PPT s03-machine vision-s2
- 1. Course: Machine Vision Light and Color Session 03 D5627 – I Gede Putra Kusuma Negara, B.Eng., PhD
- 2. Outline • The Physics of Color • Human Color Perception • Representing Color • A Model of Image Color
- 3. The Physics of Color
- 4. What is color? • Color is a psychological property of our visual experiences when we look at objects and lights, not a physical property of those objects or lights (S. Palmer, Vision Science: Photons to Phenomenology) • Color is the result of interaction between physical light in the environment and our visual system
- 5. Electromagnetic spectrum • Why do we see light at these (visible) wavelengths? Because that’s where the sun radiates electromagnetic energy
- 6. The Physics of Light • Any source of light can be completely described physically by its spectrum: the amount of energy emitted (per time unit) at each wavelength 400 - 700 nm
- 7. Interaction of light and surfaces • The color of surfaces is a result of a large variety of mechanisms, including differential absorption at different wavelengths, refraction, diffraction, and bulk scattering
- 8. Interaction of light and surfaces
- 10. The Human Eye • Diameter: 20 mm • 3 membranes enclose the eye • Cornea & sclera • Choroid • Lens • Retina • The human eye is a camera: photoreceptor cells (rods and cones) in the retina have the same function as film/CCD/CMOS sensor in a camera
- 11. Light Receptors • Cones – Cones are located in the fovea and are sensitive to color – Each one is connected to its own nerve end – Cone vision is called photopic (or bright-light vision) • Rods – Rods are giving a general, overall picture of the field of view and are not involved in color vision – Several rods are connected to a single nerve and are sensitive to low levels of illumination (scotopic or dim-light vision)
- 12. Human Color Vision • Unlike rods, which contain a single photopigment, there are three types of cones that differ in the photopigment they contain • Each of these photopigments has a different sensitivity to light of different wavelengths • For this reason are referred to as “blue,” “green,” and “red,” or, more appropriately, short (S), medium (M), and long (L) wavelength cones
- 13. Color Perception • Rods and cones act as filters on the spectrum • To get the output of a filter, multiply its response curve by the spectrum, integrate over all wavelengths • Each cone yields one number • The sensitivity of the human eye to luminous intensities related to the three primary colors is not the same
- 14. Color Sensing in Camera (RGB) Bayer filter Inside a digital SLR
- 15. Practical Color Sensing: Bayer Grid
- 17. Standardizing color experience • Accurate color reproduction is commercially valuable • Many products are identified by color (e.g. “golden” arches) • Only few color names are widely recognized by English speakers: – About 10; other languages have fewer/more, but not many more. – It’s common to disagree on appropriate color names. • Color reproduction problems increased by prevalence of digital imaging
- 18. Color Matching Experiment • Most studies of the three-color nature of human vision are based on some variation of this simple apparatus • One part of the screen is illuminated by a lamp of a target color; the other by a mixture of three colored lamps. • The test subject (observer) adjusts the intensities of the three lamps until the mixture appears to match the target color.
- 19. Trichromacy • Three numbers seem to be sufficient for encoding color • In color matching experiments, most people can match any given light with three primaries • Exception: color blindness • For the same light and same primaries, most people select the same weights • Trichromatic color theory dates back to 18th century (Thomas Young)
- 20. Additive vs. Subtractive Color Matching Additive • Many colors can be represented as a mixture of A, B, C: M=a A + b B + c C • Where the = sign should be read as “matches” • Gives a color description system: two people who agree on A, B, C need only supply (a, b, c) to describe a color Subtractive • Some colors can’t be matched like this M=a A + b B + c C, instead, must write: M+a A = b B+c C • Interpret this as (-a, b, c) • Problem for building monitors: Choose R, G, B such that positive linear combinations match a large set of colors
- 21. Grassman’s Laws • Additive matching is linear • If two test lights can be matched with the same set of weights, then they match each other: – Suppose A = u1 P1 + u2 P2 + u3 P3 and B = u1 P1 + u2 P2 + u3 P3. Then A = B. • If we mix two test lights, then mixing the matches will match the result: – Suppose A = u1 P1 + u2 P2 + u3 P3 and B = v1 P1 + v2 P2 + v3 P3. Then A+B = (u1+v1) P1 + (u2+v2) P2 + (u3+v3) P3. • If we scale the test light, then the matches get scaled by the same amount: – Suppose A = u1 P1 + u2 P2 + u3 P3. Then kA = (ku1) P1 + (ku2) P2 + (ku3) P3.
- 22. Color Matching Functions • Choose primaries, say A, B, C • Given energy function, E(λ) what amounts of primaries will match it? • For each wavelength, determine how much of A, of B, and of C is needed to match light of that wavelength alone a(λ) b(λ) c(λ) • These are color matching functions • Then the match is a(l)E(l)dl ò { }A+ b(l)E(l)dl ò { }B+ c(l)E(l)dl ò { }C
- 23. Linear Color Spaces: RGB • Linear color space that uses single wavelength primaries (645.16 nm for R, 526.32 nm for G, and 444.44 nm • Available colors are usually represented as a unit cube— usually called the RGB cube— whose edges represent the R, G, and B weights • Subtractive matching required for some wavelengths
- 24. Linear color spaces: CIE XYZ • Established in 1931 by the International Commission on Illumination • Primaries are imaginary, but matching functions are everywhere positive • 2D visualization: draw (x,y), where x = X/(X+Y+Z) y = Y/(X+Y+Z)
- 25. Nonlinear color spaces: Uniform color spaces • Unfortunately, differences in x,y coordinates do not reflect perceptual color differences • CIE u’v’ is a projective transform of x,y to make the ellipses more uniform u' = 4x -2x +12y+3 v' = 9y -2x +12y+3
- 26. Nonlinear color spaces: HSV • First described by Alvy Ray Smith in 1978, HSV seeks to depict relationships between colors, and improve upon the RGB color model • Perceptually meaningful dimensions: Hue, Saturation, Value (Intensity)
- 27. A Model of Image Color
- 28. The Use of Color in Computer Vision • Color histogram for indexing and retrieval Swain and Ballard, Color Indexing, IJCV 1991 • Skin detection M. Jones and J. Rehg, Statistical Color Models with Application to Skin Detection, IJCV 2002 • Image segmentation and retrieval C. Carson, S. Belongie, H. Greenspan, and Ji. Malik, Blobworld: Image segmentation using Expectation-Maximization and its application to image querying, ICVIS 1999 • Robot soccer M. Sridharan and P. Stone, Towards Eliminating Manual Color Calibration at RoboCup. RoboCup-2005: Robot Soccer World Cup IX, Springer Verlag, 2006
- 29. Acknowledgment Some of slides in this PowerPoint presentation are adaptation from various slides, many thanks to: 1. D.A. Forsyth, Department of Computer Science, University of Illinois at Urbana – Champaign (http://luthuli.cs.uiuc.edu/~daf/) 2. Professor William Hoff, Department of Electrical Engineering & Computer Science (http://inside.mines.edu/~whoff/) 3. James Hays, Computer Science Department, Brown University, (http://cs.brown.edu/~hays/)
- 30. Thank You