数字图像打印.docx
《数字图像打印.docx》由会员分享，可在线阅读，更多相关《数字图像打印.docx（19页珍藏版）》请在冰豆网上搜索。
数字图像打印
（1）名词解释
RGBRedGreenBlue，红绿蓝三原色
CMYKCyanMagentayellowblacK,青、品红、黄、黑，用于印刷的四分色
HISHorizontalSituationIndicator水平位置指示器
FFTFastFourierTransformAlgorithm（method）快速傅氏变换算法
CWTcontinuouswavelettransform连续小波变换
DCTDiscreteCosineTransform离散余弦变换
DWTDiscreteWaveletTransform离散小波变换
CCDChargeCoupledDevice电荷耦合装置
Pixel:
adigitalimageiscomposedofafinitenumberofelements,eachofwhichhasaparticularlicationandvalue,theseelementsarecalledpixel像素
DCcomponentinfrequencydomain频域直流分量
GLH GrayLevelHistogram灰度直方图
Mather（basic）wavelet：
afunction（wave）usedtogenerateasetofwavelets,母小波，用于产生小波变换所需的一序列子小波
Basisfunctionsbasisimage:
thereisonlyonesetofαkforanygivenf（x）,thentheψk（x）arecalledbasisfunctions基函数基图像
Multi-scaleanalysis多尺度分析
Gaussianfunction:
isafunctionoftheform:
forsomerealconstantsa0,b,c0,ande≈2.718281828（Euler’snumber）.对于一些真正的常量0,b,c0,和e≈2.718281828（欧拉数）。
高斯函数
sharpeningfilter锐化滤波器
Smoothingfilter/convolution平滑滤波器/卷积
smoothing filter are used for blurring and for noise reduction平滑滤波器用于模糊处理和降低噪声/卷积
Imageenhancement/imagerestoration图像增强和图像恢复
空间域滤波Spatialdomainfiltering：
频率域滤波Frequencydomainfiltering：
Frequencydomainfilteringwithavariablefrequencyforthesignalfiltering以频率作为变量对信号进行滤波
空间分辨率：
spatialresolutionisameasureofthesmallestdiscernibledetailinanimage.图像中可辨别的最小细节的度量
灰度分辨率：
Intensityresolutionreferstothesmallestdiscerniblechangeinintensitylevel.灰度分辨率是指在灰度级中可分辨的最小变化
取样sampling:
Digitizingthecoordinatevaluesiscalledsampling.对坐标值进行数字化
量化quantization：
Digitizingtheamplitudevaluesiscalledquantization.对幅值数字化
图像压缩：
Imagecompression,theartandscienceofreducingtheamountofdatarequiredtorepresentanimage.图像压缩是一种减少描绘一幅图像所需数据量的技术和科学.
（2）问答题
1.Citeoneexampleofdigitalimageprocessing
Answer:
Inthedomainofmedicalimageprocessingwemayneedtoinspectacertainclassofimagesgeneratedbyanelectronmicroscopetoeliminatebright,isolateddotsthatarenointerest.
2.Citeoneexampleofspatialoperation举一个空间操作的例子
Answer:
Inthedomainofmedicalimageprocessingwemayneedtoinspectacertainclassofimagesgeneratedbyanelectronmicroscopetoeliminatebright,isolateddotsthatarenointerest.
3.Citeoneexampleoffrequencydomainoperationfromthefollowingprocessingresult,makeageneralcommentaboutidealhighpassfilter（figureB）andGaussianhighpassfilter（figureD）
A.Originalimage
B.idealhighpassfilter
Incontrasttotheideallowpassfilter,itistoletallthesignalsabovethecutofffrequencyfcwithoutloss,andtomakeallthesignalsbelowthecutofffrequencyofFCwithoutlossof.
C.theresultofidealhighpassfilter
D.Gaussianhighpassfilter
Highpassfilter,alsoknownas"lowresistancefilter",itisaninhibitoryspectrumofthelowfrequencysignalandretainhighfrequencysignalmodel（ordevice）.Highpassfiltercanmakethehighfrequencycomponents,whilethehigh-frequencypartofthefrequencyintheimageofthesharpchangeinthegrayarea,whichisoftentheedgeoftheobject.Sohighpassfiltercanmaketheimagegetsharpeningprocessing
E.TheresultofGaussianfilter
3.Theoriginalimage,theideallowpassfilterandGaussianlowpassfilterareshownbelowBndC.DandEaretheresultoftheeitherfilterBorC
A.Drawlinestoconnectthefilterwiththeirresult
B.Explainthedifferenceofthetwofilters
Duetoexcessivecharacteristicsoftheideallow-passfiltertoofastJun,itwillproducearingingphenomenon.OvercharacteristicsofGaussfilterisveryflat,soitisnotringing
4.Whatistheresultwhenapplyinganaveragingmaskwiththesize1X1?
Nochange
5.StatetheconceptoftheNyquistsamplingtheoremfromthefigurebelovy
Thelawofsamplingprocessshouldbefollowed,alsocalledthesamplingtheoremandthesamplingtheorem.Thesamplingtheoremshowstherelationshipbetweenthesamplingfrequencyandthesignalspectrum,anditisthebasicbasisofthecontinuoussignaldiscretization.Inanalog/digitalsignalconversionprocess,whenthesamplingfrequencyfs.maxgreaterthan2timesthehighestfrequencypresentinthesignalFmaxfs.max>2fmax,samplingdigitalsignalcompletelyretainedtheinformationintheoriginalsignal,thegeneralpracticalapplicationassurancesamplingfrequencyis5~10timeshigherthanthatofthesignalofthehighfrequency;samplingtheorem,alsoknownastheNyquisttheorem
6.Ameanfilterisalinearfilterbutamedianfilterisnot,why?
Thebasicprincipleoflinearfilteringistoreplacetheoriginalimagewiththemeanvalueofeachpixel,butmedianfilterreplacetheoriginalimagewiththemedianvalueofeachpixel.Thevalueofmeanandmedianisdifferent.
7.Fundamental Steps in images Digital image Processing 数字图像图像处理的基本步骤
image acquisition—>image enhancement—>image restoration—>Color image processing—>wavelets—>compression（压缩）—>morphological processing（形态学理）—>segmentation（分割）—>representation and description（表示与描述）—>recognition（识别）
8.WiththechromaticitydiagrambellowgiveabriefdescriptiontotheRGBcolormodel.Andthesethreecolorsenoughtocomposeallvisiblecolors?
Answer：
ImagesrepresentedintheRGBcolormodelconsistofthreecomponentimages,oneforeachprimarycolor.
Thesethreecolorsenoughtocomposeallvisiblecolors
（3）算法题
1.ThefollowingmatrixAisa3*3imageandBis3*3Laplacianmask,whatwillbetheresultingimage?
（Notethattheelementsbeyondtheborderremainunchanged）
2.DevelopanalgorithmtoobtaintheprocessingresultBfromoriginalimageA
3.Developanalgorithmwhichcomputesthepseudocolorimageprocessingbymeansoffouriertramsform
Answer：
Thestepsoftheprocessareasfollow:
（1）Multiplytheinputimagef（x,y）by（-1）x+ytocenterthetransform;
（2）ComputetheDFToftheimagefrom
（1）togetpowerspectrumF（u,v）ofFouriertransform.
（3）Multiplybyafilterfunctionh（u,v）.
（4）ComputetheinverseDFToftheresultin（3）.
（5）Obtaintherealpartoftheresultin（4）.
（6）Multiplytheresultin（5）by（-1）x+y
4.Developanalgorithmtogenerateapproximationimageseriesshowninthefollowingfigureb**meansofdownsampling
5.DevelopanalgorithmwhichimplementsfrequencydomainfilteringbymeansofFouriertransform.
Answer：
Thestepsoftheprocessareasfollow:
（1）Multiplytheinputimagef（x,y）by（-1）x+ytocenterthetransform;
（1）将输入图像f（x，y）的（-1）x+y为中心的变换；
（2）ComputetheDFToftheimagefrom
（1）togetpowerspectrumF（u,v）ofFouriertransform.
计算图像的DFT从
（1）得到的功率谱f（u，v）的傅里叶变换。
Multiplybyafilterfunctionh（u,v）
乘以一个滤波器函数h（u,v）.
ComputetheinverseDFToftheresultin（3）.
计算（3）中的结果DFT的逆
Obtaintherealpartoftheresultin（4）.
获得（4）结果中的实部
Multiplytheresultin（5）by（-1）x+y
（-1）x+y乘以（5）中的结果.
5.Losslessapproaches—HoffmanCoding无损方法-霍夫曼编码
步骤：
（1）createofsourcereductionsbyorderingthesymbolsunderconsiderationandcombiningthelowestprobabilitysymbolsintoasinglesymbolsthatreplacestheminthenextsourcereduction.
（2）Codeeachreducedsource,startingwiththesmallestsourceandworkingbacktotheoriginalsource.
（4）编程题
1）Therearetwosatellitephotosofnightasblew.WriteaprogramwithMATLABtotellwhichisbrighter
代码：
A=imread（’1.jgp’）;
B=imread（‘2.jpg’）;
[m,n]=size（A）;
fori=1:
m
forj=1:
n
sum1=sum1+A[I,j];
end
end
avg1=sum1/m*n;
[r,c]=size（B）;
fori=1:
m
forj=1:
n
sum2=sum2+B[I,j];
end
end
avg2=sum2/m*n;
2）An8*8imagef（i,i）hasgraylevelsgivenbythefollowingequation:
f（i,i）=|i-j|,i,j=0,1….,7
Writeaprogramtofindtheoutputimageobtainedbyapplyinga3*3medianfilterontheimagef（i,j）;notethattheborderpixelsremainunchanged.
Ansewr:
function[r]=avgfilter（gray,n）
a（1:
n,1:
n）=1;
[row,col]=size（gray）;
gray1=double（gray）;
gray2=gray1;
fori=1:
row-n+1
forj=1:
col-n+1
c=gray1（i:
i+（n-1）,j:
j+（n-1））.*a;
s=sum（sum（c））;
gray2（i+（n-1）/2,j+（n-1）/2）=s/（n*n）;
end
end
r=uint8（gray2）;
>>avg3=avgfilter（noise,3）;
>>avg5=avgfilter（noise,5）;
>>avg7=avgfilter（noise,7）;
>>subplot（221）;imshow（noise）;title（'原噪声图'）;
>>subplot（222）;imshow（avg3）;title（'3*3均值滤波图'）;
>>subplot（223）;imshow（avg5）;title（'5*5均值滤波图'）;
>>subplot（224）;imshow（avg7）;title（'7*7均值滤波图'）;
1．DesignanadaptivelocalnoisereductionfilterandapplyittoanimagewithGaussiannoise.Comparetheperformanceoftheadaptivelocalnoisereductionfilterwitharithmeticmeanandgeometricmeanfilter.
Answer:
clear
closeall;
rt=imread（'E:
\数字图像处理\yy.bmp'）;
gray=rgb2gray（rt）;
subplot（2,3,1）;imshow（rt）;
title（'原图像'）;
subplot（2,3,2）;imshow（gray）;
title（'原灰度图像'）;
rtg=im2double（gray）;
rtg=imnoise（rtg,'gaussian',0,0.005）%加入均值为0，方差为0.005的高斯噪声
subplot（2,3,3）;imshow（rtg）;
title（'高噪点处理后的图像'）;
[a,b]=size（rtg）;
n=3;
smax=7;
nrt=zeros（a+（smax-1）,b+（smax-1））;
fori=（（smax-1）/2+1）:
（a+（smax-1）/2）
forj=（（smax-1）/2+1）:
（b+（smax-1）/2）
nrt（i,j）=rtg（i-（smax-1）/2,j-（smax-1）/2）;
end
end
figure;
imshow（nrt）;
title（'扩充后的图像'）;
nrt2=zeros（a,b）;
fori=n+1:
a+n
forj=n+1:
b+n
form1=3:
2
m2=（m1-1）/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））;
A1=Zmed-Zmin;
A2=Zmed-Zmax;
if（A1>0&&A2<0）
B1=nrt2（i,j）-Zmin;
B2=nrt2（i,j）-Zmax;
if（B1>0&&B2<0）
nrt2（i,j）=nrt2（i,j）;
else
nrt2（i,j）=Zmed;
end
continue;
end
end
end
end
nrt3=im2uint8（nrt2）;
figure;
imshow（nrt3）;
title（'自适应中值滤波图'）;
2.ImplementWienerfilterwith“wiener2”functionofMatLabtoanimagewithGaussiannoiseandcomparetheperformancewithadaptivelocalnoisereductionfilter.
代码如下：
>>I=imread（'E:
\数字图像处理\yy.bmp'）;
>>J=rgb2gray（I）;
>>K=imnoise（J,'gaussian',0,0.005）;
>>L=wiener2（K,[55]）;
升级会员 