2/3/2009. Color. Today! Sensing Color Coding color systems Models of Reflectance Applications

Transcription

1 Color Today! Sensing Color Coding color systems Models of Reflectance Applications 1

2 Color Complexity Many theories, measurement techniques, and standards for colors, yet no one theory of human color perception is universally accepted Color of object depends on object itself but also on light source illuminating it, on color of surrounding area, and on human visual system (the eye/brain mechanism) Some objects reflect light (wall, desk, paper), while others also transmit light (cellophane, glass) Achromatic & Chromatic Light Achromatic light: intensity (quantity of light) Chromatic light: visual color sensations 2

3 The Elements of Colour >80% incident light from white source reflected from white object. <3% from black object. Narrow bandwidth reflected perceived as colour The Visible Spectrum 3

4 Gamma Gamma (γ) is a measure of the nonlinearities of a display Term often used incorrectly to refer to nonlinearity of image data We need to maintain color consistency across different platforms and hardware devices (monitor, printer, etc.) Even the same type/brand of monitors change gamma value over time Example: PC monitors have a gamma of roughly 2.5, while Mac monitors have a gamma of 1.8, so Mac images appear dark on PC s Nonlinearities To achieve equal steps in brightness, space logarithmically rather than linearly, so that I I j + 1 j = I I j j 1 = r Use the following relations: 2 3 I = I0, I1 = ri0, I2 = ri1 = r I0, I3 = ri2 = r I0, 0 K, I = r I = 1 4

5 Chromatic Color Hue distinguishes among colors such as red, green, purple, and yellow Saturation refers to how pure the color is, how much white/gray is mixed with it Lightness: perceived achromatic intensity of reflecting object Brightness: perceived intensity of a self luminous object, such as a light bulb, the sun Color Mixture subtractive mixture additive mixture 5

6 Imaging Process Factors that Affect Perception Light: the spectrum of energy that illuminates the object surface Reflectance: ratio of reflected light to incoming light Specularity: highly specular (shiny) vs. matte surface Distance: distance to the light source Angle: angle between surface normal and light source Sensitivity: how sensitive is the sensor 6

7 Sensing Color light beam splitter 3 CCD Coding color systems RGB is an additive system (add colors to black) used for displays CMY is a subtractive system for printing HSV is good a good perceptual space for art, psychology, and recognition YIQ used for TV is good for compression 7

8 Comparing color codes RGB color cube 8

9 Color palette and normalized RGB Color hexagon for HSI (HSV) Color is coded relative to the diagonal of the color cube Hue is encoded as an angle Saturation is the relative distance from the diagonal Intensity is height 9

10 Editing saturation of colors (Left) Image of food originating from a digital camera; (center) saturation value of each pixel decreased 20%; (right) saturation value of each pixel increased 40%. Properties of HSI (HSV) Separates out intensity I from the coding p y g Two values (H & S) encode chromaticity Convenient for designing colors Hue H is defined by an angle Saturation S models the purity of the color S=1 for a completely pure or saturated S=0 for a shade of gray 10

11 YIQ and YUV for TV signals Have better compression properties Luminance Y encoded using more bits than chrominance values I and Q; humans more sensitive to Y than I,Q NTSC TV uses luminance Y; chrominance values I and Q Luminance used by black/white TVs All 3 values used by color TVs YUV encoding used in some digital video and JPEG and MPEG compression Conversion from RGB to YIQ We often use this for color to gray-tone conversion. 11

12 Colors can be used for image segmentation Can cluster on color values and pixel locations Can use connected components and an approximate color criteria to find regions Can train an algorithm to look for certain colored regions for example, skin color Color Clustering by K means Algorithm Form K-means clusters from a set of n-dimensional vectors 1. Set i c (iteration count) to 1 2. Choose randomly a set of K means m 1 (1),, m K (1). 3. For each vector x i, compute D(x i,m k (i c )), k=1, K and assign x i to the cluster C j with nearest mean. j 4. Increment i c by 1, update the means to get m1(i c ),,m K (i c ). 5. Repeat steps 3 and 4 until C k (i c ) = C k (i c +1) for all k. 12

13 K means Clustering Example Original RGB Image Color Clusters by K-Means Extracting white regions Program learns white from training set of sample pixels. Aggregate similar neighbors to form regions. Components might be classified as characters. input RGB image output is a labeled image. 13

14 Skin color in RGB space Purple region shows skin color samples from several people. Blue and yellow regions show skin in shadow or behind a beard. Finding a face in video frame (left) input video frame ( ) p (center) pixels classified according to RGB space (right) largest connected component with aspect similar to a face (all work contributed by Vera Bakic) 14

15 Color histograms Histogram is fast and easy to compute. Size can easily be normalized so that different image histograms can be compared. Can match color histograms for database query or classification. Color histograms 15

16 How to make a color histogram Make 3 histograms and concatenate them Create a single pseudo color between bt 0 and 255 by using 3 bits of R, 3 bits of G and 2 bits of B Can normalize histogram to hold frequencies so that bins total