Digital Image Processing. Lecture # 4 Image Enhancement (Histogram)

Transcription

1 Digital Image Processing Lecture # 4 Image Enhancement (Histogram) 1

2 Histogram of a Grayscale Image Let I be a 1-band (grayscale) image. I(r,c) is an 8-bit integer between 0 and 255. Histogram, h I, of I: a 256-element array, h I h I (g) = number of pixels in I that have value g. for g = 0,1, 2, 3,, 255 2

3 HISTOGRAM A discrete function h(r k )=n k r k is the k th gray level n k is the number of pixels having gray level r k in the image Ex: n k r k 3

4 4 UNIQUENESS

5 Histogram of a Grayscale Image Histogram of a digital image with gray levels in the range [0,L-1] is a discrete function Where r k = k th gray level h( r ) k n k n k = number of pixels in the image having gray level r k h(r k ) = histogram of an image having r k gray levels 5

6 Normalized Histogram Dividing each of histogram at gray level r k by the total number of pixels in the image, n p( r ) n / n for k 0,1,, L 1 k k p(r k ) gives an estimate of the probability of occurrence of gray level r k The sum of all components of a normalized histogram is equal to 1 6

7 Histogram of a Grayscale Image 16-level (4-bit) image lower RHC: number of pixels with intensity g black marks pixels with intensity g 7

8 Histogram of a Grayscale Image Black marks pixels with intensity g Plot of histogram: number of pixels with intensity g 8

9 Histogram of a Grayscale Image Black marks pixels with intensity g Plot of histogram: number of pixels with intensity g 9

10 Histogram of a Grayscale Image hi g the number of pixels in I with graylevel g. 10

11 Histogram of a Color Image If I is a 3-band image then I(r,c,b) is an integer between 0 and 255. I has 3 histograms: h R (g) = # of pixels in I(:,:,1) with intensity value g h G (g) = # of pixels in I(:,:,2) with intensity value g h B (g) = # of pixels in I(:,:,3) with intensity value g 11

12 Histogram of a Color Image There is one histogram per color band R, G, & B. Luminosity histogram is from 1 band = (R+G+B)/3 12

13 Histogram of a Color Image 13

14 Histogram of a Color Image 14

15 Histogram: Example Dark Image How would the histograms of these images look like? Bright Image 15

16 Histogram: Example Dark image Components of histogram are concentrated on the low side of the gray scale Bright image Components of histogram are concentrated on the high side of the gray scale 16

17 17 HISTOGRAM INSIGHT INTO CONTRAST

18 Histogram: Example Low Contrast Image How would the histograms of these images look like? High Contrast Image 18

19 Histogram: Example Low contrast image Histogram is narrow and centered toward the middle of the gray scale High contrast image Histogram covers broad range of the gray scale and the distribution of pixels is not too far from uniform with very few vertical lines being much higher than the others 19

20 Contrast Stretching Improve the contrast in an image by `stretching' the range of intensity values it contains to span a desired range of values, e.g. the the full range of pixel values 20

21 Contrast Stretching If r max and r min are the maximum and minimum gray level of the input image and L is the total gray levels of output image, the transformation function for contrast stretch will be s T r r r L 1 ( ) ( min ) rmax rmin L 0 r min 127 r max

22 Contrast Stretching 22

23 Histogram Equalization Histogram equalization re-assigns the intensity values of pixels in the input image such that the output image contains a uniform distribution of intensities 23

24 24 HISTOGRAM EQUALIZATION

25 AERIAL PHOTOGRAPH OF THE PENTAGON 25 Resulting image uses more of dynamic range. Resulting histogram almost, but not completely, flat.

26 The Probability Distribution Function Let 255 g0 I A h g Note that since I with value g, hi g of an Image is the number of pixels in A is the number of pixels in I. That is if I is R rows by C columns then A RC. Then, 1 pi g hi g A This is the probability that an arbitrary pixel from I has value g. 26

27 The Probability Distribution Function of an Image p(g) is the fraction of pixels in an image that have intensity value g. p(g) is the probability that a pixel randomly selected from the given image has intensity value g. Whereas the sum of the histogram h(g) over all g from 0 to 255 is equal to the number of pixels in the image, the sum of p(g) over all g is 1. p is the normalized histogram of the image 27

28 The Cumulative Distribution Function of an Image Let q = I(r,c) be the value of a randomly selected pixel from I. Let g be a specific gray level. The probability that q g is given by g g 1 P g p h I I I 0 A 0 g h h I I, where h I (γ ) is the histogram of image I. This is the probability that any given pixel from I has value less than or equal to g. 28

29 The Cumulative Distribution Function of an Image Let q = I(r,c) be the value of a randomly selected pixel from I. Let g be a specific gray level. The probability that q g is given by g g 1 P g p h I I I 0 A 0 g h h I I, Also called CDF for Cumulative Distribution Function. where h I (γ ) is the histogram of image I. This is the probability that any given pixel from I has value less than or equal to g. 29

30 The Cumulative Distribution Function of an Image P(g) is the fraction of pixels in an image that have intensity values less than or equal to g. P(g) is the probability that a pixel randomly selected from the given band has an intensity value less than or equal to g. P(g) is the cumulative (or running) sum of p(g) from 0 through g inclusive. P(0) = p(0) and P(255) = 1; 30

31 Histogram Equalization Task: remap image I so that its histogram is as close to constant as possible Let PI be the cumulative (probability) distribution function of I. The CDF itself is used as the LUT. 31

32 Histogram Equalization pdf The CDF (cumulative distribution) is the LUT for remapping. CDF 32

33 Histogram Equalization pdf The CDF (cumulative distribution) is the LUT for remapping. LUT 33

34 Histogram Equalization pdf The CDF (cumulative distribution) is the LUT for remapping. LUT 34

35 Histogram Equalization 35

36 Histogram Equalization Luminosity before J r, c 255 PI I r, c. after 36

37 HISTOGRAM EQUALIZATION IMPLEMENTATION Gray levels Counts (h(r k )) r 0 r 1 r 2 r 3 r 4 r 5 r 6 Normalized h (P(r k )) 5/20 4/ /20 1/20 3/20 0 4/20 1/20 cdf F(r k ) 5/20 9/20 11/20 12/20 15/20 19/20 20/20 s k =round(9 F(r k )) s 0 s 1 s 2 s 3 s 4 s 5 s 6 37

38 Histogram Equalization: Example An 8x8 image 38

39 Histogram Equalization: Example Fill in the following table/histogram Image Histogram (Non-zero values) 39

40 Histogram Equalization: Example Image Histogram (Non-zero values shown) 40

41 Histogram Equalization: Example 41

42 Histogram Equalization: Example Cumulative Distribution Function (cdf) Image Histogram/Prob Mass Function 42

43 Histogram Equalization: Example Cumulative Distribution Function (cdf) 43

44 Histogram Equalization: Example Cumulative Distribution Function (cdf) 44

45 Histogram Equalization: Example Normalized Cumulative Distribution Function (cdf) Divide each value by total number of pixels (64) to get the normalized cdf 45

46 Histogram Equalization: Example If cdf is normalized If cdf is NOT normalized s s J r, c 255 PI I r, c. s round(255. cdf ( r)) cdf () r round (255. ) M N round ( / 64 ) s 183 Original Image

47 Histogram Equalization: Example 47

48 Histogram Equalization: Example Original Image Corresponding histogram (red) and cumulative histogram (black) Image after histogram equalization Corresponding histogram (red) and cumulative histogram (black) 48

49 Bright image Dark image Histogram Equalization: Example Equalized Histogram Equalized Histogram 49

50 High Contrast Low contrast Histogram Equalization: Example Equalized Histogram Equalized Histogram 50

51 HISTOGRAM MATCHING (SPECIFICATION) HISTOGRAM EQUALIZATION DOES NOT ALLOW INTERACTIVE IMAGE ENHANCEMENT AND GENERATES ONLY ONE RESULT: AN APPROXIMATION TO A UNIFORM HISTOGRAM. SOMETIMES THOUGH, WE NEED TO BE ABLE TO SPECIFY PARTICULAR HISTOGRAM SHAPES CAPABLE OF HIGHLIGHTING CERTAIN GRAY-LEVEL RANGES. 54

52 HISTOGRAM SPECIFICATION 55

53 HISTOGRAM SPECIFICATION THE PROCEDURE FOR HISTOGRAM-SPECIFICATION BASED ENHANCEMENT IS: EQUALIZE THE LEVELS OF THE ORIGINAL IMAGE USING: s T( r k ) k j0 n j n n: total number of pixels, nj: number of pixels with gray level rj, L: number of discrete gray levels 56

54 HISTOGRAM SPECIFICATION SPECIFY THE DESIRED DENSITY FUNCTION AND OBTAIN THE TRANSFORMATION FUNCTION G(z): v k k k pz zi sk G z i0 pz: specified desirable PDF for output 57

55 HISTOGRAM SPECIFICATION THE NEW, PROCESSED VERSION OF THE ORIGINAL IMAGE CONSISTS OF GRAY LEVELS CHARACTERIZED BY THE SPECIFIED DENSITY p z (z). In essence: z G 1 ( s) z G 1 [ T( r)] 58

56 59 MAPPINGS

57 HISTOGRAM SPECIFICATION OBTAIN THE HISTOGRAM OF THE GIVEN IMAGE MAP EACH LEVEL r K TO A LEVEL S K OBTAIN THE TRANSFORMATION FUNCTION G FROM THE GIVEN P Z (Z) PRECOMPUTE Z K FOR EACH VALUE OF S K FOR EACH PIXEL IN THE ORIGINAL IMAGE, IF THE VALUE OF THAT PIXEL IS r k MAP THIS VALUE TO ITS CORRESPONDING LEVEL S K, THEN MAP LEVEL S K INTO THE FINAL VALUE Z K 60

58 HISTOGRAM SPECIFICATION k n k p r (r k ) s k p z (z k ) v k n k A 64X64 (4096 PIXELS) IMAGE WITH 8 GRAY LEVELS 61

59 62 IMAGE ENHANCEMENT IN THE SPATIAL DOMAIN

60 63 IMAGE ENHANCEMENT IN THE SPATIAL DOMAIN

61 64

62 GLOBAL/LOCAL HISTOGRAM EQUALIZATION IT MAY BE NECESSARY TO ENHANCE DETAILS OVER SMALL AREAS IN THE IMAGE THE NUMBER OF PIXELS IN THESE AREAS MAY HAVE NEGLIGIBLE INFLUENCE ON THE COMPUTATION OF A GLOBAL TRANSFORMATION WHOSE SHAPE DOES NOT NECESSARILY GUARANTEE THE DESIRED LOCAL ENHANCEMENT DEVISE TRANSFORMATION FUNCTIONS BASED ON THE GRAY LEVEL DISTRIBUTION IN THE NEIGHBORHOOD OF EVERY PIXEL IN THE IMAGE THE PROCEDURE IS: DEFINE A SQUARE (OR RECTANGULAR) NEIGHBORHOOD AND MOVE THE CENTER OF THIS AREA FROM PIXEL TO PIXEL. AT EACH LOCATION, THE HISTOGRAM OF THE POINTS IN THE NEIGHBORHOOD IS COMPUTED AND EITHER A HISTOGRAM EQUALIZATION OR HISTOGRAM SPECIFICATION TRANSFORMATION FUNCTION IS OBTAINED. THIS FUNCTION IS FINALLY USED TO MAP THE GRAY LEVEL OF THE PIXEL CENTERED IN THE NEIGHBORHOOD. THE CENTER IS THEN MOVED TO AN ADJACENT PIXEL LOCATION AND THE PROCEDURE IS REPEATED. 65

63 66 GLOBAL/LOCAL HISTOGRAM EQUALIZATION

64 USE OF HISTOGRAM STATISTICS FOR IMAGE ENHANCEMENT (Global) LET r REPRESENT A GRAY LEVEL IN THE IMAGE [0, L-1], AND LET p(r i ) DENOTE THE NORMALIZED HISTOGRAM COMPONENT CORRESPONDING TO THE i th VALUE OF r. THE n th MOMENT OF r ABOUT ITS MEAN IS DEFINED AS n L 1 i0 r r m pr WHERE m IS THE MEAN VALUE OF r (AVERAGE GRAY LEVEL) i n i m L 1 0 r p r i i i 67

65 USE OF HISTOGRAM STATISTICS FOR IMAGE ENHANCEMENT (Global) THE SECOND MOMENT IS GIVEN BY L r r m pr 2 i0 WHICH IS THE VARIANCE OF r 1 MEAN AS A MEASURE OF AVERAGE GRAY LEVEL IN THE IMAGE VARIANCE AS A MEASURE OF AVERAGE CONTRAST i 2 i 68

66 USE OF HISTOGRAM STATISTICS FOR IMAGE ENHANCEMENT (Local) LET (x,y) BE THE COORDINATES OF A PIXEL IN AN IMAGE, AND LET S X,Y DENOTE A NEIGBORHOOD OF SPECIFIED SIZE, CENTERED AT (x,y) THE MEAN VALUE m SXY OF THE PIXELS IN S X,Y IS m THE GRAY LEVEL VARIANCE OF THE PIXELS IN REGION S X,Y IS GIVEN BY r p r sxy s, t s, t s, t S xy S 2 2 r s t ms p r xy, s, s, t S xy xy t 69

67 USE OF HISTOGRAM STATISTICS FOR IMAGE ENHANCEMENT THE GLOBAL MEAN AND VARIANCE ARE MEASURED OVER AN ENTIRE IMAGE AND ARE USEFUL FOR GROSS ADJUSTMENTS OF OVERALL INTENSITY AND CONTRAST. A USE OF THESE MEASURES IN LOCAL ENHANCEMENT IS, WHERE THE LOCAL MEAN AND VARIANCE ARE USED AS THE BASIS FOR MAKING CHANGES THAT DEPEND ON IMAGE CHARACTERISTICS IN A PREDEFINED REGION ABOUT EACH PIXEL IN THE IMAGE. 70

68 71 TUNGSTEN FILAMENT IMAGE

69 USE OF HISTOGRAM STATISTICS FOR IMAGE ENHANCEMENT A PIXEL AT POINT (x,y) IS CONSIDERED IF: m SXY k 0 M G, where k0 is a positive constant less than 1.0, and M G is global mean σ sxy k 2 D G, where D G is the global standard deviation and k 2 is a positive constant k 1 D G σ sxy,, with k1 < k2 A PIXEL THAT MEETS ALL ABOVE CONDITIONS IS PROCESSED SIMPLY BY MULTIPLYING IT BY A SPECIFIED CONSTANT, E, TO INCREASE OR DECREASE THE VALUE OF ITS GRAY LEVEL RELATIVE TO THE REST OF THE IMAGE. THE VALUES OF PIXELS THAT DO NOT MEET THE ENHANCEMENT CONDITIONS ARE LEFT UNCHANGED. 72

70 73 IMAGE ENHANCEMENT IN THE SPATIAL DOMAIN

71 74 IMAGE ENHANCEMENT IN THE SPATIAL DOMAIN

72 Readings from Book (3 rd Edn.) 3.3 Histogram

73 Material in these slides has been taken from, the following resources Acknowledgements Digital Image Processing, Rafael C. Gonzalez & Richard E. Woods, Addison-Wesley,