matlab-(含教程)基于matlab的指纹识别算法仿真与分析

%clc,clear,close all; function w=tuxiangyuchuli(image) f=imread(image); f=imresize(f,[363,312]); figure;imshow(f); gray=double(rgb2gray(f)); figure;imshow(uint8(gray)); [m,n]=size(gray); %归一化，灰度值限制在某一范围 M=0;var=0; %均值 for x=1:m for y=1:n M=M+gray(x,y); end end M1=M/(m*n); %方差 for x=1:m for y=1:n var=var+(gray(x,y)-M1).^2; end; end; var1=var/(m*n); for x=1:m for y=1:n if gray(x,y)>M1 gray(x,y)=150+sqrt(2000*(gray(x,y)-M1)/var1); else gray(x,y)=150-sqrt(2000*(M1-gray(x,y))/var1); end end end figure;imshow(uint8(gray)); %分割 分成多个3*3的块大小 M=3; H=floor(m/M);L=floor(n/M); aveg1=zeros(H,L); var1=zeros(H,L); %计算每一块的平均值 for x=1:H for y=1:L aveg=0;var=0; %每一块的均值 for i=1:M for j=1:M aveg=gray(i+(x-1)*M,j+(y-1)*M)+aveg; end; end; aveg1(x,y)=aveg/(M*M); %每一块的方差值 for i=1:M for j=1:M var=(gray(i+(x-1)*M,j+(y-1)*M)-aveg1(x,y)).^2+var; end; end; var1(x,y)=var/(M*M); end; end; %所有块的平均值和方差 Gmean=0;Vmean=0; for x=1:H for y=1:L Gmean=Gmean+aveg1(x,y); Vmean=Vmean+var1(x,y); end end Gmean1=Gmean/(H*L); Vmean1=Vmean/(H*L); %每一小块和整块相比，再次求均值方差 % 前景（黑色） gtemp=0;gtotle=0;vtotle=0;vtemp=0; for x=1:H for y=1:L if Gmean1>aveg1(x,y) gtemp=gtemp+1; gtotle=gtotle+aveg1(x,y); end if Vmean1<var1(x,y) vtemp=vtemp+1; vtotle=vtotle+var1(x,y); end end end % 前景均值 G1=gtotle/gtemp; % 前景方差 V1=vtotle/vtemp; %再次与刚刚产生的值相比 % 求得背景（白色）均值方差 增加可靠性 gtemp1=0;gtotle1=0;vtotle1=0;vtemp1=0; for x=1:H for y=1:L if G1<aveg1(x,y) gtemp1=gtemp1+1; gtotle1=gtotle1+aveg1(x,y); end if 0<var1(x,y)<V1 vtemp1=vtemp1+1; vtotle1=vtotle1+var1(x,y); end end end % 背景均值 G2=gtotle1/gtemp1; % 背景方差 V2=vtotle1/vtemp1; %构建矩阵（H*L） e=zeros(H,L); for x=1:H for y=1:L if aveg1(x,y)>G2 && var1(x,y)<V2 % 背景 e(x,y)=1; end % 前景中的更接近黑色的变为白色 if aveg1(x,y)<G1-100 && var1(x,y)<V2 e(x,y)=1; end end end %该点八邻域小于四为0 for x=2:H-1 for y=2:L-1 if e(x,y)==1 if e(x-1,y) + e(x,y+1)+e(x+1,y+1)+e(x-1,y+1)+e(x+1,y)+e(x+1,y-1)+e(x,y-1)+e(x-1,y-1) <=4 e(x,y)=0; end end end end %构建m*m矩阵 Icc=ones(m,n); for x=1:H for y=1:L if e(x,y)==1 for i=1:M for j=1:M gray(i+(x-1)*M,j+(y-1)*M)=G1; Icc(i+(x-1)*M,j+(y-1)*M)=0; end end end end end figure,imshow(uint8(gray)); figure,imshow(Icc); %找指纹脊线方向并二值化 temp=(1/9)*[1,1,1;1,1,1;1,1,1];%模板系数 均值滤波 Im=gray; In=zeros(m,n); for a=2:m-1 for b=2:n-1 In(a,b)=Im(a-1,b-1)*temp(1,1)+Im(a-1,b)*temp(1,2)+Im(a-1,b+1)*temp(1,3)+Im(a,b-1)*temp(2,1)... +Im(a,b)*temp(2,2)+Im(a,b+1)*temp(2,3)+Im(a+1,b-1)*temp(3,1)+Im(a+1,b)*temp(3,2)+Im(a+1,b+1)*temp(3,3); end end gray=In; Im=zeros(m,n); %求八个方向每个方向的和 for x=5:m-5 for y=5:n-5 %0-7方向的和 sum1=gray(x,y-4)+gray(x,y-2)+gray(x,y+2)+gray(x,y+4); sum2=gray(x-2,y+4)+gray(x-1,y+2)+gray(x+1,y-2)+gray(x+2,y-4); sum3=gray(x-2,y+2)+gray(x-4,y+4)+gray(x+2,y-2)+gray(x+4,y-4); sum4=gray(x-2,y+1)+gray(x-4,y+2)+gray(x+2,y-1)+gray(x+4,y-2); sum5=gray(x-2,y)+gray(x-4,y)+gray(x+2,y)+gray(x+4,y); sum6=gray(x-4,y-2)+gray(x-2,y-1)+gray(x+2,y+1)+gray(x+4,y+2); sum7=gray(x-4,y-4)+gray(x-2,y-2)+gray(x+2,y+2)+gray(x+4,y+4); sum8=gray(x-2,y-4)+gray(x-1,y-2)+gray(x+1,y+2)+gray(x+2,y+4); sumi=[sum1,sum2,sum3,sum4,sum5,sum6,sum7,sum8]; %最大值 summax=max(sumi); %最小值 summin=min(sumi); %和 &&平均值 summ=sum(sumi); b=summ/8; if(summax+summin+4*gray(x,y))> (3*b) sumf=summin; else sumf=summax; end if sumf>b Im(x,y)=128; else Im(x,y)=255; end end end % imshow(Im); %两个矩阵点乘 for i=1:m for j=1:n Icc(i,j)=Icc(i,j)*Im(i,j); end end %转换为二值图 for i=1:m for j=1:n if (Icc(i,j)==128) Icc(i,j)=0; else Icc(i,j)=1; end end end figure;imshow(double(Icc)); title('Icc'); %去除空洞和毛刺 u=Icc; for x=2:m-1 for y=2:n-1 if u(x,y)==0 %该点的4邻域点（上下左右） 如果三个或以上都是白点（1）则该点为毛刺 if u(x,y-1)+u(x-1,y)+u(x,y+1)+u(x+1,y)>=3 u(x,y)=1; end else u(x,y)=u(x,y); end end end figure;imshow(u); title('去除毛刺'); %去除空洞 for a=2:m-1 for b=2:n-1 if u(a,b)==1 %寻找端点 if abs(u(a,b+1)-u(a-1,b+1))+abs(u(a-1,b+1)-u(a-1,b))+abs(u(a-1,b)-u(a-1,b-1))... +abs(u(a-1,b-1)-u(a,b-1))+(abs(u(a,b-1)-u(a+1,b-1)))+abs(u(a+1,b-1)-u(a+1,b))... +abs(u(a+1,b)-u(a+1,b+1))+abs(u(a+1,b+1)-u(a,b+1))~=1 if (u(a,b+1)+u(a-1,b+1)+u(a-1,b))*(u(a,b-1)+u(a+1,b-1)+u(a+1,b))+(u(a-1,b)+u(a-1,b-1)+u(a,b-1))... *(u(a+1,b)+u(a+1,b+1)+u(a,b+1))==0 %去除空洞 u(a,b)=0; end end end end end figure;imshow(u); title('去除空洞'); %图像细化 v=~u; figure;imshow(v); se=strel('square',3); %对图像进行开闭操作 fo=imopen(v,se); figure;imshow(fo); % 先腐蚀后膨胀，作用是：可以使边界平滑，消除细小的尖刺，断开窄小的连接,保持面积大小不变 title('开运算') v=imclose(fo,se); figure;imshow(v); title('闭运算') w=bwmorph(v,'thin',Inf);%对图像进行细化 figure;imshow(w); title('细化图'); imshow(Icc)