Image Degradation Model (Linear/Additive)

Size: px
Start display at page:

Download "Image Degradation Model (Linear/Additive)"

Transcription

1 Image Degradation Model (Linear/Additive),,,,,,,, g x y f x y h x y x y G u v F u v H u v N u v 1

2 Source of noise Objects Impurities Image acquisition (digitization) Image transmission Spatial properties of noise Statistical behavior of the gray-level values of pixels Noise parameters, correlation with the image Frequency properties of noise Fourier spectrum White noise (a constant Fourier spectrum) 2

3 Some Noises Distribution: 3

4 Test Pattern Histogram has three spikes (Impulse) For two independent random variables (x, y) z x y p z p x p y Z X Y 4

5 Noisy Images(1): 5

6 Noisy Images (2): 6

7 Periodic 1 Noise: Electronic Devices 1 Disturbance 7

8 Estimation of Noise Parameters Periodic noise Observe the frequency spectrum Random noise with PDFs Case 1: Imaging system is available Capture image of flat environment Case 2: Single noisy image is available Take a strip from constant area Draw the histogram and observe it Estimate PDF parameters or measure the mean and variance 8

9 Noise Estimation: Shape: Histogram of a subimage (Background) 9

10 Noise Estimation: Parameters: N S N Maximum Likelihood me tho d: zi arg max p z, i 1 i i1 Momentum z p z 2 z S i i i 2 z p z i i zis Solve? Power Spectrum Estimation: 2 PDF Parameters N 1 i 2 2 N u, v E x, y x, y N i i1 xy, : Flat (No object) Subimages 10

11 Noise-only spatial filter: Mean Filters:,,,,,, g x y f x y x y G u v F u v N u v A m n mask centered at (x, y): S xy An algebraic operation on the mask pixels! 11

12 Mean Filter: Arithmetic mean: 1 fˆ x, y g s, t Noise Reduction (uncorrelated, zero mean) vs. Blurring Geometric mean: mn ( s, t) S xy Same smoothing with less detail degradation: fˆ x, y g s, t ( s, t) S xy 1 mn 12

13 Mean Filter: Harmonic mean: Salt Reduction (Pepper will increase)/ Good for Gaussian like Contra-harmonic mean: fˆ x, y fˆ x, y Q>0 for pepper and Q<0 for salt noise ( s, t) S xy ( s, t) S xy mn 1 g s, t ( s, t) S xy g s, t g s, t Q1 Q 13

14 Example (1): Additive Noise Mean Arithmetic Geometric Original Noisy More-Less Blurring Arithmetic Geometric 14

15 Example (2): Noisy (Pepper) Noisy (Salt) Impulsive Contra-harmonic Correct use Q=+1.5 Q=

16 Example (3): Impulsive Noise Conrtra-harmonic Wrong use Q=-1.5 Q=

17 Noise-only spatial filter: Mean Filters:,,, g x y f x y x y A m n mask centered at (x, y): S xy An Ordering/Ranking operation on mask members! 17

18 Median: Most Familiar OS filter: fˆ x, y mediang s, t Little Blurring/Single/Double valued noises Max/min: ( s, t) S xy ˆ fˆ x, y Max g s, t, f x, y min g s, t ( s, t) S xy ( s, t) S xy Find Brightest/Darkest Points (Pepper/Salt Reducer) 18

19 Midpoint: OS+Mean Properties (Gaussian, Uniform) 1 fˆ x, y, min, 2 Max g s t g s t ( s, t) S xy ( s, t) S xy 19

20 Alpha Trimmed mean: Delete (d/2) lowest and highest samples d = 0 Mean Filter ˆ 1 f x, y g s, t mn d ( s, t) S d = mn-1 Median Filter Others: Combinational Properties xy r 20

21 Example (1): Noisy (Salt & Pepper) One Pass Impulsive Noise Median Filter passes Less Blurring Two Pass Three Pass 21

22 Example (2): Max and min Filters: 3 3 Max Filter 3 3 min Filter 22

23 Example (3): Noise: Uniform (a) and Impulsive (b) Mask size: 5 5 Filters Arithmetic mean (c) Geometric mean (d) Median (e) Alpha Trimmed (f) 23

24 Adaptive, local noise reduction: 2 If is small, return g(x, y) 2 2 If L,return value close to g(x, y) 2 2 If L,return the arithmetic mean m L 2 Non-stationary noise or poor estimation: 2 L fˆ x, y g x, y g x, y ml x, y xy, 2 max 1 2 L xy, 24

25 Example: N(0,100) Filters: Arithmetic Mean Geometric Mean Local Original Noisy A. Mean G- Mean Local 25

26 Adaptive Median: zmed Median of intensity in Sxy A1 Zmed Zmin, A2 Zmed Zmax zxy Intensity value at ( x, y) if A1 0 and A2< 0 then SMax Maximum allowed size of B1 Z xy Zmin, B2 Z xy Zmax if B1 0 and B2< 0 then Zˆ else ˆ xy Z xy Z xy Zmed else increase Mask size: ( Smax ): Check " A" condition. ( S ˆ max ): Z xy Zm ed z z min max Minimum intensity in S Maximum intensity in S xy xy S xy 26

27 Example: Noise: Salt and Pepper (Pa=Pb=0.25) Noisy Median (7 7) Adaptive Median (Smax=7) 27

28 Band Reject Filter (Periodic Noises) Ideal, Butterworth, and Gaussian IBRF BBRF(1) GBRF 28

29 Band Reject Filter (Periodic Noises) W 1 Du, v D0 2 W W H I u, v 0 D0 D u, v D0 2 2 W 1 Du, v D0 2 1 H B u, v n 2n Du, vw D u, v D0 H G D u, v D 0 u, v 1 exp 2 Du, vw 2 29

30 Example: Noisy Spectrum BBRF(1) DeNoised 30

31 Band Pass Filter (Extract Periodic Noises) Analysis of Spectrum in different bands H u, v 1 H u, v Noise Pattern: bp br 31

32 Notch Reject Filter: INRF GNRF BNRF NPF=1-NRF 32

33 Notch Filter (Single Frequency Removal):, 2 2, 2 2, u, v 2 2 D u v u M u v N v D u v u M u v N v H H H I B G u, v u, v 0 D u, v D and D u, v D 1 O.W., u, v D0 1 D u v D D u v D u, v 1 exp D0 33

34 Example Horizontal Pattern in space Vertical Pattern in Frequency Spectrum Notch in Vertical lines Notch Passed Spectrum Vertical Notch Rejected Spectrum 34

35 Several Single Frequency Noises: Images Spectrum 35

36 Multiple Notch will disturb the image: Extract interference pattern using Notch Pass 1 N( u, v) H NP ( u, v) G( u, v) x, y H NP( u, v) G( u, v) Now Perform Filter Function in Spatial Domain as An Adaptive-Local Point Process: f ˆ x, y g x, y w x, y x, y Optimization Criteria: minimum variance of filtered images min 2ˆ f 36

37 Optimum Notch Filter: 2 fˆ 2 d d 1 x, y ˆ, ˆ f x s y t f x, y 2d 1 sd td 2 fˆ 2 d d ˆ 1 f x, y ˆ 2 fxs, y t 2d 1 sd td d d 1 x, y g x s, y t w x s, y t x s, y t 2d 1 sd td Smoothness of w( x, y):,,, g x y w x y x y w x s, y t w x, y, w x, y x, y w x, y x, y

38 Optimum Notch Filter: d d 1 x, y g x s, y t wx, yx s, y t 2d 1 sd td 2 fˆ 2,,, g x y w x y x y ˆ xy, f g x, y x, y g x, y x, y 0 w x, y w x, y x, y x, y Computed for each pixel using surrounded mask 2 38

39 Result: Former images Spectrum (without fftshift) 39

40 Result: N(u,v) η(x,y) 40

41 Result: Filtered Image 41

42 Linear Degradation: g x, y H f x, y x, y L-System: H f x, y f, H x, y d d h x, y,, H x, y H f x, y f, h x, y dd LSI-System: g x, y f x, y h x, y x, y G u, v F u, v H u, v N u, v 42

43 Degradation Estimation: Image Observation: Look at the image and Experiments: Acquire image using well defined object (pinhole) Modeling: Introduce certain model for certain degradation using physical knowledge. 43

44 Degradation (Using Experimental PSF) Original Object Degraded Object H s u, v G u, v A PSF 44

45 Atmospheric Turbulence: NASA 45

46 Modeling of turbulence in atmospheric images: 2 2, exp H u v k u v 56 46

47 Motion Blurring: Camera Limitation; Object movement. 47

48 Motion Blurring Modeling:,, T g x y f x x t y y t dt 0 T T j 2 ux vy G u, v f x x 0t, y y 0t dt e dxdy 0 j 2 ux t vy t 0 0 G u, v F u, v e dt F u, v H u, v 48

49 Linear one/two dimensional motion blurring: at T j ua x0 t, tmax T H u, v sin uae T ua at bt T x0t, y0t H u, v sin ua vb e T T ua vb j uavb 49

50 Motion Blurring Example (1): a=b=0.1 and T=1 Original (Left) and Notion Blurred (Right) 50

51 Motion Blurring Example (2): 51

52 Uniform Motion Blurring (?): 52

53 Out of Focus Blurring: h x, y circ r 1 circ 2 R 1 r 1 0 r 1 x y 2 2 R 53

54 Performance Metrics in Image Restoration: Definition f(x, y): Clean Image n(x, y): zero mean uncorrelated noise with variance σ 2 g(x, y): Blurred Image, f(x, y)(x, y) f x, y : Restored Image SNR (Signal to Noise Ratio), Image Quality: BSNR (Blurred SNR) SNR 10log BSNR 10log f 2 xy, 10 2 g 2 x, y xy, 10 2 x, y 54

55 Performance Metrics in Image Restoration: ISNR (Improvement in SNR): ISNR 10log,, xy, 10 2 xy, f x y g x y, ˆ, f x y f x y 2 55

56 Inverse Filtering: Without Noise: Hˆ u v ˆ G u, v F u, v H u, v F u, v F u, v Hˆ u, v, Problem of division by zero or NaN! 56

57 Inverse Filtering: With Noise: ˆ ˆ G u, v F u, v H u, v N u, v N u, v F u, v F u, v Hˆ u, v H u, v Hˆ u, v Problem of division by zero! Impossible to recover even if H(.,.) is known!! 57

58 Pseudo Inverse (Constrained) Filtering: Set infinite (large) value to zero; G u, v Hˆ u, v ˆ ˆ F u, v H u, v 0 Hˆ u, v H H THR THR 58

59 Example: Full Band (a) Radius 50 (b) Radius 70 (c) Radius 85 (d) 59

60 Threshold Effect: Ghost 60

61 Phase Problem: 61

62 Wiener Filtering in 2D case: Degradation f(x,y) g(x,y) g(x,y) g x y s x y x y G u v S u v N u v,,,,,, W u v G u v,, ˆ,,,., 2, Fˆ u, v,., E u v F u v F u v F u v W u v G u v E E u v E F WG F WG E FF WGG W WGF FW G 2 2 Restoration E F W G W WGF W G F f(x,y) 62

63 Wiener Filtering in 2D case: W 2 E E u, v P WP W W P WP 2,, 2 FF GG FG GF E E u, v PFG u, v 0 W u, v P u, v P u, v E X :Spectral Estimation XX P u, v E XY : Cross Spectral Estimation XY P u v P u v XY YX GG 63

64 Wiener Filtering in 2D case: Special Cases: Noise Only:,,,,,, g x y f x y x y G u v F u v N u v Uncorrelated "zero mean noise" and "image": FG W u, v * 2 * FF * 2 2 * *, P u v E F F N E F E F E N P, P u v E F G F N E F E N E F E N E F E N P P GG FF NN FF PFF u, v,, P u v P u v NN 64

65 Degradation plus Noise:,,,,,,,, g x y f x y h x y x y G u v F u v H u v N u v Uncorrelated Noise and Image: W FF u, v 2 P H PFF H PNN H 1 H 1 P H P 2 2 H 2 NN H 2 PFF FF 2 H H P P NN FF H P NN SNR 1 65

66 Degradation plus Noise: White Noise W u, v 1 H H 2 H 2 K Select Interavtively 66

67 Wiener Filter is known as: Wiener-Hopf Minimum Mean Square Error Least Square Error Problems with Wiener: P FF P NN 67

68 Phase in Wiener Filter: W H W 2 2 PNN 1 H H PFF P 1 H H 2 H NN PFF W H No Phase compensation! 68

69 Wiener Filter vs. Inverse Filter: 1 H H H 0 W lim W 2 H 2 P P NN NN H H 0 H 0 P FF 69

70 Other Wiener Related Filter: W W S H, 0 1 H 2 PNN H P FF 1 1 exp j 2 P NN H H P FF 70

71 Nonlinear Filter:, 1 ˆ j G u v,,,, F u v G u v e G u v G u v 1 HP like 1 LP like Fˆ u, v j Gu, v log G u, v e G u, v G 0 O.W. min 71

72 Example: Full Inverse Pseudo Inverse Wiener 72

73 Example: Degraded Inverse Wiener Motion Blurring+Noise Noise Decrease 73

74 Algebraic Approach A Matrix Review: Matrix Diagonalization 1 : Each square matrix A (with n linearly independent eigenvectors) can be decomposed into the very special form: A PDP 1 P: Matrix composed of the eigenvectors of. D: Diagonal matrix constructed from the corresponding eigenvalues of A

75 Algebraic Approach: Matrix Representation of Circular Convolution:, Periodicity: In Matrix Form: N 1 y n f n g n f j g n j g i g i N j0 y0 g0 gn 1 gn2 g1 f0 y g g g g f N y 2 g2 g1 g0 g 3 f 2 y g g g g f N 1 N 1 N 2 N 3 0 N 1 C :, A Circulant Matrix! C k j g k j C 75

76 Algebraic Approach: Diagonalization of Circular Convolution: C = WDW WDW, W W W m, n 1 1 e 2 jm n N G G 1 0 D G k 0 0 G N 1 DFT g n K 76

77 Algebraic Approach: Diagonalization of Circular Convolution: f g WDW f WDW f DFT Multiply IDFT Extension to 2D case is similar! 77

78 Algebraic Approach to Noise Reduction: Least Square Approach: Matrix Representation of Degradation Model: g Hf n n g Hf g Hf g Hf fˆ arg min n arg min g Hf f f 1 ˆ T T f H H H g Pseudo Inverse (Left) T 1 T H H H H = I g Hf g Hf g Hf f f T 0 T 2H g Hf 0 T 78

79 Least Square Approach: If H is square and H -1 exists then: This is Inverse Filter! fˆ 1 H g f x, y H u, vg u, v Gu, v 2 Hu, v H u, v * H H H ˆ T T f Hg g 79

80 Constrained Least Square (CLS) Approach: Q is a linear operator (High pass version of image, Laplacian and etc.) Using Lagrange Multiplier: f ˆ arg mi n Qf Const. t o g - Hf n f J f Qf g - Hf n J f f T T 2Q Qf - 2H g - Hf T 1 T T T T 1 T ˆ f H H Q Q H g H H Q Q H g : Lagrange Multiplier which should be be satisfy: g - Hf 2 2 n 80

81 CLS in frequency domain: Diagonalization: 2 1 * T * H = WDHW H H = W DH W T * 1 * * T 2 * H H Q H = WD W H = WD W = WD W Q Q = W D W 1 2 T T T 2 * * * H H Q Qf H g W DH DQ W f WDH W g 1 Multiply by : 2 2 * * * 2 2 * D ˆ H DQ W f DH W g H u, v C u, v F u, v H u, vg u, v W C u, v : Fourier Transform of Q (Linear Operator!) * H u, v, C u, v Fˆ u, v G u, v 2 2 H u v 81

82 CLS in frequency domain: * H uv, 2, C u, v Fˆ u, v G u, v 2 H u v ˆ f H 1 T 1 T T H QQ H g 82

83 Wiener Filter vc. CLS filter: T * H u, v, C u, v T ˆ R E f ff F u, v G 2 2 u, v H u v R E nn n 1 T 1 ˆ T 1 T T Q Q R f Rn f H H Q Q H g 0 Inverse Filter 1 Wiener Filter (General form) 83

84 For Laplacian as a measure of smoothness: P x, y C0 CM 1 CM2 C1 C1 C0 CM 1 C 2 Q C2 C1 C0 C 3 CM 1 CM 2 CM 3 C 0 C j,0, 1,1 Pe j Pe j N Pe j P e j N Pe j, 1,0 84

85 How to estimate γ: Residual Vector: r = g - Hfˆ T It can be shown that r r r r is monotically increasing with. 2 2 We seek for such that r n, is our accuracy 1. Initial Guess on 2. Calculate 3. If criteria r r g - Hfˆ n is not reached : 3.a if r n increase. 3.b if r n decrease. 4. Re-calculate new filter and repeat until criteria is meeted 85

86 How to estimate γ: Computational Tips: 2 2 We need r and n. 2 2,,, ˆ, r, 2 n n M1N1 x0 y0 M1N1 x0 y0 M1N1 2 2 x0 y MN n mn Need only n x, y n x, y x, y and 2 2 n n n 2 M1N1 R u v G u v H u v F u v r x y m n 1 MN 1 MN n n m m x0 y0 86

87 Example: CLS High Noise Mid Noise Low Noise 87

88 Example: Correct Noise Parameter Incorrect Noise Parameter 88

89 Computation Efforts: g Hf n n g Hf T 1 ˆ T f H H H g H is size of MNMN ( )!!!! Gradient Descent Methods of function minimization: x GD min x x x x x x Convergence Condition: 0 : Eignvalue of E ( k 1) k k k xx T 2 max k 89

90 Iterative LS or CLS (Tikhonov-Miller): LS : g Hf n n g Hf Φ f 1 g Hf g Hf Φ 2 f H g Hf 1. f T H g 2 T f f H g Hf H g I H H f f I H H f CLS : T T T T k 1 k k k 0 k J f J f Qf g - Hf n 2 f T T Q Qf - H g - Hf 1. f T H g T T T T T H Q Q 2. f f H g - Hf Q Qf H g I H k k k k k f 90

91 Example (Uniform Noise): Blurred by a 7 7 Disc (BSNR=20dB) Restored Using Generalized Inverse Filter with T=10-3 (ISNR=-32.9dB) 91

92 Example (Gaussian Noise): Blurred by a 5 Gaussian (σ 2 =1) (BSNR=20dB) Restored Using Generalized Inverse Filter with T=10-3 (ISNR=-36.6dB) 92

93 Example (Uniform Noise): Blurred by a 7 7 Disc (BSNR=20dB) Restored Using Generalized Inverse Filter with T=10-1 (ISNR=+0.61 db) 93

94 Example (Gaussian Noise): Blurred by a 5 Gaussian (σ 2 =1) (BSNR=20dB) Restored Using Generalized Inverse Filter with T=10-1 (ISNR=-1.8 db) 94

95 Example (Uniform Noise): Blurred by a 7 7 Disc (BSNR=20dB) Restored Using CLS Filter with ϒ=1 (ISNR=+2.5 db) 95

96 Example (Gaussian Noise): Blurred by a 5 Gaussian (σ 2 =1) (BSNR=20dB) Restored Using CLS Filter with ϒ=1 (ISNR=+1.3 db) 96

97 Example (Uniform Noise): Blurred by a 7 7 Disc (BSNR=20dB) Restored Using CLS Filter with ϒ=10-4 (ISNR=-21.9 db) 97

98 Example (Gaussian Noise): Blurred by a 5 Gaussian (σ 2 =1) (BSNR=20dB) Restored Using CLS Filter with ϒ=10-4 (ISNR=-22.1 db) 98

99 Example: 99

100 Wiener Filter Example (input): Original (Left) and degraded with SNR =7dB (Right) 100

101 Wiener Filter Example (output): Known Noisy Image (Left) and Know Clear Image (Right) 101

102 Iterative Wiener Filter: Wiener in Matrix form (Spatial Filter): g = Hf + n fˆ Wf T 1, T is Hermitian Transfor W R H HR H R R R f n T f f n : Autocorrelation of f : Autocorrelation of n m! 102

103 Iterative Wiener Filter: Conditions: 1. The original image and noise are statistically uncorrelated. 2. The power spectral density (or autocorrelation) of the original image and noise are known. 3. Both original image and the noise are zero mean. 103

104 Iterative Wiener Filter: Iterative Algorithm: 1. R 0 f T i i i T 2. W 1 R H HR H R i i 3. fˆ 1 W 1 g ˆ ˆT i E i i 4. R 1 = f 1 f 1 f = R g 5. Repeat 2,3,4 until convergence. f f n 1 104

105 Example Iterative Wiener Filter): 105

106 Adaptive Wiener Filter: Image are Non Stationary! Need Adaptive WF which is locally optimal. Assume small region which image are stationary Image Model in each region:,,, f x y x y x y : zero-mean white noise with unit variance!, : Constant over each region. f f Noisy Image: f f g x, y f x, y v x, y, v :Constant over each region 106

107 Local Wiener Filter in each region: W a u, v P f,, f xy f x y g x y x y 2 ff f 2 2 ff Pvv f v 2 2 f v fˆ x, y g x, y w x, y f a f ˆ f x, y g x, y f f 2 f 2 2 f v, : Low-pass filtered on noisy image.,,, Zero mean assumption g x, y, : Hi-pass filtered on noisy image. f P 2 wa x y x y ˆ xy, f x, y HP x, y LP x, y 2 f 2 2 f xy, v 107

108 Parameter Estimation: 2 g 2 v Local Noisy Image Variance Variance in a smooth image region or background x y Max x y,,, f g v 108

109 Example (1): 109

110 Example (1): 110

111 Results: 111

112 Results: 112

113 Spatial Transform: x y r(x,y) s(x,y) x y tiepoints f(x,y) g(x,y ) original distorted 113

114 An example of Spatial Transform: x y c c 1 5 x c 2 x c 6 y c 3 y c xy c 7 4 xy c 8 114

115 Gray Level Interpolation: Main Problem: Non-Integer coordination! 115

116 Gray Level Interpolation: Nearest neighbor Bilinear interpolation Use 4 nearest neighbors Cubic convolution interpolation Fit a surface over a large number of neighbors 116

117 Example (1): original Tiepoints after distortion Nearest neighbor restored Bilinear restored 117

118 Example (2): original distorted Difference restored 118

119 Matlab Image Restoration Command: deconvblind: Restore image using blind deconvolution deconvlucy: Restore image using accelerated Richardson-Lucy algorithm deconvreg: Restore image using Regularized filter deconvwnr: Restore image using Wiener filter wiener2: Perform 2-D adaptive noise-removal filtering edgetaper: Taper the discontinuities along the image edges 119

Image Degradation Model (Linear/Additive)

Image Degradation Model (Linear/Additive) Image Degradation Model (Linear/Additive),,,,,,,, g x y h x y f x y x y G uv H uv F uv N uv 1 Source of noise Image acquisition (digitization) Image transmission Spatial properties of noise Statistical

More information

Digital Image Processing

Digital Image Processing Digital Image Processing Part 3: Fourier Transform and Filtering in the Frequency Domain AASS Learning Systems Lab, Dep. Teknik Room T109 (Fr, 11-1 o'clock) achim.lilienthal@oru.se Course Book Chapter

More information

Computer Vision & Digital Image Processing

Computer Vision & Digital Image Processing Computer Vision & Digital Image Processing Image Restoration and Reconstruction I Dr. D. J. Jackson Lecture 11-1 Image restoration Restoration is an objective process that attempts to recover an image

More information

COMP344 Digital Image Processing Fall 2007 Final Examination

COMP344 Digital Image Processing Fall 2007 Final Examination COMP344 Digital Image Processing Fall 2007 Final Examination Time allowed: 2 hours Name Student ID Email Question 1 Question 2 Question 3 Question 4 Question 5 Question 6 Total With model answer HK University

More information

Digital Image Processing. Filtering in the Frequency Domain

Digital Image Processing. Filtering in the Frequency Domain 2D Linear Systems 2D Fourier Transform and its Properties The Basics of Filtering in Frequency Domain Image Smoothing Image Sharpening Selective Filtering Implementation Tips 1 General Definition: System

More information

Wiener Filter for Deterministic Blur Model

Wiener Filter for Deterministic Blur Model Wiener Filter for Deterministic Blur Model Based on Ch. 5 of Gonzalez & Woods, Digital Image Processing, nd Ed., Addison-Wesley, 00 One common application of the Wiener filter has been in the area of simultaneous

More information

Empirical Mean and Variance!

Empirical Mean and Variance! Global Image Properties! Global image properties refer to an image as a whole rather than components. Computation of global image properties is often required for image enhancement, preceding image analysis.!

More information

Image Enhancement: Methods. Digital Image Processing. No Explicit definition. Spatial Domain: Frequency Domain:

Image Enhancement: Methods. Digital Image Processing. No Explicit definition. Spatial Domain: Frequency Domain: Image Enhancement: No Explicit definition Methods Spatial Domain: Linear Nonlinear Frequency Domain: Linear Nonlinear 1 Spatial Domain Process,, g x y T f x y 2 For 1 1 neighborhood: Contrast Enhancement/Stretching/Point

More information

ERRORS. 2. Diffraction of electromagnetic waves at aperture stop of the lens. telephoto lenses. 5. Row jittering synchronization of frame buffer

ERRORS. 2. Diffraction of electromagnetic waves at aperture stop of the lens. telephoto lenses. 5. Row jittering synchronization of frame buffer ERRORS סטיה העדשה (Hubble) 1. Lens Aberration 2. Diffraction of electromagnetic waves at aperture stop of the lens למקד 3. Defocusing תנועות ותנודות של המצלמה 4. Motions and vibrations of the camera telephoto

More information

Histogram Processing

Histogram Processing Histogram Processing The histogram of a digital image with gray levels in the range [0,L-] is a discrete function h ( r k ) = n k where r k n k = k th gray level = number of pixels in the image having

More information

MATHEMATICAL MODEL OF IMAGE DEGRADATION. = s

MATHEMATICAL MODEL OF IMAGE DEGRADATION. = s MATHEMATICAL MODEL OF IMAGE DEGRADATION H s u v G s u v F s u v ^ F u v G u v H s u v Gaussian Kernel Source: C. Rasmussen Gaussian filters pixel 5 pixels 0 pixels 30 pixels Gaussian filter Removes high-frequency

More information

Prof. Mohd Zaid Abdullah Room No:

Prof. Mohd Zaid Abdullah Room No: EEE 52/4 Advnced Digital Signal and Image Processing Tuesday, 00-300 hrs, Data Com. Lab. Friday, 0800-000 hrs, Data Com. Lab Prof. Mohd Zaid Abdullah Room No: 5 Email: mza@usm.my www.eng.usm.my Electromagnetic

More information

LPA-ICI Applications in Image Processing

LPA-ICI Applications in Image Processing LPA-ICI Applications in Image Processing Denoising Deblurring Derivative estimation Edge detection Inverse halftoning Denoising Consider z (x) =y (x)+η (x), wherey is noise-free image and η is noise. assume

More information

EE 367 / CS 448I Computational Imaging and Display Notes: Image Deconvolution (lecture 6)

EE 367 / CS 448I Computational Imaging and Display Notes: Image Deconvolution (lecture 6) EE 367 / CS 448I Computational Imaging and Display Notes: Image Deconvolution (lecture 6) Gordon Wetzstein gordon.wetzstein@stanford.edu This document serves as a supplement to the material discussed in

More information

Machine vision, spring 2018 Summary 4

Machine vision, spring 2018 Summary 4 Machine vision Summary # 4 The mask for Laplacian is given L = 4 (6) Another Laplacian mask that gives more importance to the center element is given by L = 8 (7) Note that the sum of the elements in the

More information

Convolution Spatial Aliasing Frequency domain filtering fundamentals Applications Image smoothing Image sharpening

Convolution Spatial Aliasing Frequency domain filtering fundamentals Applications Image smoothing Image sharpening Frequency Domain Filtering Correspondence between Spatial and Frequency Filtering Fourier Transform Brief Introduction Sampling Theory 2 D Discrete Fourier Transform Convolution Spatial Aliasing Frequency

More information

Machine vision. Summary # 4. The mask for Laplacian is given

Machine vision. Summary # 4. The mask for Laplacian is given 1 Machine vision Summary # 4 The mask for Laplacian is given L = 0 1 0 1 4 1 (6) 0 1 0 Another Laplacian mask that gives more importance to the center element is L = 1 1 1 1 8 1 (7) 1 1 1 Note that the

More information

Local enhancement. Local Enhancement. Local histogram equalized. Histogram equalized. Local Contrast Enhancement. Fig 3.23: Another example

Local enhancement. Local Enhancement. Local histogram equalized. Histogram equalized. Local Contrast Enhancement. Fig 3.23: Another example Local enhancement Local Enhancement Median filtering Local Enhancement Sometimes Local Enhancement is Preferred. Malab: BlkProc operation for block processing. Left: original tire image. 0/07/00 Local

More information

Local Enhancement. Local enhancement

Local Enhancement. Local enhancement Local Enhancement Local Enhancement Median filtering (see notes/slides, 3.5.2) HW4 due next Wednesday Required Reading: Sections 3.3, 3.4, 3.5, 3.6, 3.7 Local Enhancement 1 Local enhancement Sometimes

More information

Lecture 4 Filtering in the Frequency Domain. Lin ZHANG, PhD School of Software Engineering Tongji University Spring 2016

Lecture 4 Filtering in the Frequency Domain. Lin ZHANG, PhD School of Software Engineering Tongji University Spring 2016 Lecture 4 Filtering in the Frequency Domain Lin ZHANG, PhD School of Software Engineering Tongji University Spring 2016 Outline Background From Fourier series to Fourier transform Properties of the Fourier

More information

Intensity Transformations and Spatial Filtering: WHICH ONE LOOKS BETTER? Intensity Transformations and Spatial Filtering: WHICH ONE LOOKS BETTER?

Intensity Transformations and Spatial Filtering: WHICH ONE LOOKS BETTER? Intensity Transformations and Spatial Filtering: WHICH ONE LOOKS BETTER? : WHICH ONE LOOKS BETTER? 3.1 : WHICH ONE LOOKS BETTER? 3.2 1 Goal: Image enhancement seeks to improve the visual appearance of an image, or convert it to a form suited for analysis by a human or a machine.

More information

UNIT III IMAGE RESTORATION Part A Questions 1. What is meant by Image Restoration? Restoration attempts to reconstruct or recover an image that has been degraded by using a clear knowledge of the degrading

More information

Introduction to Computer Vision. 2D Linear Systems

Introduction to Computer Vision. 2D Linear Systems Introduction to Computer Vision D Linear Systems Review: Linear Systems We define a system as a unit that converts an input function into an output function Independent variable System operator or Transfer

More information

Lecture 04 Image Filtering

Lecture 04 Image Filtering Institute of Informatics Institute of Neuroinformatics Lecture 04 Image Filtering Davide Scaramuzza 1 Lab Exercise 2 - Today afternoon Room ETH HG E 1.1 from 13:15 to 15:00 Work description: your first

More information

Image Segmentation: Definition Importance. Digital Image Processing, 2nd ed. Chapter 10 Image Segmentation.

Image Segmentation: Definition Importance. Digital Image Processing, 2nd ed. Chapter 10 Image Segmentation. : Definition Importance Detection of Discontinuities: 9 R = wi z i= 1 i Point Detection: 1. A Mask 2. Thresholding R T Line Detection: A Suitable Mask in desired direction Thresholding Line i : R R, j

More information

Fourier Methods in Digital Signal Processing Final Exam ME 579, Spring 2015 NAME

Fourier Methods in Digital Signal Processing Final Exam ME 579, Spring 2015 NAME Fourier Methods in Digital Signal Processing Final Exam ME 579, Instructions for this CLOSED BOOK EXAM 2 hours long. Monday, May 8th, 8-10am in ME1051 Answer FIVE Questions, at LEAST ONE from each section.

More information

Computer Vision. Filtering in the Frequency Domain

Computer Vision. Filtering in the Frequency Domain Computer Vision Filtering in the Frequency Domain Filippo Bergamasco (filippo.bergamasco@unive.it) http://www.dais.unive.it/~bergamasco DAIS, Ca Foscari University of Venice Academic year 2016/2017 Introduction

More information

Reading. 3. Image processing. Pixel movement. Image processing Y R I G Q

Reading. 3. Image processing. Pixel movement. Image processing Y R I G Q Reading Jain, Kasturi, Schunck, Machine Vision. McGraw-Hill, 1995. Sections 4.-4.4, 4.5(intro), 4.5.5, 4.5.6, 5.1-5.4. 3. Image processing 1 Image processing An image processing operation typically defines

More information

Fourier Transforms 1D

Fourier Transforms 1D Fourier Transforms 1D 3D Image Processing Alireza Ghane 1 Overview Recap Intuitions Function representations shift-invariant spaces linear, time-invariant (LTI) systems complex numbers Fourier Transforms

More information

Deconvolution. Parameter Estimation in Linear Inverse Problems

Deconvolution. Parameter Estimation in Linear Inverse Problems Image Parameter Estimation in Linear Inverse Problems Chair for Computer Aided Medical Procedures & Augmented Reality Department of Computer Science, TUM November 10, 2006 Contents A naive approach......with

More information

EECS490: Digital Image Processing. Lecture #11

EECS490: Digital Image Processing. Lecture #11 Lecture #11 Filtering Applications: OCR, scanning Highpass filters Laplacian in the frequency domain Image enhancement using highpass filters Homomorphic filters Bandreject/bandpass/notch filters Correlation

More information

TRACKING and DETECTION in COMPUTER VISION Filtering and edge detection

TRACKING and DETECTION in COMPUTER VISION Filtering and edge detection Technischen Universität München Winter Semester 0/0 TRACKING and DETECTION in COMPUTER VISION Filtering and edge detection Slobodan Ilić Overview Image formation Convolution Non-liner filtering: Median

More information

G52IVG, School of Computer Science, University of Nottingham

G52IVG, School of Computer Science, University of Nottingham Image Transforms Fourier Transform Basic idea 1 Image Transforms Fourier transform theory Let f(x) be a continuous function of a real variable x. The Fourier transform of f(x) is F ( u) f ( x)exp[ j2πux]

More information

Adaptive Filtering. Squares. Alexander D. Poularikas. Fundamentals of. Least Mean. with MATLABR. University of Alabama, Huntsville, AL.

Adaptive Filtering. Squares. Alexander D. Poularikas. Fundamentals of. Least Mean. with MATLABR. University of Alabama, Huntsville, AL. Adaptive Filtering Fundamentals of Least Mean Squares with MATLABR Alexander D. Poularikas University of Alabama, Huntsville, AL CRC Press Taylor & Francis Croup Boca Raton London New York CRC Press is

More information

Single Exposure Enhancement and Reconstruction. Some slides are from: J. Kosecka, Y. Chuang, A. Efros, C. B. Owen, W. Freeman

Single Exposure Enhancement and Reconstruction. Some slides are from: J. Kosecka, Y. Chuang, A. Efros, C. B. Owen, W. Freeman Single Exposure Enhancement and Reconstruction Some slides are from: J. Kosecka, Y. Chuang, A. Efros, C. B. Owen, W. Freeman 1 Reconstruction as an Inverse Problem Original image f Distortion & Sampling

More information

Name of the Student: Problems on Discrete & Continuous R.Vs

Name of the Student: Problems on Discrete & Continuous R.Vs Engineering Mathematics 05 SUBJECT NAME : Probability & Random Process SUBJECT CODE : MA6 MATERIAL NAME : University Questions MATERIAL CODE : JM08AM004 REGULATION : R008 UPDATED ON : Nov-Dec 04 (Scan

More information

Today s lecture. Local neighbourhood processing. The convolution. Removing uncorrelated noise from an image The Fourier transform

Today s lecture. Local neighbourhood processing. The convolution. Removing uncorrelated noise from an image The Fourier transform Cris Luengo TD396 fall 4 cris@cbuuse Today s lecture Local neighbourhood processing smoothing an image sharpening an image The convolution What is it? What is it useful for? How can I compute it? Removing

More information

Name of the Student: Problems on Discrete & Continuous R.Vs

Name of the Student: Problems on Discrete & Continuous R.Vs Engineering Mathematics 08 SUBJECT NAME : Probability & Random Processes SUBJECT CODE : MA645 MATERIAL NAME : University Questions REGULATION : R03 UPDATED ON : November 07 (Upto N/D 07 Q.P) (Scan the

More information

Slide a window along the input arc sequence S. Least-squares estimate. σ 2. σ Estimate 1. Statistically test the difference between θ 1 and θ 2

Slide a window along the input arc sequence S. Least-squares estimate. σ 2. σ Estimate 1. Statistically test the difference between θ 1 and θ 2 Corner Detection 2D Image Features Corners are important two dimensional features. Two dimensional image features are interesting local structures. They include junctions of dierent types Slide 3 They

More information

Inverse problem and optimization

Inverse problem and optimization Inverse problem and optimization Laurent Condat, Nelly Pustelnik CNRS, Gipsa-lab CNRS, Laboratoire de Physique de l ENS de Lyon Decembre, 15th 2016 Inverse problem and optimization 2/36 Plan 1. Examples

More information

Digital Image Processing COSC 6380/4393

Digital Image Processing COSC 6380/4393 Digital Image Processing COSC 6380/4393 Lecture 11 Oct 3 rd, 2017 Pranav Mantini Slides from Dr. Shishir K Shah, and Frank Liu Review: 2D Discrete Fourier Transform If I is an image of size N then Sin

More information

HST.582J/6.555J/16.456J

HST.582J/6.555J/16.456J Blind Source Separation: PCA & ICA HST.582J/6.555J/16.456J Gari D. Clifford gari [at] mit. edu http://www.mit.edu/~gari G. D. Clifford 2005-2009 What is BSS? Assume an observation (signal) is a linear

More information

Image Enhancement in the frequency domain. GZ Chapter 4

Image Enhancement in the frequency domain. GZ Chapter 4 Image Enhancement in the frequency domain GZ Chapter 4 Contents In this lecture we will look at image enhancement in the frequency domain The Fourier series & the Fourier transform Image Processing in

More information

ENERGY METHODS IN IMAGE PROCESSING WITH EDGE ENHANCEMENT

ENERGY METHODS IN IMAGE PROCESSING WITH EDGE ENHANCEMENT ENERGY METHODS IN IMAGE PROCESSING WITH EDGE ENHANCEMENT PRASHANT ATHAVALE Abstract. Digital images are can be realized as L 2 (R 2 objects. Noise is introduced in a digital image due to various reasons.

More information

Time Series Analysis: 4. Linear filters. P. F. Góra

Time Series Analysis: 4. Linear filters. P. F. Góra Time Series Analysis: 4. Linear filters P. F. Góra http://th-www.if.uj.edu.pl/zfs/gora/ 2012 Linear filters in the Fourier domain Filtering: Multiplying the transform by a transfer function. g n DFT G

More information

MIT 2.71/2.710 Optics 10/31/05 wk9-a-1. The spatial frequency domain

MIT 2.71/2.710 Optics 10/31/05 wk9-a-1. The spatial frequency domain 10/31/05 wk9-a-1 The spatial frequency domain Recall: plane wave propagation x path delay increases linearly with x λ z=0 θ E 0 x exp i2π sinθ + λ z i2π cosθ λ z plane of observation 10/31/05 wk9-a-2 Spatial

More information

Digital Image Processing COSC 6380/4393

Digital Image Processing COSC 6380/4393 Digital Image Processing COSC 6380/4393 Lecture 13 Oct 2 nd, 2018 Pranav Mantini Slides from Dr. Shishir K Shah, and Frank Liu Review f 0 0 0 1 0 0 0 0 w 1 2 3 2 8 Zero Padding 0 0 0 0 0 0 0 1 0 0 0 0

More information

AdaptiveFilters. GJRE-F Classification : FOR Code:

AdaptiveFilters. GJRE-F Classification : FOR Code: Global Journal of Researches in Engineering: F Electrical and Electronics Engineering Volume 14 Issue 7 Version 1.0 Type: Double Blind Peer Reviewed International Research Journal Publisher: Global Journals

More information

Atmospheric Turbulence Effects Removal on Infrared Sequences Degraded by Local Isoplanatism

Atmospheric Turbulence Effects Removal on Infrared Sequences Degraded by Local Isoplanatism Atmospheric Turbulence Effects Removal on Infrared Sequences Degraded by Local Isoplanatism Magali Lemaitre 1, Olivier Laligant 1, Jacques Blanc-Talon 2, and Fabrice Mériaudeau 1 1 Le2i Laboratory, University

More information

EE5356 Digital Image Processing

EE5356 Digital Image Processing EE5356 Digital Image Processing INSTRUCTOR: Dr KR Rao Spring 007, Final Thursday, 10 April 007 11:00 AM 1:00 PM ( hours) (Room 111 NH) INSTRUCTIONS: 1 Closed books and closed notes All problems carry weights

More information

3. ESTIMATION OF SIGNALS USING A LEAST SQUARES TECHNIQUE

3. ESTIMATION OF SIGNALS USING A LEAST SQUARES TECHNIQUE 3. ESTIMATION OF SIGNALS USING A LEAST SQUARES TECHNIQUE 3.0 INTRODUCTION The purpose of this chapter is to introduce estimators shortly. More elaborated courses on System Identification, which are given

More information

Noise, Image Reconstruction with Noise!

Noise, Image Reconstruction with Noise! Noise, Image Reconstruction with Noise! EE367/CS448I: Computational Imaging and Display! stanford.edu/class/ee367! Lecture 10! Gordon Wetzstein! Stanford University! What s a Pixel?! photon to electron

More information

Filtering in the Frequency Domain

Filtering in the Frequency Domain Filtering in the Frequency Domain Outline Fourier Transform Filtering in Fourier Transform Domain 2/20/2014 2 Fourier Series and Fourier Transform: History Jean Baptiste Joseph Fourier, French mathematician

More information

IMAGE ENHANCEMENT II (CONVOLUTION)

IMAGE ENHANCEMENT II (CONVOLUTION) MOTIVATION Recorded images often exhibit problems such as: blurry noisy Image enhancement aims to improve visual quality Cosmetic processing Usually empirical techniques, with ad hoc parameters ( whatever

More information

Machine Learning. A Bayesian and Optimization Perspective. Academic Press, Sergios Theodoridis 1. of Athens, Athens, Greece.

Machine Learning. A Bayesian and Optimization Perspective. Academic Press, Sergios Theodoridis 1. of Athens, Athens, Greece. Machine Learning A Bayesian and Optimization Perspective Academic Press, 2015 Sergios Theodoridis 1 1 Dept. of Informatics and Telecommunications, National and Kapodistrian University of Athens, Athens,

More information

Fourier series: Any periodic signals can be viewed as weighted sum. different frequencies. view frequency as an

Fourier series: Any periodic signals can be viewed as weighted sum. different frequencies. view frequency as an Image Enhancement in the Frequency Domain Fourier series: Any periodic signals can be viewed as weighted sum of sinusoidal signals with different frequencies Frequency Domain: view frequency as an independent

More information

ITK Filters. Thresholding Edge Detection Gradients Second Order Derivatives Neighborhood Filters Smoothing Filters Distance Map Image Transforms

ITK Filters. Thresholding Edge Detection Gradients Second Order Derivatives Neighborhood Filters Smoothing Filters Distance Map Image Transforms ITK Filters Thresholding Edge Detection Gradients Second Order Derivatives Neighborhood Filters Smoothing Filters Distance Map Image Transforms ITCS 6010:Biomedical Imaging and Visualization 1 ITK Filters:

More information

From Fourier Series to Analysis of Non-stationary Signals - II

From Fourier Series to Analysis of Non-stationary Signals - II From Fourier Series to Analysis of Non-stationary Signals - II prof. Miroslav Vlcek October 10, 2017 Contents Signals 1 Signals 2 3 4 Contents Signals 1 Signals 2 3 4 Contents Signals 1 Signals 2 3 4 Contents

More information

Representing regions in 2 ways:

Representing regions in 2 ways: Representing regions in 2 ways: Based on their external characteristics (its boundary): Shape characteristics Based on their internal characteristics (its region): Both Regional properties: color, texture,

More information

SRI VIDYA COLLEGE OF ENGINEERING AND TECHNOLOGY UNIT 3 RANDOM PROCESS TWO MARK QUESTIONS

SRI VIDYA COLLEGE OF ENGINEERING AND TECHNOLOGY UNIT 3 RANDOM PROCESS TWO MARK QUESTIONS UNIT 3 RANDOM PROCESS TWO MARK QUESTIONS 1. Define random process? The sample space composed of functions of time is called a random process. 2. Define Stationary process? If a random process is divided

More information

ELEG 3143 Probability & Stochastic Process Ch. 6 Stochastic Process

ELEG 3143 Probability & Stochastic Process Ch. 6 Stochastic Process Department of Electrical Engineering University of Arkansas ELEG 3143 Probability & Stochastic Process Ch. 6 Stochastic Process Dr. Jingxian Wu wuj@uark.edu OUTLINE 2 Definition of stochastic process (random

More information

No. of dimensions 1. No. of centers

No. of dimensions 1. No. of centers Contents 8.6 Course of dimensionality............................ 15 8.7 Computational aspects of linear estimators.................. 15 8.7.1 Diagonalization of circulant andblock-circulant matrices......

More information

Computer Vision & Digital Image Processing. Periodicity of the Fourier transform

Computer Vision & Digital Image Processing. Periodicity of the Fourier transform Computer Vision & Digital Image Processing Fourier Transform Properties, the Laplacian, Convolution and Correlation Dr. D. J. Jackson Lecture 9- Periodicity of the Fourier transform The discrete Fourier

More information

Statistical and Adaptive Signal Processing

Statistical and Adaptive Signal Processing r Statistical and Adaptive Signal Processing Spectral Estimation, Signal Modeling, Adaptive Filtering and Array Processing Dimitris G. Manolakis Massachusetts Institute of Technology Lincoln Laboratory

More information

CHAPTER 4 PRINCIPAL COMPONENT ANALYSIS-BASED FUSION

CHAPTER 4 PRINCIPAL COMPONENT ANALYSIS-BASED FUSION 59 CHAPTER 4 PRINCIPAL COMPONENT ANALYSIS-BASED FUSION 4. INTRODUCTION Weighted average-based fusion algorithms are one of the widely used fusion methods for multi-sensor data integration. These methods

More information

Modern Navigation. Thomas Herring

Modern Navigation. Thomas Herring 12.215 Modern Navigation Thomas Herring Estimation methods Review of last class Restrict to basically linear estimation problems (also non-linear problems that are nearly linear) Restrict to parametric,

More information

Super-Resolution. Dr. Yossi Rubner. Many slides from Miki Elad - Technion

Super-Resolution. Dr. Yossi Rubner. Many slides from Miki Elad - Technion Super-Resolution Dr. Yossi Rubner yossi@rubner.co.il Many slides from Mii Elad - Technion 5/5/2007 53 images, ratio :4 Example - Video 40 images ratio :4 Example Surveillance Example Enhance Mosaics Super-Resolution

More information

DIFFERENTIATION RULES

DIFFERENTIATION RULES 3 DIFFERENTIATION RULES DIFFERENTIATION RULES 3. The Product and Quotient Rules In this section, we will learn about: Formulas that enable us to differentiate new functions formed from old functions by

More information

Tutorial 9 The Discrete Fourier Transform (DFT) SIPC , Spring 2017 Technion, CS Department

Tutorial 9 The Discrete Fourier Transform (DFT) SIPC , Spring 2017 Technion, CS Department Tutorial 9 The Discrete Fourier Transform (DFT) SIPC 236327, Spring 2017 Technion, CS Department The DFT Matrix The DFT matrix of size M M is defined as DFT = 1 M W 0 0 W 0 W 0 W where W = e i2π M i =

More information

Conjugate gradient acceleration of non-linear smoothing filters Iterated edge-preserving smoothing

Conjugate gradient acceleration of non-linear smoothing filters Iterated edge-preserving smoothing Cambridge, Massachusetts Conjugate gradient acceleration of non-linear smoothing filters Iterated edge-preserving smoothing Andrew Knyazev (knyazev@merl.com) (speaker) Alexander Malyshev (malyshev@merl.com)

More information

SIMG-782 Digital Image Processing Homework 6

SIMG-782 Digital Image Processing Homework 6 SIMG-782 Digital Image Processing Homework 6 Ex. 1 (Circular Convolution) Let f [1, 3, 1, 2, 0, 3] and h [ 1, 3, 2]. (a) Calculate the convolution f h assuming that both f and h are zero-padded to a length

More information

Inverse Problems in Image Processing

Inverse Problems in Image Processing H D Inverse Problems in Image Processing Ramesh Neelamani (Neelsh) Committee: Profs. R. Baraniuk, R. Nowak, M. Orchard, S. Cox June 2003 Inverse Problems Data estimation from inadequate/noisy observations

More information

Statistical signal processing

Statistical signal processing Statistical signal processing Short overview of the fundamentals Outline Random variables Random processes Stationarity Ergodicity Spectral analysis Random variable and processes Intuition: A random variable

More information

Time Series Analysis: 4. Digital Linear Filters. P. F. Góra

Time Series Analysis: 4. Digital Linear Filters. P. F. Góra Time Series Analysis: 4. Digital Linear Filters P. F. Góra http://th-www.if.uj.edu.pl/zfs/gora/ 2018 Linear filters Filtering in Fourier domain is very easy: multiply the DFT of the input by a transfer

More information

Chapter 4: Filtering in the Frequency Domain. Fourier Analysis R. C. Gonzalez & R. E. Woods

Chapter 4: Filtering in the Frequency Domain. Fourier Analysis R. C. Gonzalez & R. E. Woods Fourier Analysis 1992 2008 R. C. Gonzalez & R. E. Woods Properties of δ (t) and (x) δ : f t) δ ( t t ) dt = f ( ) f x) δ ( x x ) = f ( ) ( 0 t0 x= ( 0 x0 1992 2008 R. C. Gonzalez & R. E. Woods Sampling

More information

GENG2140, S2, 2012 Week 7: Curve fitting

GENG2140, S2, 2012 Week 7: Curve fitting GENG2140, S2, 2012 Week 7: Curve fitting Curve fitting is the process of constructing a curve, or mathematical function, f(x) that has the best fit to a series of data points Involves fitting lines and

More information

ECG782: Multidimensional Digital Signal Processing

ECG782: Multidimensional Digital Signal Processing Professor Brendan Morris, SEB 3216, brendan.morris@unlv.edu ECG782: Multidimensional Digital Signal Processing Filtering in the Frequency Domain http://www.ee.unlv.edu/~b1morris/ecg782/ 2 Outline Background

More information

Feature detectors and descriptors. Fei-Fei Li

Feature detectors and descriptors. Fei-Fei Li Feature detectors and descriptors Fei-Fei Li Feature Detection e.g. DoG detected points (~300) coordinates, neighbourhoods Feature Description e.g. SIFT local descriptors (invariant) vectors database of

More information

SYDE 575: Introduction to Image Processing. Image Compression Part 2: Variable-rate compression

SYDE 575: Introduction to Image Processing. Image Compression Part 2: Variable-rate compression SYDE 575: Introduction to Image Processing Image Compression Part 2: Variable-rate compression Variable-rate Compression: Transform-based compression As mentioned earlier, we wish to transform image data

More information

INTRODUCTION Noise is present in many situations of daily life for ex: Microphones will record noise and speech. Goal: Reconstruct original signal Wie

INTRODUCTION Noise is present in many situations of daily life for ex: Microphones will record noise and speech. Goal: Reconstruct original signal Wie WIENER FILTERING Presented by N.Srikanth(Y8104060), M.Manikanta PhaniKumar(Y8104031). INDIAN INSTITUTE OF TECHNOLOGY KANPUR Electrical Engineering dept. INTRODUCTION Noise is present in many situations

More information

Aperture Photometry in Band Limited. Data. Robert Lupton. March 4, Band limited data, that is, data that has non-zero Fourier components in

Aperture Photometry in Band Limited. Data. Robert Lupton. March 4, Band limited data, that is, data that has non-zero Fourier components in Aperture Photometry in Band Limited Data Robert Lupton March 4, 996 Band Limited Data Band limited data, that is, data that has non-zero Fourier components in only a restricted part of k-space jkj < k,

More information

Digital Image Processing. Chapter 4: Image Enhancement in the Frequency Domain

Digital Image Processing. Chapter 4: Image Enhancement in the Frequency Domain Digital Image Processing Chapter 4: Image Enhancement in the Frequency Domain Image Enhancement in Frequency Domain Objective: To understand the Fourier Transform and frequency domain and how to apply

More information

Image enhancement. Why image enhancement? Why image enhancement? Why image enhancement? Example of artifacts caused by image encoding

Image enhancement. Why image enhancement? Why image enhancement? Why image enhancement? Example of artifacts caused by image encoding 13 Why image enhancement? Image enhancement Example of artifacts caused by image encoding Computer Vision, Lecture 14 Michael Felsberg Computer Vision Laboratory Department of Electrical Engineering 12

More information

HST.582J / 6.555J / J Biomedical Signal and Image Processing Spring 2007

HST.582J / 6.555J / J Biomedical Signal and Image Processing Spring 2007 MIT OpenCourseWare http://ocw.mit.edu HST.582J / 6.555J / 16.456J Biomedical Signal and Image Processing Spring 2007 For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.

More information

12.1 Taylor Polynomials In Two-Dimensions: Nonlinear Approximations of F(x,y) Polynomials in Two-Variables. Taylor Polynomials and Approximations.

12.1 Taylor Polynomials In Two-Dimensions: Nonlinear Approximations of F(x,y) Polynomials in Two-Variables. Taylor Polynomials and Approximations. Week 12 Lecture Overheads Math2080 Week 12 Page 1 12.1 Taylor Polynomials In Two-Dimensions: Nonlinear Approximations of F(x,y) Polynomials in Two-Variables. Taylor Polynomials and Approximations. 12.2

More information

Erkut Erdem. Hacettepe University February 24 th, Linear Diffusion 1. 2 Appendix - The Calculus of Variations 5.

Erkut Erdem. Hacettepe University February 24 th, Linear Diffusion 1. 2 Appendix - The Calculus of Variations 5. LINEAR DIFFUSION Erkut Erdem Hacettepe University February 24 th, 2012 CONTENTS 1 Linear Diffusion 1 2 Appendix - The Calculus of Variations 5 References 6 1 LINEAR DIFFUSION The linear diffusion (heat)

More information

Sparse Matrix Techniques for MCAO

Sparse Matrix Techniques for MCAO Sparse Matrix Techniques for MCAO Luc Gilles lgilles@mtu.edu Michigan Technological University, ECE Department Brent Ellerbroek bellerbroek@gemini.edu Gemini Observatory Curt Vogel vogel@math.montana.edu

More information

Feature detectors and descriptors. Fei-Fei Li

Feature detectors and descriptors. Fei-Fei Li Feature detectors and descriptors Fei-Fei Li Feature Detection e.g. DoG detected points (~300) coordinates, neighbourhoods Feature Description e.g. SIFT local descriptors (invariant) vectors database of

More information

Signals and Spectra - Review

Signals and Spectra - Review Signals and Spectra - Review SIGNALS DETERMINISTIC No uncertainty w.r.t. the value of a signal at any time Modeled by mathematical epressions RANDOM some degree of uncertainty before the signal occurs

More information

Using Entropy and 2-D Correlation Coefficient as Measuring Indices for Impulsive Noise Reduction Techniques

Using Entropy and 2-D Correlation Coefficient as Measuring Indices for Impulsive Noise Reduction Techniques Using Entropy and 2-D Correlation Coefficient as Measuring Indices for Impulsive Noise Reduction Techniques Zayed M. Ramadan Department of Electronics and Communications Engineering, Faculty of Engineering,

More information

Lecture 8: Interest Point Detection. Saad J Bedros

Lecture 8: Interest Point Detection. Saad J Bedros #1 Lecture 8: Interest Point Detection Saad J Bedros sbedros@umn.edu Last Lecture : Edge Detection Preprocessing of image is desired to eliminate or at least minimize noise effects There is always tradeoff

More information

LINEARIZED BREGMAN ITERATIONS FOR FRAME-BASED IMAGE DEBLURRING

LINEARIZED BREGMAN ITERATIONS FOR FRAME-BASED IMAGE DEBLURRING LINEARIZED BREGMAN ITERATIONS FOR FRAME-BASED IMAGE DEBLURRING JIAN-FENG CAI, STANLEY OSHER, AND ZUOWEI SHEN Abstract. Real images usually have sparse approximations under some tight frame systems derived

More information

Image Enhancement in the frequency domain. Inel 5046 Prof. Vidya Manian

Image Enhancement in the frequency domain. Inel 5046 Prof. Vidya Manian Image Enhancement in the frequency domain Inel 5046 Prof. Vidya Manian Introduction 2D Fourier transform Basics of filtering in frequency domain Ideal low pass filter Gaussian low pass filter Ideal high

More information

Basics on 2-D 2 D Random Signal

Basics on 2-D 2 D Random Signal Basics on -D D Random Signal Spring 06 Instructor: K. J. Ray Liu ECE Department, Univ. of Maryland, College Park Overview Last Time: Fourier Analysis for -D signals Image enhancement via spatial filtering

More information

Image Noise: Detection, Measurement and Removal Techniques. Zhifei Zhang

Image Noise: Detection, Measurement and Removal Techniques. Zhifei Zhang Image Noise: Detection, Measurement and Removal Techniques Zhifei Zhang Outline Noise measurement Filter-based Block-based Wavelet-based Noise removal Spatial domain Transform domain Non-local methods

More information

Numerical Linear Algebra and. Image Restoration

Numerical Linear Algebra and. Image Restoration Numerical Linear Algebra and Image Restoration Maui High Performance Computing Center Wednesday, October 8, 2003 James G. Nagy Emory University Atlanta, GA Thanks to: AFOSR, Dave Tyler, Stuart Jefferies,

More information

Part III Super-Resolution with Sparsity

Part III Super-Resolution with Sparsity Aisenstadt Chair Course CRM September 2009 Part III Super-Resolution with Sparsity Stéphane Mallat Centre de Mathématiques Appliquées Ecole Polytechnique Super-Resolution with Sparsity Dream: recover high-resolution

More information

BME 50500: Image and Signal Processing in Biomedicine. Lecture 5: Correlation and Power-Spectrum CCNY

BME 50500: Image and Signal Processing in Biomedicine. Lecture 5: Correlation and Power-Spectrum CCNY 1 BME 50500: Image and Signal Processing in Biomedicine Lecture 5: Correlation and Power-Spectrum Lucas C. Parra Biomedical Engineering Department CCNY http://bme.ccny.cuny.edu/faculty/parra/teaching/signal-and-image/

More information

EE538 Final Exam Fall :20 pm -5:20 pm PHYS 223 Dec. 17, Cover Sheet

EE538 Final Exam Fall :20 pm -5:20 pm PHYS 223 Dec. 17, Cover Sheet EE538 Final Exam Fall 005 3:0 pm -5:0 pm PHYS 3 Dec. 17, 005 Cover Sheet Test Duration: 10 minutes. Open Book but Closed Notes. Calculators ARE allowed!! This test contains five problems. Each of the five

More information

Final Examination. CS 205A: Mathematical Methods for Robotics, Vision, and Graphics (Fall 2013), Stanford University

Final Examination. CS 205A: Mathematical Methods for Robotics, Vision, and Graphics (Fall 2013), Stanford University Final Examination CS 205A: Mathematical Methods for Robotics, Vision, and Graphics (Fall 2013), Stanford University The exam runs for 3 hours. The exam contains eight problems. You must complete the first

More information