Digital Image Processing, Gonzalez & Woods 1 Digital Image Processing STMIK MDP.

Digital Image Processing, Gonzalez & Woods

1

Digital Image Processing

STMIK MDP

Digital Image Processing, Gonzalez & Woods 2


Image Distortion Model

• Restoration depends on distortion Common model: convolve plus noise Special case: noise alone (no convolution)


Thoughts on Restoration

• Enhancement vs restoration Enhancement: cosmetic Restoration: substantive In reality, there’s a continuum

• Restoration relies on prior knowledge• Example of restoration

Inverse filtering Noise removal Nonlinear processing


Chapter 6, Color Image Processing

• Color fundamentals and models• Pseudocolor

Slicing False-color maps Index color Multispectral color models

• Color transformations• Smoothing and sharpening• Color segmentation


Retinal Physiology and Color

• Human retinas have (at least) four types of photoreceptors Three types of ‘cones’

High light level, high acuity vision Each type of cone has a different spectral response

One type of ‘rods’ Low-light level and peripheral vision

• There is substantive genetic diversity in color receptors Different spectral response of photoreceptor Absence of one of the pigments Many more phenomena...


Spectral Response of Cones


Color Matching Theory

• Young’s observation Any uniform color can be matched by projecting three

different light sources onto a screen• In the 1920’s, an international effort on the part

of a number of physicists let to the CIE standard color theory

• The effort attempted to simplify and standardize• There have been many refinements, but the basic

theory still stands• Theory useful for matching pigments (all paint

stores have spectrophotometers) and design of color media systems


Tristimulus Values

QuickTime™ and aTIFF (Uncompressed) decompressor

are needed to see this picture.

I() — spectal energy distribution of a source

X,Y ,Z — tristimulus values of the source

X x ()I()d400

700

Y y()I()d

400

700

Z z ()I()d

400

700

Chromaticities - trichromatic coefficients

x X

X Y Z

y Y

X Y Z

z Z

X Y Zx y z 1


Chromaticity Diagram

• From http://www.efg2.com/Lab/Graphics/Colors/Chromaticity.htm

• White is (1/3, 1/3, 1/3)• Y is the subjective

intensity.

http://www.efg2.com/Lab/Graphics/Colors/Chromaticity.htm


Standard Observer


Better Subjective Color

• Constant distances in x, y space do not correspond to ‘constant’ changes of color

• A better linear space is provided by the Y-U-V coordinates


RGB Coordinates

• With the advent of color television, the RGB coordinates were introduced to conveniently produce color on a color cinescope (cathode ray tube)


RGB Gamut

• Color space produced by a specific set of phosphors or LCD filters is limited


Constant-Color-Difference Coordinates

• For modulation (analog TV) and encoding (digital TV, JPEG), Y-U-V-like coordinates are produced

• R’, G’, B’ are gamma corrected RGB. These are used in electronic media.


Chapter 6Color Image Processing












































































































































Digital Image Processing, Gonzalez & Woods 1 Digital Image Processing STMIK MDP.

Documents