FUNDAMENDALS OF DIGITAL IMAGE PROCESSING

Fundamendals of image processing

1. Digital Image Processing
Dr. B.SARITHA
ASSISTANT PROFESSOR,ECE
JAISHRIRAM ENGINEERING COLLEGE
TIRUPPUR

2. 2
Description
Restoration is a process of reconstructing or
recovering an image that has been degraded by
using a priori knowledge of the degradation
phenomenon.
It is an objective process.

3. 3

4. 4
Image Restoration
ImageRestorationaimsatreducingorremovingvarious
typesofdistortionsintheimageofinterest.
ImageEnhancementisasubjectiveprocessconcernedwith
thepleasingaspectsofaviewer.
Imagerestorationtriestoreconstructorrecoveranimage
whichwasdegradedusingaprioriknowledgeof
degradation.
Herewemodelthedegradationandapplytheinverse
processtorecovertheoriginalimage.

5. 5
Image Degradation / Restoration Process
Inthismodel,weassumethatthereisanadditivenoiseterm,
operatingonaninputimagef(x,y)toproduceadegradedimage
g(x,y).
Giveng(x,y)andsomeinformationaboutdegradationfunctionH,
andknowledgeaboutnoisetermη(x,y).
Theaimofrestorationistogetanestimatef(x,y)oforiginalimage.
ThemoreweknowaboutHandtermη,wegetbetterresults.

6. 6
Image Denoising
Involvesformulaorcriteriathatwillyieldtheoriginalimage
fromthenoisecorruptedsignal
Variationsofintensitythathavenobearingontheinformation
intheimagearecallednoise
HowNoise?Potentialdegradationoccurbecauseof
Sensornonlinearities
Dustintheoptics
Nonlinearityintroducedbyelectroniccomponents
Improperhandlingorloadingoffilm
Poorcleaningofscreen
ContaminatedhandsandFingerprints
Improperloadingoffilms
Imagemotionblurandgeometricdistortion

7. 7
Origin of Noises
Gaussian noise arises in an image such as electronic circuit
noise and sensor noise due to poor illumination and/or high
temperature.
Rayleigh noise arises in range imaging.
Exponentialand Gamma noises appear in laser imaging.
Impulse noise is found in places where quick transients,
such as faulty switching take place during imaging.
Rician noise appear in MRI images, Gaussianand Poission
noisesarises in X-ray and Mammograms.
Speckle noise is seen in Ultrasound images

8. 8

9. 9

10. 10
Image Denoising

11. K S R College of Engineering 11

12. Image Denoising
Involves formula or criteria that will yield the original image
from the noise corrupted signal
Mean filters
Arithmetic mean filter
Geometric mean filter
Harmonic mean filter
Contra-harmonic mean filter
Order statistics filters
Median filter
Max and min filters
Mid-point filter
Alpha-Trimmed filters
12

13. Original Image
Median Filter Wiener Filter
Salt & Pepper Noise
13

14. 14

15. K S R College of Engineering 15

16. 16

17. K S R College of Engineering 17

18. 18
Inverse Filtering
Motion Blur After Inverse Filtering

19. 19
Original image Image blurred by a
Gaussian-shaped
Point Spread Function
Result of Inverse
Filtering
Inverse Filtering

20. 20
Image Segmentation
Segmentationdealswiththeprocessoffragmentingtheimage
intohomogeneousmeaningfulparts,regionsorsub-images.
Imagesegmentationisaprocessinwhichregionsorfeatures
sharingsimilarcharacteristicsareidentifiedandgrouped
together.

21. K S R College of Engineering 21
Image Segmentation

22. 22
Image Segmentation

23. 23
Discussions

24. 24K S R College of Engineering