1、圖像處理與模式識(shí)別作業(yè)4- 圖像 濾波姓名：班級(jí)：自動(dòng)化學(xué)院：電信學(xué)院聯(lián)系方式： 日期：2012/4/6【運(yùn)行環(huán)境：vs平臺(tái)opencv】分別用高斯濾波器和中值濾波器去平滑測(cè)試圖像test1和2，模板大小分別是3x3 ， 5x5 ，7x7 ;1). matlab高斯濾波器程序代碼：si=3;ss=(si+1)/2;sig2=1.5*1.5;mins=exp(-2.0*ss*ss/(2*sig2);%令最小項(xiàng)系數(shù)為1he=0;%整數(shù)系數(shù)和guass1=ones(si);for i=1:si for j=1:si temp=(i-ss)*(i-ss)+(j-ss)*(j-ss); guass1(i

2、,j)=round(exp(-temp/(2*sig2)/mins); he=he+guass1(i,j); endendI=imread(F:/圖像處理作業(yè)/第四次作業(yè)/第四次作業(yè)/test2.tif);h,w,x=size(I);J=I;for i=ss:h+1-ss for j=ss:w+1-ss tempp=double(I(i-ss+1:i+ss-1,j-ss+1:j+ss-1); z=tempp.*guass1; J(i,j)=round(sum(sum(z)/he); endendfigure;subplot(121);imshow(I);title(原圖像)subplot(12

3、2);imshow(J);title(高斯濾波后)執(zhí)行結(jié)果：3x3:5x5:以3x3高斯濾波器為例，test1.pgm:test2.tif:2). c中值濾波器程序代碼：si=3;ss=(si+1)/2;I=imread(F:/圖像處理作業(yè)/第四次作業(yè)/第四次作業(yè)/test2.tif);h,w,x=size(I);J=I;for i=ss:h+1-ss for j=ss:w+1-ss tempp=; for k=1:si tempp=tempp,(I(i-ss+1:i+ss-1,j-ss+k:j-ss+k); end tempp=sort(tempp); J(i,j)=tempp(round(

4、si*si+1)/2); endendfigure;subplot(121);imshow(I);title(原圖像)subplot(122);imshow(J);title(中值濾波后)執(zhí)行結(jié)果： test1.pgm（3x3中值濾波器）:test1.pgm（5x5中值濾波器）:test1.pgm（7x7中值濾波器）:test2.tif（3x3中值濾波器）:test2.tif（5x5中值濾波器）:test2.tif（7x7中值濾波器）:2.利用高通濾波器濾波測(cè)試圖像test3,4：包括unsharp masking, Sobel edge detector, and Laplace edge

5、detection ; 1).unsharp maskingMatlab程序代碼：si=3;ss=(si+1)/2;I=imread(F:/圖像處理作業(yè)/第四次作業(yè)/第四次作業(yè)/test4 copy.bmp);h,w,x=size(I);J=I;for i=ss:h+1-ss for j=ss:w+1-ss tempp=double(I(i-ss+1:i+ss-1,j-ss+1:j+ss-1); z=tempp.*1 1 1;1 -9 1;1 1 1; J(i,j)=round(0.1*sum(sum(z)+I(i,j); endendfigure;subplot(121);imshow(I)

6、;title(原圖像)subplot(122);imshow(J);title(unsharp masking):test4.tif:2). Sobel edge detector(閾值取100):Matlab 程序代碼：si=3;ss=(si+1)/2;I=imread(F:/圖像處理作業(yè)/第四次作業(yè)/第四次作業(yè)/test4 copy.bmp);h,w,x=size(I);J=I;for i=ss:h+1-ss for j=ss:w+1-ss tempp=double(I(i-ss+1:i+ss-1,j-ss+1:j+ss-1); te=-1 -2 -1;0 0 0;1 2 1; z1=te

7、mpp.*te; z2=tempp.*te; t1=sum(sum(z1);t2=sum(sum(z2); s=sqrt(t12+t22); if(s50) J(i,j)=255; else J(i,j)=0; end endendfigure;subplot(121);imshow(I);title(原圖像)subplot(122);imshow(J);title(soble edge):test4.tif: 3). Laplace edge detection (閾值取10): Matlab 程序代碼：si=3;ss=(si+1)/2;I=imread(F:/圖像處理作業(yè)/第四次作業(yè)/第四

8、次作業(yè)/test3_corrupt.pgm);h,w,x=size(I);J=I;%sss=;k=1;for i=ss:h+1-ss for j=ss:w+1-ss tempp=double(I(i-ss+1:i+ss-1,j-ss+1:j+ss-1); z=tempp.*0,-1,0;-1,4,-1;0,-1,0; s=sum(sum(z); % sss(k)=s;k=k+1; if(s10) J(i,j)=255; else J(i,j)=0; end endendfigure;subplot(121);imshow(I);title(原圖像)subplot(122);imshow(J);

9、title(Laplace edge detection):test4.tif:3.利用canny算法提取測(cè)試圖像test3,4邊緣：.Matlab程序源碼：clear;clc;i=imread(test4 copy.bmp);%k=rgb2y(i);%獲取h分量，即亮度分量k=double(i);k1=gaosi(k);%橫向?yàn)V波k1=k1;%對(duì)圖像進(jìn)行轉(zhuǎn)置，為下一步縱向?yàn)V波作準(zhǔn)備（縱向?yàn)V波=轉(zhuǎn)置后橫向?yàn)V波）k1=gaosi(k1);k1=k1;%還原%計(jì)算梯度的大小和方向h,w=size(k);for m=2:h-1 for n=2:w-1 zz1=k1(m,n-1)+k1(m+1,n-1

10、); zz2=k1(m,n)+k1(m+1,n); zz3=k1(m,n-1)+k1(m,n); zz4=k1(m+1,n-1)+k1(m+1,n); kp(m,n)=0.5*(zz2-zz1); kq(m,n)=0.5*(zz3-zz4); kfu(m,n)=sqrt(kp(m,n)2)+(kq(m,n)2);%梯度大小 kjiao(m,n)=atan(kq(m,n)/kp(m,n)+0.001); endend%非極大值抑制%首先將梯度方向劃分為4個(gè)方向0,45,90,135（以及他們的反向延長(zhǎng)線）for m=2:h-1 for n=2:w-1 if kjiao(m,n)=3/8*pi k

11、jiao(m,n)=2; else if kjiao(m,n)=1/8*pi kjiao(m,n)=1; else if kjiao(m,n)=-1/8*pi kjiao(m,n)=0; else if kjiao(m,n)=-3/8*pi kjiao(m,n)=3; else kjiao(m,n)=2; end end end end endendk2=k1;for m=2:h-1 for n=2:w-1 if kjiao(m,n)=0 if k1(m,n)k1(m,n-1)&k1(m,n)k1(m,n+1); else k2(m,n)=0; end end if kjiao(m,n)=1

12、if k1(m,n)k1(m+1,n-1)&k1(m,n)k1(m-1,n+1); else k2(m,n)=0; end end if kjiao(m,n)=2 if k1(m,n)k1(m-1,n)&k1(m,n)k1(m+1,n); else k2(m,n)=0; end end if kjiao(m,n)=3 if k1(m,n)k1(m-1,n-1)&k1(m,n)k1(m+1,n+1); else k2(m,n)=0; end end endend%兩次閾值分割k3=k2;%以t1為閾值分割后的矩陣k4=k2;%以t2為閾值分割后的矩陣t1=10;t2=30;for m=2:h-1

13、 for n=2:w-1 if kfu(m,n)t1 k3(m,n)=0; else k3(m,n)=255; end if kfu(m,n)t2 k4(m,n)=0; else k4(m,n)=255; end endendfigure;subplot(221);imshow(i);title(原圖像);subplot(222);imshow(k3,);title(閾值為50的分割圖像);subplot(223);imshow(k4,);title(閾值為100的分割圖像);flag=zeros(h,w);%標(biāo)記該點(diǎn)是否以檢測(cè)過(guò)，1表示檢測(cè)過(guò)for m=2:h-1 for n=2:w-1 if k4(m,n)=0&flag(m,n)=0 k4,flag=findline(k3,k4,flag,m,n); end endend subplot(224);imshow(k4,);title(修正后的分割圖像);imwrite(k4,C:try.jpg);其中findline函數(shù)代碼如下：function ff,flag1=findline(k3,k4,flag,m,n)flag1=flag; m1=m+1;n1=n+1; while(m=m1|n=n1) flagg=0; for i=1:3 if(f