-------------------------------------- pushbutton5_Callback(回調(diào)函數(shù)) % --- Executes on button press in pushbutton5. function pushbutton5_Callback(hObject, eventdata, handles) % hObject handle to pushbutton5 (see GCBO) % eventdata reserved - to be defined in a future version of MATLAB % handles structure with handles and user data (see GUIDATA) global im imycbcr skin1 skin2 lpf %----------------宣布為全域變量 if get (gcbo, 'Value' ) ==1; ? ?sumarea = bwarea(skin2) ; %-----------計算對象的總面積， 為利用面積進行判別做準(zhǔn)備 ? ?[L,numobj] =bwlabel(skin2, 8) ; ? ?avearea = sumarea/numobj; %-----------計算出對象的平均面積 ? ?A = zeros(4, numobj) ;%179 ? ?n = 1; ? ? ? ? ?% ------------------------獲得一副只包含該區(qū)域的圖像， 讓圖像其他區(qū)域為黑色 ? ? ? ?bwsegment = bwselect(skin2, y, x, 8) ; ? ? ? ?% --------------------- 計算出該區(qū)域內(nèi)分割塊的數(shù)目 ? ? ? ?[L, numobjs] = bwlabel(bwsegment, 4) ; ? ? ? ?% ---------------------------------得到區(qū)域孔的數(shù)目 ? ? ? ?numfeatures = bweuler(bwsegment, 4) ; ? ? ? ?numholes = 1 - numfeatures; ? ? ? ?% --------------------------------掃描得出區(qū)域坐標(biāo) ? ? ? ?top = -1; ? ? ? ?buttom =-1; ? ? ? ?left = -1; ? ? ? ?right = -1; ? ? ? ?dim=size(bwsegment); ? ? ? ?for i=1:dim(1) ? ? ? ? ? ?for j=1:dim(2) ? ? ? ? ? ? ? ?if(bwsegment(i,j) ~= 0) ? ? ? ? ? ? ? ? ? ?top=i; ? ? ? ? ? ? ? ? ? ?break; ? ? ? ? ? ? ? ?end; ? ? ? ? ? ?end; ? ? ? ? ? ?if (top ~= -1) break; end; ? ? ? ?end;

三、運行結(jié)果

?