Dip Practical File

Lab Manual ECS 702 Digital Image Processing

Even Semester 2013 B.ec! "CSE# $II Semester

Prere%uisite & Stu'ent !as (no)le'ge about t!e c an' matlab. !e* s!oul' be able to )rite an' im+lement t!e +rograms in c language )it! turbo c com+ilers. !e basic ,unctional as+ect o, Matlab s!oul' be (no)n b* t!e stu'ents.

-utcomes& ,ter (no)ing all t!e mentione' belo) +rograms eit!er in C or in Matlab t!e stu'ents )ill become to 'o researc! b* a++l*ing 'i,,erent tec!ni%ues o, +roblem i'enti,ication an' t!eir solution )it! innovative met!o'ologies.

List -, Programs "Practicals#

Sl /o. Program /ame

Language

1

C or Matlab

Program in Matlab for digital Negative.

2

C or Matlab A. Program For Mean Filter

B. Program for Median Filter 

C or Matlab

3

Program for Histogram Equiliation

Matlab

!

Program for "o# Pass Filter

Matlab

$

Program for Hig% Pass Filter

Matlab

&

Program for Boundar' E(tra)tion

Matlab

*

Program for +ra%m,s -)an "ine Algorit%m

Matlab



Program For Edge /ete)tion

Matlab

0

Program for "ine /ete)tion

Matlab

Program 1. Im+lement t!e +rogram in Matlab ,or 'igital /egative. clear all, close all

e)%o on  1. load a gra' s)ale image (iind45imread67images89&!.bm97:; n%nv45sie6(i:;  sie of t%e image (5ind2gra'6(iind:;  9i(el value in < 2$$4 figure61: ims%o#6(:title67original image7:  9osi5get6179osition7:;  a 1 ( ! ve)tor gives lo#er left )orner and sie of image set6179osition7< $!< 9osi63=!:4:  >mage negatives= ?6r: 5 "@1@r "52$& (neg52$$@(; figure62:ims%o#6(neg:title67image negatives7:  9osi5get6279osition7:;  a 1 ( ! ve)tor gives lo#er left )orner and sie of image set6279osition71&< $!< 9osi63=!:4:  9ause   "et us no# )onsider some transformation fun)tion  r5<=<.<1=14; s5<.$r6find6r<.2:: <.11.$6r6find6<.2 5 r D r 5 <.*::@<.2:...   1<.$6r6find6r<.*::@1:4; figure6*:sub9lot6131:9lot6rs7@7:title67mid@range stre)%7: set6*79osition72*$ 2&$ *12 10!4:  (15double6(:2$$;  (1 range to < and 1 '56<.$(1:.(1<.246<.11.$6(1@<.2::.<.2 5 (1 D (1 5 <.*4 ... 61<.$6(1@1::.(1  <.*4; (midind'45)munique6':; figure63:ims%o#6(midind':title67mid@range stret)%ed7:  9osi5get6379osition7:;  a 1 ( ! ve)tor gives lo#er left )orner and sie of image set6379osition732< $!< 9osi63=!:4:  9ause ) 5 2$$log1<612$$:;  ) is )%osen to ensure onl' 2$& gra' levels after transform r5<=2$$4; s5)log1<61abs6r::; figure6*:sub9lot6132:9lot6rs7@7:title67logarit%m )om9ression7: a(is6< 2$& < 2$&4: (log5uint6round6)log1<61double6(::::; figure6!:ims%o#6(log:title67log@transformed7:  9osi5get6!79osition7:;  a 1 ( ! ve)tor gives lo#er left )orner and sie of image set6!79osition7!< $!< 9osi63=!:4:

Program 2a. Mean ilter )it! 3  3 mas( 

)lear all; )lose all; a5imread67)ar.G9g7:; i5rgb2gra'6a:; sub9lot6321:; ims%o#6i: title67org image7:; n5imnoise6i7salt D 9e99er7<.<$:;adding salt D 9e99er noise sub9lot6322:; ims%o#6n:; title67#it% noise7: 5medfilt26n:;a99l'ing mean filter  sub9lot6323:; ims%o#6:; title67mean filter7: % 5 fs9e)ial67average7$:;averager of $($ 9i(el dimension %15 uint6filter26%n::;a99l'ing 2/ filter over t%e noise image 7n7 #it% t%e averager 7%7 and t%en )onverting double to uint sub9lot632!:; ims%o#6%1:; title67avg7:; m5#iener26n:;a99l'ing #iener filter  sub9lot632$:; ims%o#6m:; title67#iener7:; rms15sqrt6sum6sum6n@i:.I2::lengt%6n:;a99l'ing root mean square P-NJ1noise52<  log1<62$$rms1:;finding 9ea signal to noise ratio rms25sqrt6sum6sum6i@:.I2::lengt%6n:; P-NJ2mean52<  log1<62$$rms2:; rms35sqrt6sum6sum6i@m:.I2::lengt%6n:; P-NJ3#einer52<  log1<62$$rms3:; rms!5sqrt6sum6sum6i@%1:.I2::lengt%6n:; P-NJavg52<  log1<62$$rms!:;

Program 2b. Me'ian ilter )it! 3  3 mas( 

fun)tion  img2 4 5 MedianFilter6 img : Kidt%Heig%t45sie6img:; img2 5 eros6Kidt%Heig%t:; for ()oord51=1=Kidt% leftbound5()oord@1; rig%tbound5()oord1; if leftbound1 leftbound51; end if rig%tboundKidt% rig%tbound5Kidt%; end for ')oord51=1=Heig%t to9bound5')oord@1;  bottombound5')oord1; if to9bound1 to9bound51; end if bottomboundHeig%t  bottombound5Heig%t; end arrsie 5 6bottombound@to9bound:6rig%tbound@leftbound:; arr 5 eros6arrsie:; (inde(5leftbound; 'inde(5bottombound; inde(51; #%ile inde(arrsie arr6inde(:5img26(inde('inde(:; (inde(5(inde(1; if (inde( rig%tbound (inde(51; 'inde(5'inde(1; end if 'inde(bottombound  brea  end end sortedarr 5 -ele)tion-ort6arr:; img26()oord')oord1:5sortedarr6)eil66arrsie:2::; end end end

fun)tion  numbers 4 5 -ele)tion-ort6 numbers :

lengt%5sie6numbers:; for i 5 lengt%= @1= 1 ma( 5 i; for G 5 i@1=@1=1 if numbers6ma(:  numbers6G: ma( 5 G; end end tem9 5 numbers6ma(:; numbers6ma(: 5 numbers6i:; numbers6i: 5 tem9; end end

Program 3. Plot t!e regular !istogram o, M  / Image o, 3 bits in Matlab. lso e%ualie an' +lot t!e lat !istogram. )lear all )lose all load 9&!int.t(t;  &! ( &! image " 5 2$& (<59&!int; "52$&; mn45sie6(<:; len5mn; (5res%a9e6(<len1:; (9df5%ist6(<="@14:;  9df 1 ( " tr5round6(9dftriu6ones6"::6"@1:len:;  )df range from < to "@1 '<5eros6mn:; for i51=" if (9df6i:<   '<5'<(<55i@14tr6i:; end end '9df5%ist6res%a9e6'<len1:<="@14:;  9df of ' 1 ( " figure61:sub9lot6211:stem6<="@14(9df:title67%istogram original7: a(is6< 2$& < $<<4: sub9lot6212:stem6<="@14'9df:title67%istogram equalied7:a(is6< 2$& < $<<4:  figure62:sub9lot6121:image6(<:)olorma967gra'7:  sub9lot6122:image6'<:)olorma967gra'7: figure62:sub9lot6121:ims%o#6uint6(<::title67before7: sub9lot6122:ims%o#6uint6'<::title67after7: figure63: stairs6<="@14tr:title67transformation7:a(is6< 2$& < 2$&4:

 Algorit%m=  1. )om9ute t%e P/F and t%en C/F 6?6r::of (<  2. 9oint@#ise gra'@level transform a))ording to ?6r: if nargin3 mode5<; end  default not 9lotting figures if nargin2 "52$&; end  default "52$&

 bins5<="@14; mn45sie6(<:; len5mn; (5res%a9e6(<len1:; (9df5%ist6(bins:;  9df 1 ( " tr5round6(9dftriu6ones6"::6"@1:len:;  )df range from < to "@1 '<5eros6mn:; for i51=" if (9df6i:<   '<5'<(<55i@14tr6i:; end end '9df5%ist6res%a9e6'<len1:<="@14:;  9df of ' 1 ( " if mode551 figure61: sub9lot6211:stem6<="@14(9df:title67%istogram original7: a(is6< 2$& < $<<4: sub9lot6212:stem6<="@14'9df:title67%istogram equalied7: a(is6< 2$& < $<<4:   figure62:   stairs6<="@14tr:title67transformation7: a(is6< 2$& < 2$&4:   figure63:sub9lot6121:images)6uint6(<::title67before7:)olorma967gra'7:   sub9lot6122:images)6uint6'<::title67after7:)olorma967gra'7: end

Program . 4rite a +rogram to im+lement t!e lo) +ass ,ilter in Matlab. Luestion No=$ >/EA" "K@PA-- F>"?EJ  fun)tion idealfilter6P: f5imread6:; MN45sie6f:; F5fft26double6f::; u5<=6M@1:; v5<=6N@1:; id(5find6uM2:; u6id(:5u6id(:@M; id'5find6vN2:; v6id':5v6id':@N; O45mes%grid6vu:; /5sqrt6.I2O.I2:; H5double6/5P:; +5F; for i 5 1= +6==i: 5 H.F6==i:; end g5uint6real6ifft26double6+::::; ims%o#6f:figureims%o#6g:; end

Program 5. 4rite a +rogram to im+lement t!e !ig! +ass ,ilter in Matlab. fun)tion g4 5 FF?PF1/6 binsie f P:   M5lengt%6:;  get t%e fft 9osition of )utoff frequen)e 6referen)e  frequen)': f 5 f  binsie;   (id(51=1=M; fftref 5 abs6fft6sin6(id( . 2 . 9i .f:::; bounder 5 find6 6ma(6fftref: @ fftref:  6ma(6fftref: . 1<<< ::;  do t%e filter  fft( 5 fft6: if 6 P :   fft(6bounder61:=1=bounder62::5<   else fft(61=1=bounder61:: 5 <; fft(6bounder62:=1=M: 5 <; end   g5real6ifft6fft(::; end

Program 6. 4rite a +rogram to im+lement t!e boun'ar* etraction algorit!m in Matlab.  boundar' demo.m  6C: 2<<& b' Qu Hen Hu  )reated= /e). 10 2<<&  demonstrate mor9%ologi)al boundar' e(tra)tion  )lear all )lose all A<5imread67m's%a9!.bm97:; ims%o#6A<:;  a %eart s%a9e %and dra#ing title67original image7:;  9ause  A< )ontains mostl' 1s and t%e dra#ing )ontains
Program 7. 4rite a +rogram to im+lement t!e gra!ms scan line algorit!m in Matlab. >n9ut= a set of 9oints - 5 RP 5 6P.(P.':S -ele)t t%e rig%tmost lo#est 9oint P < in -. -ort - angularl' about P < as a )enter. For ties dis)ard t%e )loser 9oints. "et PN4 be t%e sorted arra' of 9oints. Pus% P<45P< and P14 onto a sta) K. #%ile i  N R "et P?1 5 t%e to9 9oint on K "et P?2 5 t%e se)ond to9 9oint on K if 6Pi4 is stri)tl' left of t%e line P ?2 to P?1: R Pus% Pi4 onto K i  in)rement i S   else Po9 t%e to9 9oint P ?1 off t%e sta)  S ut9ut= K 5 t%e )onve( %ull of -.

Program 8. 4rite a +rogram ,or e'ge 'etection algorit!m in Matlab  9rogram for edge dete)tion )l); )lear all; )lose all; read an image

filename 9at%name filterinde(4 5 uigetfile6 ... R 7.G9g77TPE+ 6.G9g:7; ... 7.bm977Kindo#s Bitma9 6.bm9:7; ... 7.fig77Figures 6.fig:7; ... 7.7 7All Files 6.:7S ... 7C%oose image6s: to be 9ro)essed7 ... 7Multi-ele)t7 7off7:; if filterinde(55< brea;end filename5)ellstr6filename:; '5 imread6%or)at69at%name)%ar6filename:::;  '5imread67)ar3.G9g7:; mm5in9ut67>n9ut 'our t%res%old bet#een < and 2$$  )onvert )olor image to gra' s)ale (5 rgb2gra'6':; ims%o#6(:; r)45sie6(:; ma(5<; min5<;

for G52=r@1 for i52=)@1   95i@1;   q5i1;   a5G@1;   b5G1;   d15abs6(6Gi:@(6G9::;   d25(6Gi:@(6Gq:;   e15(6Gi:@(6ai:;   e25(6Gi:@(6bi:;   f15(6Gi:@(6a9:;   f25(6Gi:@(6bq:;

7:;

     

 

 

 

 

 

 

 

           

g15(6Gi:@(6aq:; g25(6Gi:@(6b9:; if 6d1  <: d15d16@1:; end if 6d2  <: d25d26@1:; end if 6e1  <: e15e16@1:; end if 6e2  <: e25e26@1:; end if 6f1  <: f15f16@1:; end if 6f2  <: f25f26@1:; end if 6g1  <: g15g16@1:; end if 6g2  <: g25g26@1:; end

if 6d1 5 mm: DD 6d2 5 mm: (6Gi:52$$; elseif 6e1 5 mm: DD 6e2 5 mm: (6Gi:52$$; elseif 6f1 5 mm: DD 6f2 5 mm: (6Gi:52$$; elseif 6g1 5 mm: DD 6g2 5 mm: (6Gi:52$$; else (6Gi:5<;

end end end figure ims%o#6(:; BK 5 im2b#6( <.$:; figure ims%o#6BK:

Program 9. 4rite a Program ,or line 'etection algorit!m in Matlab. fun)tion /ra#"inesU2Ends6lineseg varargin: /ra# line segments 9arameteried as 6(1 (2 '1 '2: on gra9%   /ra#"inesU2Ends6lineseg varargin:  A sim9le fun)tion for dra#ing line segments on gra9%. Made as an  au(iliar' tool for fun)tion 7...4 5 Houg%U+rd6...:7.   >NP?= 6lineseg 9ro9erties:  lineseg= Parameters 6(1 (2 '1 '2: of line segments to dra#.  >s a Ns@b'@! matri( #it% ea)% ro# )ontains t%e 9arameters  6(1 (2 '1 '2: t%at define t%e t#o ends of a line  segment. ?%e out9ut 7lineseg7 from t%e fun)tion  7...4 5 Houg%U+rd6...:7 )an be 9ut %ere dire)tl'.  9ro9erties= 69tional:  A string of line dra#ing 9ro9erties. Kill be transferred  to fun)tion 79lot7 #it%out modifi)ation for line dra#ing.   ?P?= None   B+ JEPJ?=  Please send 'our bug re9orts )omments and suggestions to  9engtaoVglue.umd.edu . ?%ans.  Aut%or= ?ao Peng  /e9artment of Me)%ani)al Engineering  niversit' of Mar'land College Par Mar'land 2<*!2 -A  9engtaoVglue.umd.edu  Oersion= al9%a Jevision= /e). <2 2<<$

%old on; for  5 1 = sie6lineseg 1:  ?%e image origin defined in fun)tion 7...4 5 Houg%U+rd6...:7 is  different from #%at is defined in Matlab off b' 6<.$ <.$:. if nargin  1 9lot6lineseg61=2:<.$ lineseg63=!:<.$ vararginR1S:;   else 9lot6lineseg61=2:<.$ lineseg63=!:<.$ 7"ineKidt%7 2:; end end %old off;