face detection code in image

Please any one suggest how to detect face in a image..in following code its detecting faces but closely standing persons face is not detecting properly..two closely faces is displaying as single face..here i used matlab..not cv toolbox. i need to detect individual faces from image and have to extract..please where is the mistake tell me..take any image having so many faces and check it..

clc; clear all; close all; I=imread('test2.jpg'); I=double(I); R=I(:,:,1); G=I(:,:,2); B=I(:,:,3);

cb = 0.148* I(:,:,1) - 0.291* I(:,:,2) + 0.439 * I(:,:,3) + 128; %calculating cb value using formula cr = 0.439 * I(:,:,1) - 0.368 * I(:,:,2) -0.071 * I(:,:,3) + 128; [w h]=size(I(:,:,1)); for i=1:w for j=1:h if 140<=cr(i,j) && cr(i,j)<=165 && 140<=cb(i,j) && cb(i,j)<=195 segment(i,j)=1; else segment(i,j)=0; % end end end

segment(200:end,1:end)=0; %lower plane masking % I(200:end,1:end,:)=0; I1=rgb2gray(I); % imshow(segment);

[m n]=size(segment); morp=bwareaopen(segment,150); %MORPHOLOGICAL Processing%

imag(:,:,1)=R.*morp; imag(:,:,2)=G.*morp; imag(:,:,3)=B.*morp; % figure(1),imshow(uint8(imag));

RI=imag(:,:,1); GI=imag(:,:,2); BI=imag(:,:,3);

L = bwlabel(morp,8); BB = regionprops(L, 'BoundingBox'); bboxes= cat(1, BB.BoundingBox); widths=bboxes(:,3); heights=bboxes(:,4); hByW=heights./widths;

lenRegions=size(bboxes,1); foundFaces=zeros(1,lenRegions);

rgb=label2rgb(L); %figure,imshow(rgb); %title('face candidates');

%%%%%%%%%%%%%%%% CHECK FACE CRITERIONS %%%%%%%%%%%%%%%%%%%%%%%%%%% % figure(2) arr=[]; x=0; for i=1:lenRegions if (hByW(i)>2 hByW(i)<0.75) continue; end

      if (heights(i)<10 && widths(i)<10)
          continue;
      end

      CurBB=bboxes(i,:);
      XStart=CurBB(1);
      YStart=CurBB(2);
      WCur=CurBB(3);
      HCur=CurBB(4);

      % crop current region%
      rangeY=int32(YStart):int32(YStart+HCur-1);
      rangeX= int32(XStart):int32(XStart+WCur-1);
      RIC11=RI(rangeY, rangeX);
      GIC=GI(rangeY, rangeX);
      BIC=BI(rangeY, rangeX);
      RIC1=imresize(RIC11,[100,100]);

      R1=R(rangeY, rangeX);
      R2=G(rangeY, rangeX);
      R3=B(rangeY, rangeX);
      R11=imresize(R1,[100,100]);
      R22=imresize(R2,[100,100]);
      R33=imresize(R3,[100,100]);

      RIC(1:100,1:100,1)=histeq(uint8(R11));       %histogram equalization   
      RIC(1:100,1:100,2)=histeq(uint8(R22));          
      RIC(1:100,1:100,3)=histeq(uint8(R33)); 

      arr=[arr,RIC];
%       subplot(5,4,i)
%       imshow(uint8(RIC));   
end

figure(3) imshow(arr);

j=0; temp=0; N=length(arr)/100; for i=1:length(arr)/100 temp=temp+arr(1:100,j+1:(i*100),:)./N; %calculating template j=i*100; end temp1=rgb2gray(temp);

[Image Analyst]

And just how are we supposed to be able to test your code?