Color Image Processing

Transcription

1 Color Image Processing

2 Color Fundamentals 2/27/2014 2

3 Color Fundamentals 2/27/2014 3

4 Color Fundamentals 6 to 7 million cones in the human eye can be divided into three principal sensing categories, corresponding roughly to red, green, and blue. 65%: red 33%: green 2%: blue (blue cones are the most sensitive) 2/27/2014 4

5 Color Fundamentals 2/27/2014 5

6 2/27/2014 6

7 Color Fundamentals The characteristics generally used to distinguish one color from another are brightness, hue, and saturation brightness: the achromatic notion of intensity. hue: dominant wavelength in a mixture of light waves, represents dominant color as perceived by an observer. saturation: relative purity or the amount of white light mixed with its hue. 2/27/2014 7

8 Color Fundamentals Tristimulus Red, green, and blue are denoted X, Y, and Z, respectively. A color is defined by its trichromatic coefficients, defined as x y z = = = X X + Y + Z Y X + Y + Z Z X + Y + Z 2/27/2014 8

9 CIE Chromaticity Diagram It shows color composition as a function of x (red) and y (green) 2/27/2014 9

10 RGB Color Model 2/27/

11 RGB Color Model Pixel depth The total number of colors in a 24-bit RGB image is (2 8 ) 3 = 16,777,216 2/27/

12 2/27/

13 Safe RGB colors (or safe Web colors) are reproduced faithfully, reasonably independently of viewer hardware capabilities 2/27/

14 2/27/

15 The CMY and CMYK Color Models C 1 R M 1 G = Y 1 B Equal amounts of the pigment primaries, cyan, magenta, and yellow should produce black. In practice, combining these colors for printing produces a muddy-looking black. To produce true black, the predominant color in printing, the fourth color, black, is added, giving rise to the CMYK color model. 2/27/

16 CMY vs. CMYK 2/27/

17 HSI Color Model brightness: the achromatic notion of intensity. hue: dominant wavelength in a mixture of light waves, represents dominant color as perceived by an observer. saturation: relative purity or the amount of white light mixed with its hue. 2/27/

18 HSI Color Model 2/27/

19 HSI Color Model 2/27/

20 HSI Color Model 2/27/

21 Converting Colors from RGB to HSI Given an image in RGB color format, the H component of each RGB pixel is obtained using the equation H θ if B G = 360 θ if B>G θ 1 [ ( R G ) + ( R B ) ] 2 R G + ( R B)( G B) 1 = cos 2 1/2 ( ) 2/27/

22 Converting Colors from RGB to HSI Given an image in RGB color format, the saturation component is given by S 3 = 1 min(,, ) ( R+ G+ B) [ RGB] 2/27/

23 Converting Colors from RGB to HSI Given an image in RGB color format, the intensity component is given by 1 I = R+ G+ B 3 ( ) 2/27/

24 Converting Colors from HSI to RGB RG sector (0 H < 120 ) B= I(1 S) Scos H R= I 1 + cos(60 H ) and G = 3 I ( R+ B) 2/27/

25 Converting Colors from HSI to RGB RG sector (120 H < 240 ) H = H 120 R= I(1 S) Scos H G = I 1 + cos(60 H ) and B= 3 I ( R+ G) 2/27/

26 Converting Colors from HSI to RGB RG sector (240 H 360 ) H = H 240 G = I(1 S) Scos H B= I 1 + cos(60 H ) and R= 3 I ( G+ B) 2/27/

27 2/27/

28 2/27/

29 2/27/

30 Pseudocolor Image Processing The process of assigning colors to gray values based on a specified criterion. Intensity Slicing f( xy, ) = c if f( xy, ) V k k 2/27/

31 2/27/

32 2/27/

33 2/27/

34 2/27/

35 2/27/

36 Pseudocolor Image Processing Intensity to Color Transformation 2/27/

37 The images are obtained from an airport X-ray scanning system. The left contains ordinary articles and the right contains the same articles as well as a block of simulated plastic explosives. 2/27/

38 2/27/

39 2/27/

40 2/27/

41 Pseudocolor by combining several of the sensor images from the Galileo spacecraft, some of which are in spectral regions not visible to the eye. Bright red depicts materials newly ejected from an active volcano on Io, and the surrounding yellow materials are older sulfur deposits. 2/27/

42 Basics of Full-Color Image Processing Let c represent an arbitrary vector in RGB color space: cr R c= c G G = c B B At coordinates ( xy, ), cr ( xy, ) Rxy (, ) cxy (, ) = cg ( xy, ) Gxy (, ) = cb ( xy, ) Bxy (, ) 2/27/

43 Basics of Full-Color Image Processing 2/27/

44 Color Transformations gxy (, ) = T[ f( xy, )] s = T( r, r,..., r ), i = 1,2,..., n. i i 1 2 n 2/27/

45 2/27/

46 gxy (, ) = kf( xy, ) s i = kr, i = 1, 2,3. i s = kr + (1 k), i i = 1, 2,3. i 3 3 2/27/ s = kr

47 2/27/

48 2/27/

49 Color slicing Highlighting a specific range of colors in an image If the colors of interest are enclosed by a cube of width W and centered at a protypical color with components ( a, a,..., a ), the necessary set of transformations is 1 2 n 0.5 if rj aj W /2 any 1 s i = > ri otherwise j n 2/27/

50 Color slicing If a sphere is used to specify the colors of interest, R is the radius of the enclosing of its center. 0 The transformations is s i n 0.5 if = j= 1 ri ( ) j j 2 r a > R 2 0 otherwise 2/27/

51 Color slicing 2/27/

52 Tone and Color Corrections 2/27/

53 2/27/

54 2/27/

55 Color Image Smoothing Let S xy denote the set of coordinates defining a neighborhood centered at ( xy, ) in an RGB color image. The average of the RGB component vectors in this neighborhood is 1 cxy (, ) = cst (,) K ( st,) S xy 1 Rst (,) K ( st,) Sxy 1 = Gst (,) K ( st,) Sxy 1 Bst (,) K ( st,) Sxy 2/27/

56 2/27/

57 2/27/

58 2/27/

59 Color Image Sharpening The Laplacian of vector c is 2 Rxy (, ) (, ) = Gxy (, ) 2 Bxy (, ) [ cxy] 2 2 2/27/

60 2/27/

61 Image Segmentation Based on Color: Segmentation in HSI Color Space 2/27/

62 Segmentation in RGB Vector Space Let the average color of interest is denoted by the RGB vector a. Let z denote an arbitrary point in RGB space. T D( z, a) = z a = ( z a) ( z a) 1/2 = ( z a ) + ( z a ) + ( z a ) R R G G B B 1/2 2/27/

63 2/27/

64 Color Edge Detection (1) Let r, g, and b be unit vectors along the R, G, and B axis of RGB color space, and define vectors R G B u = r + g + b x x x and R G B v = r + g + b y y y 2/27/

65 Color Edge Detection (2) 2/27/

66 Color Edge Detection (3) The direction of maximum rate of change of c( xy, ) is given by the angle 2g 1 xy 1 θ ( xy, ) = tan 2 gxx gyy The value of the rate of change at ( xy, ) in the direction of θ ( xy, ), is given by 1 F θ ( xy, )= 2 ( g ) ( ) xx + gyy + gxx gyy cos2 θ( xy, ) + 2gxy sin2 θ( xy, ) 1/2 2/27/

67 2/27/

68 2/27/