VIGNAN’S INSTITUTE OF ENGINEERING FOR

WOMEN
(Kapujaggarajupeta, VSEZ(POST), Visakhapatnam-530 049)

DIGITAL IMAGE PROCESSING

Lecture by
BUDDHA HARI KUMAR
ASSISTANT PROFESSOR ,M.Tech.,(P.hD).
ECE DEPARTMENT
VIGNAN'S INSTITUTE OF ENGINEERING FOR WOMEN
 UNIT-1
Introduction
2

⚫ What is Image???????????
⚫ What is Pixel?
⚫ What is Digital Image?
Origins of Digital Image Processing
Fundamental Steps in Digital Image Processing
 Fundamental Steps in Digital Image Processing

⚫ Image Acquisition:
To deal with images and before analyzing them the
most important thing is to capture the image. This is
called as Image Acquisition.
There are 3 principal sensor arrangements (produce
an electrical output proportional to light intensity).
(i)Single imaging Sensor
(ii)Line sensor
(iii)Array sensor
Fundamental Steps in Digital Image Processing
Fundamental Steps in Digital Image Processing

Image Acquisition using a Single sensor

Fundamental Steps in Digital Image Processing

Image Acquisition using a line sensor

Fundamental Steps in Digital Image Processing

Image Acquisition using a Array sensor

 Fundamental Steps in Digital Image Processing

Image Enhancement:
⚫ Image enhancement is the procedure of
improving the quality and information content of
original data before processing.
⚫ Image enhancement is basically the improving
the perception of information in images for human
viewers and providing better input for the further
image processing techniques.
Fundamental Steps in Digital Image Processing
 Fundamental Steps in Digital Image Processing

Image Restoration:
Image restoration is the process of recovering an
image from a degraded version—usually a blurred
and noisy image.
Fundamental Steps in Digital Image Processing
Fundamental Steps in Digital Image Processing
 Fundamental Steps in Digital Image Processing

Color Image Processing:

Use the color of the image to extract features of
interest in an image
 Fundamental Steps in Digital Image Processing

Wavelets:
Wavelets Are the foundation of representing images
in various degrees of resolution. It is used for image
data compression.
 Fundamental Steps in Digital Image Processing

Compression:
Techniques for reducing the storage required to save
an image or the bandwidth required to transmit it.
 Fundamental Steps in Digital Image Processing

Morphological Processing:
Morphological Processing Tools for extracting image
components that are useful in the representation and
description of shape.
In this step, there would be a transition from
processes that output images, to processes that
output image attributes.
 Fundamental Steps in Digital Image Processing

Segmentation:
Segmentation procedures partition an image into its
constituent parts or objects.
 Fundamental Steps in Digital Image Processing

Representation and Description:

Representation: Make a decision whether the data
should be represented as a boundary or as a
complete region. It is almost always follows the
output of a segmentation stage.
Boundary Representation: Focus on external shape
characteristics, such as corners and inflections
Region Representation: Focus on internal properties,
such as texture or skeleton shape
Fundamental Steps in Digital Image Processing

Representation and Description - Choosing a

representation is only part of the solution for
transforming raw data into a form suitable for
subsequent computer processing (mainly
recognition)
Description: also called, feature selection, deals with
extracting attributes that result in some information
of interest.
Fundamental Steps in Digital Image Processing

Recognition: the process that assigns label to an

object based on the information provided by its
description.
Knowledge Base: Knowledge about a problem
domain is coded into an image processing system in
the form of a knowledge database.
Components of an Image Processing System
 Components of an Image Processing System

Image Sensors:
Two elements are required to acquire digital images.
1) physical device
2) digitizer
 Components of an Image Processing System

Specialized Image Processing Hardware:

⚫ This type of hardware sometimes is called a frontend
subsystem
 Components of an Image Processing System

Computer:
⚫ The computer in an image processing system is
general-purpose computer and can range from a PC
to a supercomputer.
 Components of an Image Processing System

Image Processing Software:

⚫ Software for image processing consists of specialized
modules that perform specific tasks.
 Components of an Image Processing System

Mass Storage Capability:

⚫ One method of providing short-term storage is
computer memory.
⚫ Another is by specialized boards, called frame
buffers, that store one or more images and can be
accessed rapidly.
 Components of an Image Processing System

Image Displays:
⚫ The displays in use today are mainly color(preferably
flat screen) TV monitors.
 Components of an Image Processing System

Hardcopy devices:
⚫ Used for recording images, include laser printers,
film cameras, heat-sensitive devices, inkjet units and
digital units, such as optical and CD-Rom disks.
 Components of an Image Processing System

Networking:
⚫ Is almost a default function in any computer system,
in use today.
⚫ Because of the large amount of data inherent in
image processing applications the key consideration
in image transmission is bandwidth.
 Digital Image Fundamentals

⚫ Image Sensing and Acquisition

⚫ Image Sampling and Quantization
⚫ Basic Relationship between pixels
Image Sampling and Quantization
Image Sampling and Quantization
Basic Relationship between pixels
Light and Electromagnetic Spectrum
Light and Electromagnetic Spectrum
Introduction to the mathematical tools used in
Digital Image Processing.
Set Operations
Set Operations
Logic operations
Spatial Frequencies in image processing,
Spatial Frequencies in image processing,
 Unit 2

Back qoundi
Inknsity tsanst mahon s: 0pe sate On Singe Pixes o an
Contiast manipl ation and

mogetheshol ding.
pahel hting: Pes or ny ogesahons On h neshborhoo of
eveNy Pnel in an imoge.

hanpening

where fu,u) is theinputimage

yIy)is the Proce sed imoge.. (outPut imase )
T i s n 6pera tor n y deh ne d oyer Sow

A 3x3 heigh boy ood obot a point

(hy in an image nthe patel dompin

x inthe imase to geneae anolp

Sehaldmoi
\mage
 The intnsiby Iyansluy matior ove Simpks rnol al monr
Intessing Teehhi,H Con be Te re sented as,

S Tlr)
Y Volue of Pixel be horero Cesing Pirel value of
(mput maye
af kr roCesSing Pixe vaeo
(ouurinage)
suved in (-0 Yay
|-D is mabix rm
So we ock up
mamng the yalue with S s imale mentee hrough
ConSidey Rbit uiormeut Con sisto¢ 254 eniiys
he Cause
256 we an hae (o to 2ss)
ohite.
ntan

Negave (eng tho

wkre
gray leuel tolt
oTereseuted as Back
LA

Bla ce te.
inpur in tensiy Iuel'y

aome Bosic Tstans fur na toy nchionN

-’ The aborHg sha Three Basic tyrey op Tans hiatin func ko
that ov Srequenty Und in Imaje Enchemet
Can cassify as
(ay Line ar ran ormahon tunchon. LNegatve &ldenht)
(b) \oqan thmic ( log and Tnvere lbg7
 imaye denhhy in Tinear tronshrnatan unch
Imase 1deaih is Very ess helpAl in jital mae ProteosinM
oill be Same
as int (mage.
the out put mas

inut Imase. yhich is less helo ful in iglal nctye alesis

mase neqatvei
Qnothey bmol Lineay Trans hrmaton tunc tiom.
3ntnsiky hue taken om o to Ll
bg the Aunc hiong=l-7
n dmate hegatve the intnsihy yalve (B) pirel balue oill be Yevene
Oich meany EPduce
to P'oduCe equealent hahi negahe'
inuene Value ot the ixe.

oy Evawmple.
L-0
S L-) ohile

Same ay ntey ne dite Pitel Val ug loi be (on Vey ted

intb ne qahe.

Log Thony thr haten

whee C- (onstant

qrayleve} ave pen he raylever

in thi's îni Pireh valae isley ’wl get outpu pive| Waluei
high
hightr Pixel Value Procuce the hiaher Cntrastimoe
ohicn will enchance fe inas
Volwey
 ontrostimos
Ibn Protue e high'
o Conttoyt Imas

fbury low han sfbrmaton uncthon

(oyrectiun]
a

Yas

tyasimas.
input value is ey it Poduced hiqh Cun
A
hign it ducd law Cantroy t itmage

hstogrom PrSCe ssing:

’ It is Used tor Tmaqr f nha ce nent
Data.
’ THis Graphi a! Reesentahon of any
ata elaled to gita mases.
Hstogran in ima9e Pocesing cd b te rgent
’ hsogyam Repreta}ahiy of 1elair Arequeng oocurance of yarisu gray
leuels gray voalu 5
4 4
Exaole
3|4
3344 5 3 ’4
4’4 9 A
5-’5
Histo gram teresentakon o gen
ig tal mase ig tal mase
isyan 4 vaiouy mase Protes,ny ari ahens libe
yalg
hs gram aluey
by plat pile.
liking
 Ixamele
44 44 441
3454 3
+ L2
3|5|553
3
244
4|514 |3
4 l4 444
Input amage
highest ixel Vale
to reRroeut hinhet Pirel value
Teqired 3bib be Cae.

Loto a]
Gayayleuels lo 4

Noof pixels J4

3 45 6 Prolbabilhy

ho o pixeo PoF CoF

valhe hstogram
E ual eve

4 (4 5-6

shs |-0

ileimasc N25
IWshoyam fQuligin.
 2 4- 5

fundamenlal of sahoy lkrin9

’ `patal hlkring is a funda mental Con Cept in mage PiD(eosing
mputs Vsion
enbance
Tt in dvey apdyjitles (B) kernel to an image to
(B estrac4 cche tatures Heve are Some Kcy tun damenBals:
Kevneli kui- ASma yatox ) mosk that defines the opeati on
to be applied to Cach Pixel in the image The sie and Valucs o
he kevne deky mine the type o knng (eg blurring, shieying,
cdy deec thon)
2) (on voluton :. The Puce of appyina the keine to te iage. it inolves
cach p'rel and the weighkd Sum
Over byira the Kernei On
keynel.
Covngutm
of piel Valuey Under the

3) Pirel neiqh bor hood' the Set of Pixel s Oed In tte Convluion.
+y pi cally Centeved amourd the Pixer being Processed. The
Sye of this heigh boy hod ig dekrmined by the keyneth dimnens
4 Filky Tyes. -

Thee Yedwe em phasi 2e orge s cale featwres by

hoie Qn!
Overagiy Pixe valugin the henh bor hood.
 harpening ilkrs nhane cds and hne de tnils by accrnhunhr
inthnsiy dierence, bel ween heih bonn Pixels
the
Deechy Flks'- Tdenhhy cags (t boundons in the imosrbs
tdy.
sari d clong in in tensity
higliqnting Signe dhSpeike toss ince
Kernels Can he de
enhance mens
the kermei aec the tet of t
5) kex nel Sie' the Siye ol
Layqer Kerne|s hae a byo de impact On the
eect
\mo
lkeingwhile waley erney t cuy on he details
enSure that
HIkrs sequire hor wolijahn to
Volues are bith in a Certoin Yange . tisis
the reSulhug Pixel oveoy Undevfl
to revemt
oC fen e Cess avy to he rel at the
t handle
Boundavy Handl'inm DeidingDeciding how Comnon
clunngonUolutio is essential.
Image bo nda y
app\oachey

Appicahny: Spathal il
teoatwe extiachon, ove Yeducton
\mage enhancemut1

) Rea tme RncesSing. EChcicat algoniths

0pratony Can be pertoimed
to enSuve that the
k\eing
tracHve applicahos.
iy vdeo reas AÔy in
quicey especaly
lo) uie poyses Cowode Ope rotiumy \may Ycquire multtple
dfferen Cernely to achieve the desired resulh
Posey with
Cavmbinig blurnn nd edge detec Ho.
Shoothing spaha

Bluring- RemoUal Sma detailsrom an image, Pior to bbjecestrachon

oode YeducHhon Burnng wih a Ineay (B) non linear Alkys
mohing Linear Filfers.
’ Out put o}Smuotiina ter iS Shipy averoge of Pxels Conta ined in the
hetghboy hod of the Filkr mosk
TA is aso tnoon a Averaging Filteo (ay Lopas Filky.
(a) (5)
1
TE X 2 4 2
q x
2

Jandara¤lter moslg somoothing Piler mosk

’ The smoEainy o Akring s the oCes of teplaci ng, eacn piney in an

whichis detned by
imast by the aug gay leue in the heigh loy hood,
the lkr mask
Arplicahi on - bibe te duc hon.
Side e ect blur edges

baluey
is standard aveYage of Pixel
mn mas

Box flkew

1s Eyonplk of
loeighted aus ctmorng
’ Pixe at te Cen ky o mask S moe impYtance.
’ This ý to ve duce oluing duing Smootng ncess.
r mase x ane mae ’m n.
i nelemenaton
Jenem
wlst)¢ h45, 1t)
gluy) =
wlsit

5
Shootins atar H krs

moo thing

Burring is Use in Pre prolesim task Svch as rcmuual o

shall de tails Arom an inae

woise Teduction Can be acto mplished by burig ith a

linear lter and albo by hon linear Hler ing:

smoothin Spaial

Linear ilers Non Uneay lteg

MeanlBox weighkd haussioan

tiler averas
hhedian ilter

thing ineor lks

They also knan aveyaging lecs (31)
pas
in the heiqh bour hoo d the Klkr maslk
Contne
theProces result tn an inage oity te du led shor

Ayansi ions Ih ntys, Hes whith uhma tely leady ho noie TeducHn.

Coecinb
hask

Sie more less) weigh t to pixes

2) weightd QVera
utt bation
hear aay om) the

2
mask
 -(4) (2n haussi an funthon)
>stan dard deuaion.

’ nos le.

2
3%3
uxd to bluy cdg and reduce Contrat:
similoy o me dian tilkr but b faster.

Uon ineoy (hder - statishe ) tkr

Yesponse is bosed on ordering (raning) the Pikels c toi nca in the
Their
the vale of the
tmag orea en Compa Sed b the Hlteray Qnd then Teplacng
Cenkr Rine with the Uoa lue deter mi'ned by the Yan ting Yesut
med|on of all the pixe yalues
) Median iltr - Find the
2) Min Aley Ane the minimum f aute Pixey yaluey
ind he maxi mum oan he th pikel yalue,
3) Ma iter

culae the Output o the

Eramolec Cansidex the imase beloo ane Cal
ing Sg heighbour hood ing
Pixel (2y2) it moo thing > done
al the 4lkrs bel:

4 4 51
oo
4|2
) Min lHe
e) Max iltey

a) box Aler
449tS t4+6+8 4240 +1
o
42
 wase
5) wiqkd aug 4lkr

5-8C25

all the aluey in assigning Order

the Genkr element i) 5

median 5
d) nin hlter ) max ley

The Pincipal Objc chve of hor pening is to hig hlight the tinate
detail ®) to enhanee details.
Tt hos wide ange of Heeli cations
ch as- Elec tonic
Pn hag medical magin3,
ndustia inspe ctions Qnd
autonom ou uidane in military systeg.
Imoe burmng Pir el averaging

harpening Seatai ierenfai on

Advotoge
Tmase dierentiation enhanee edges ane hoie 2 de tmphasiy areq
ith s y oring qray-yleuel volues.
a) First - Ordy de iyhve'

y ughe nonveo at on setofa gray level Step (Ölam

 Se Ceymd deHaHve,
aa
ero ln Hat
muyt be teo(3) Ta
ndl gray ku
SS be ro at he
onstot lho
must be 3e albng ramps ot
oCCuY is
ditance oVr hicCh changy Can
’ he shoy te,t
betweey adyecent Pixets

The basic deinihon

6yder deyahve ol a Ohe- dimensinal

Laplacan Alky
TL highliqhy roy \evd ds anHnuiig in an i mase.

leuely
Rormla

<-4}y).
Lalaian maye

-) -

G Erample
4tr On the 9iven imakon
Arply laglacian
the Cener Pixel,

5 4

Tniut @imase. masle

-(8 xo)+5)+xo) + bx1)+G K-4+(2x) +l*6) +3|44x0

0+54o+o 24. t2tot+0

-|4

4
O-4 2
2

utfut i mae
 Enhanted laplacian
Lapla flky

Enhanced

Enhaneee

Eyamele
Aely enenance La lacanilky On the iven ma9e onthe
Centkr Pixe

ihutimar
}+544 t0-5442 4|+3t4

85 4
-24 2

Eramplc Agply Laplacian ltkr n tte gien imase.

So
59
50 SoSo So
So S6KO
50 Soso 50 masle.

\00 (o0
Combining sphal en'an cemens me thods '
Combining Spaka en hante ment methodk in digibi imoge Proceisiya
in volveg agoying mutiele enhon enevt tec hniques to On Jmogr
aChieye a desiyed Yesul. HeYe are 3one Common appioachey for
gpaia enhan cemeus me thods:
Combinig

Jeu Can apply diftrent enhanre ment neltod s Sequen hally ina
apecike brder. For eramp'e, \ou mignt 3}ort wth histoaram
CQ ualing fion to im owe Con tYayt ond ollaoit with har peninq Altesto
ehanre e dges- he thdey and Com bina fiorn depend on tte srecfc
Characterisics yu lont to enhance .

) forallel Procesing'
Apol dikrent cnhanement methods to Cop ies of the ame
Itmage
nd 4he blend the Yeslts-. For instane u Can a pply one tmethod to the
luminonre Channn and amo th to the Color hanmels in an inage- Afky
ProCesin9 Ye Gm bine the Channe% ho ger tbe fing nhanced roge.
) Adaphve Entancemeut:
Aprly diferet
methods baJed On he
egions oth ini. Thij appó0dch Can Contet o£ the imoe oy Speeific
involve gmentation techniques
to ndentihy egions oe inttyest One then
nethd o
apply the moS- Suitable chhancr
coch region.
4) feed boce Lo0ps'
Uefecabace Ymechoni Sms tu lés ofvdy
imox Afker enhane e
applyinon erhan(ewne t mcthod, yu Can evalak tthe
e'yll
teCc ssay
mage Qualtyis achieved. ths Rrocess an
Con tinue untli the desired i
3) Haire Learn ing. utilie
mue Suitade Cnan
macine larving modey to Predi
cenet meths foy diksent hy o the
Oubmahirally hiy app0och involves traihing o fmags odel Cna data se
of imass ond they
Corres ponding
enhancement
 A 0x ty
inttrachely Select Ond adjust en hancemeut he
timeyQpr oaCh Can be Useul appli ations where thodsinml
Is essehas Ax hine- huna tx Verh&e
tuning tbe enhancemeny.
The Choice oe
Combining me huds
depeds on the Spepeahe Ihase
enhancement Soals, Hhe
Qrd the Comutahona choyactk istics f the imogo
re Ssurces youre Coareing wth
diekerenr Quailable Ex pen
Comeina kn s Qan hel deter mentafon and testing
mine he mast
eectie aptloac, fy
 Hlleing in the FrQueny Domaiy
Prelimino Conceo in frequncy omain
n pa ia om ain

hly) impulse reoponie.

the he
kig in the Hhequemy comai
asic stes h Alking

Invene
FouricY
ouier Trans hrm

Pre
Rcesing

Enhanced Imaye

Basic of Alteinyin the thequenty duin:

ees to louing Queshons in hegunEy Je main

by G)*T to Centey the trans brm:

mutiy the input imase

DE o the masom sep-i

Compute El) the
Fuc
tmulhly e(uy) by afilkr
DeT 0f,Yeows in 3e3
Compute the inee
Dbtain ttthe ea Pas 6l the es ulb 'n Sttp 4
Stcpts

Sho
 mage moting Using hequency cmain Ae
’hese are also alled low- Pass Alkrs
Thesx Ore ed to Smoo then the image as they alle On..
lbw tre uency ComPonenb to Pass though.
hese Ore Ud to emove nolse.

TyPeo Los-paus klkrs

(iLPE)
poss led
) Tdeat Low (BLPF)
ButtrloUyth Loo pass
Gaussian ow pas
2)
3)
deal Low Pass
the
ohere DoCutoef
is
vrequeny

Ales( BLE)
oe
where Dis the Cut Oyder
Hluyu) Areueny ana nis the
HIer
of the butr oY

hausion lou pas Gtr(GLee) ug hene Do is the Cutoffrquegl

H(u)

Osing requeney domain

Tmage ahargening
Cale& hiah -pas HIkrs
’ Thesc Ore aso
These Ore
Used to harpen the imoge as they only
allo igh rcqucney Compon enb to Pa through
\mage.
are Ued to move bacroud of an
’ ther

yPe o

) Ldea) igh pas Hky (2HpE)

3) hasian
 ohee Dos he

Rulkr rth high -Pas Cutof

where Po i) ha
ArcQueny anenis theordet

(a4ef)
Gaussian highpoyiles |2D6where Do i) the
|-ey) Fequeney
Hlb)

(,)Ore 0rder
\ote Pbint Dluu)=
>Euelidean distance.

Ex amele
tyans furm
Cnvert the given Spatar domain image uaing fóuneY
Q Per torm Ldel loo Paus kr t Shuthen the mase. hooe
Shos the Ster by ste RCe due ty ding the Same.

An imag
steei the input imasc by 4, Cenkr the
(q5) (o) (oy) (o,)
Aronsrm( . [

(Ho)
 -t4 2 -) (h3)
22s (0)o-y
4-zy2
s
(2)3)
Ritance
wask e 2Cenkr
the fm betcen
(uj dtanee Calculating
the ae
mase
Tdel Foy y)
Huy ton tunc Alks the wit h Flu nulhply
4n4
matha kerhel 4x4 kePwy)x Kenelx F(un)=
nei
froamm imas. the a'DEr theCompute
see
Ser
 2 223
2-23
4-2.4
-23
|4)
23 |41

Do =05 Value 9ten

Hluo)

GO

Compuk the inverse DFT O}"he result in stee

kernel
1

Wernel

Sters Mulhphy the Teut by ()

elechve Hkrng
aleokd onge of hequtny, CamPonab

-0rgina Sgna,

-Low-tass Hllered

Bang Pay ter

Band she lleyd

ati doal

Fluyu)4(u)
eYers to the
rec hve Altehng in inmag Pcessing
apoli cahon ot
ihin an
Qn ima, nhey
Cerdain eqiony ) features enive |mag This
vtymy to he
than Qphina fhe custmed

pulahons.
maEnhanecnenb ( ) mani
) Keqion of
Tnkvest (Ro1) tilking'
Avpyina fA lkrs (®) Proesing operatons ony to a deied regiou o) avea
of in teest tth the imase "hig
b noise e dueHon
b Sharpening Spe cite objects

) Edg eteion:
Detkehng cdo) boundaiey othin nimagc and
techoes to enhanec o e mpa Sie thoe clges is is Use fur task
ke obiect dekchen (G Seqmentohon

) Colby Bascd tlHeig'

Tsoainq SpeaAe Color s() Colbr ams within animaye and
lters o) adwt menb th thoe Colors. This an be oxe Aul
tyacking:
y tasks lite color o\ecin ) 05irct

4) Tlequy Dmain Eilking.

Analyzing the e quemy Cor
Com frne ns of an tmae 0tng technic)
tiee fousic anayss nd then an bying lers to spe etc requen
yomou ng peiodic nuise
5 Enhancim Cetain tartr ealkrmn

Aring msrphalogia opeiao n ik dilaiy eosion tospecGe

eions in the 'mave to modiy the Shapey ) SBrutu wthin
thse rejon

lectiue allos oY more Piecise ontol ves image.

ROcessng Ond Con be es ental in appliaionsllke Camuk
Vision me di ) lmaing and video rocessng

2
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
 IV B.TECH ECE I SEM EXAMINATION
DIGITAL IMAGE PROCESSING
Question Bank

Unit-1 : Introduction& Image Transforms

1. Explain the fundamental steps of DIP with a neat block diagram
2. Analyse quantization process for a 2-D function f(x,y).
3. Define the following
a) Pixel
b) Digital image
c) Resolution
4. Justify the role sampling for an image with an example
5. Define Haar Transform and derive the Haar Transformation matrix for order N = 4
6. Write about Slant transform and mentions its applications
7. Discuss the relation between pixels
8. State and prove following properties of 2-D DFT
a) Periodicity
b) Conjugate symmetry
c) Linearity
9. Explain about KL transform and write its use in image processing.
10. Explain about Hadamard transform and determine the Hadamard matrix for order N = 8.

Unit-2: Intensity Transformations and Spatial Filtering, Filtering in the frequency domain

1. Explain about Image sharpening using second order derivative operator.

2. Discuss about image smoothing in the frequency domain using ideal low pass filters.
3. Discuss about image smoothing in the frequency domain using Butterworth low pass filters.
4. Discuss about Log transformation and Power Law transformation, and write their applications.
5. Is spatial filtering useful for sharpening of an image, Justify?
6. Discuss how homomorphic filtering decreases the low frequency components and
amplifies the high frequency components simultaneously.
7. Compare and contrast frequency domain enhancement techniques and spatial domain
enhancement techniques
8. What is image smoothing? Explain spatial domain low pass filtering technique in detail.
9. Discuss smoothening of an image in frequency domain.
10. Differentiate between linear spatial filtering and non-linear spatial filtering.
11. Mention the objective of image enhancement techniques.
Unit-3: Image Restoration and Reconstruction

1. Describe II-Order Statistics filters in detail?

2. Explain the model of image degradation with neat block diagram
3. Discuss constrained least square filtering technique for image restoration
4. Describe mathematical model for constrained restoration
5. Explain inverse filtering for image restoration
6. Define image restoration
7. Explain about image restoration using minimum mean square error filtering.
8. What is image blur and discuss its causes?
9. Discuss the applications of image restoration
10. Discuss in detail the image restoration using Wiener filtering.
11. Consider the image below and calculate the output of the pixel (2,2) if smoothing is done
using 3*3 neighbourhood using all the filters below
18807 i) Box/Mean filter
47957 ii) Weighted avg filter
54686 iii) Median filter
42015 iv) Min filter
01020 v) Max fitter

Unit-4 Image compression& Wavelets and Multi-resolution Processing:

1. Write an algorithm for Huffman coding and determine Huffman code words for a message
{m1, m2, m3, m4, m5}. The probabilities of symbols are given as
Symbol m1 m2 m3 m4 m5
Probability 0.3 0.3 0.15 0.2 0.5
2. Explain the concept image pyramid.
3. Write an algorithm for arithmetic coding and determine code word for a string “CAB”
The probabilities of characters are given as
Character A B C
Probability 0.6 0.3 0.1
4. Discuss edge linking method using local processing approach
5. Explain lossless, lossy predictive coding with a neat block diagram
6. Explain Wavelet transforms in one dimensions.
7. Design golomb code from 9 with divisor 4.
8. What is an inverse filtering? Explain how it is useful for image restoration and write the
disadvantages of it.
9. With an example, explain the concept of Run Length coding
10. Explain the concept of lossless predictive coding.
Unit-5: Image segmentation, Morphological image processing & Color image processing

1. Write the applications of segmentation and explain threshold based segmentation

2. Explain about morphological opening operation with example.
3. Explain the use of intensity to color transformation in image processing.
4. Explain about morphological closing operation
5. Explain about CMY and CMYK color models, and write their applications.
6. Explain about morphological Hit-or-Miss transform
7. Discuss about converting colors from HSI to RGB.
8. What is intensity slicing and color coding? Explain their use in image processing.
9. Discuss about image segmentation using region growing.
10. Explain about color image smoothing with necessary equations.