数字图像(实验四)

《数字图像(实验四)》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。

1、福建农林大学计算机与信息学院实验报告数字图像处理(双语)课程名称：(DigitalImageProcessing)姓名：系：计算机科学与技术专业：计算机科学与技术年级：指导教师：王长缨职称：副教授2014〜2015学年下学期福建农林大学计算机与信息学院实验报告系：计算机科学与技术专业:计算机科学与技术年级：2012级姓名：学号：实验室号田家炳510计算机号实验时间：2015年5月日指导教师签字：成绩：ExperimentIVNoisereductionwithadaptivefiltering(validated)(2periods)ObjectiveEsta

2、blishthedatadrivenideafromadaptivefilterinbothspatialandfrequencydomain.MethodologyApplyspatialadaptivefilterandWienerfiltertoanoisedphotorespectivelyandmakeacommentbycomparisonoftwomethods.EnvironmentHardware:GeneralPCOperatingsystem:WindowsXFCalculatingplatform:MATLAB6.5orhigherve

3、rsionExperimentindetail:1.DesignanadaptivelocalnoisereductionfilterandapplyittoanimagewithGaussiannoise.Comparetheperformanceoftheadaptivelocalnoisereductionfilterwitharithmeticmeanandgeometricmeanfilter.代码：rt=imreadCD:tupian1.jpg’)；gray=rgb2gray(rt);subplot(2,3,1);imshow(rt);title

4、(fOriginalpicture1);subplot(2,3,2);imshow(gray);title(’Graypicture’)；rtg=im2double(gray);rtg=imnoise(rtg，’gaussian,0.005)%加入均值为0,方差为0.005的高斯噪声subplot(2,3,3);imshow(rtg);tide(’Gaussianpicture’)；[a，b]=size(rtg);n=3;smax=7;nrt=zero$(a+($max-1)，b+(smax-1));fori=(($max-l)/2+1):(a+($max

5、-l)/2)forj=((smax-l)/2+1):(b+(smax-l)/2)nrt(ij)=rtg(i-(smax-l)/2,j-(smax-l)/2);endendfigure;imshow(nrt);title(’扩充后的图像i);nrt2=zeros(a，b);fori=n+l:a+nforj=n+l:b+nforml=3:2m2=(ml-l)/2;C=nrt2(i-m2:i+m2，j-m2:j+m2);%使用7*7的滤波器Zmed=median(median(c));Zmin=min(min(c));Zmax=max(max(c));Al=Zmed

6、-Zmin;A2=Zmed-Zmax;if(Al>0&&A2<0)Bl=nrt2(i,j)-Zmin;B2=nrt2(i,j)-Zmax;if(Bl>0&&B2<0)nrt2(i，j)=nrt2(i,j);elsenrt2(ij)=Zmed;endcontinue;endendendendnrt3=im2uint8(nrt2);figure;imshow(nrt3);titlefadaptivemedianfilter’)；扩充后的图像adaptivemedianfilter2.ImplementWienerfilterwithuwiener2^functio

7、nofMatLabtoanimagewithGaussiannoiseandcomparetheperformancewithadaptivelocalnoisereductionfilter.2-3-4-5-6_7-8-9-10-11-12_原始图像萵斯喑声后的图像对加晬图像进行二维自适应维也纳•悉波3*3对加眭图傑进行二绺自适应维也纳波波对加哮图像逬行二维自适应维也纳波波9,9I=imread(?D:tupianl.jpg');H=rgb2gray(I);J=imnoise(H/gaussian’0.1，0.002);K=wiener2(J,[3,3])

8、;N=wiener2(J,[5,5])