Chapter 2: Digital Image Fundamentals - Concordia …users.encs.concordia.ca/~weiping/ELEC6641-W14/Chapter 2.pdf · Mathematical tools for DIP . Chapter 2: Digital Image Fundamentals

Chapter 2: Digital Image Fundamentals

Two types of receptors on retina: Cones & Rods Cones: sensitive to bright light; ~ photopic vision; can resolve fine details; sensitive to color; 6 to 7 million cones Rods: sensitive to dim light; ~ scotopic vision; give overall picture; not involved in color vision. 75 to 150 million rods.

Structure of the human eye


Cones: most dense in the center of retina (fovea). Rods: much more than cones, and distributed more consistently on retina. Blind spot: no cones, no rods.

Distribution of cones and rods


The thickness of lens is variable Lens becomes thicker when focused on a nearby object. Image of the object is primarily displayed in the area of the fovea.

The retinal image of an object


Subjective brightness is a logarithmic function of light intensity; HVS can adapt to a huge range of light intensity levels (order of 1010 ); But the range of simultaneous operation by HVS is small.

10

Adaptation level of HVS


The eye’s ability of identifying the bright change depends on the adaptation level; ΔI/I (Weber ratio) is used to measure the eye’s discrimination ability; A small value of ΔI/I means “good” brightness discrimination (a small percentage change is discriminable);

Discrimination of brightness change

Chapter 2: Digital Image Fundamentals Sample curve

Poor discrimination ability at low level of intensities!


Mach band effect

The brightness perceived by eye is not consistent with the actual intensity !

Chapter 2: Digital Image Fundamentals Simultaneous contrast

The brightness of the object perceived by eye depends on the background !

Chapter 2: Digital Image Fundamentals Other features of HVS -- Optical illusions

Chapter 2: Digital Image Fundamentals Visible and invisible lights


Image sensing




Image acquisition and digitization

Chapter 2: Digital Image Fundamentals Image sampling and quantization


Continuous image and the digitized one


Representation of digital image



−−−−

−−

=

)1,1()1,1()0,1(

)1,1()1,1()0,1()1,0()1,0()0,0(

),(

NMfMfMf

NfffNfff

yxf




Number of bits required to store a digital image

b=M×N ×k=N2 k k: # of bits for gray-level value


Images with different spatial resolutions

Chapter 2: Digital Image Fundamentals Images with different gray-level resolutions


Images with low gray-level resolution


How the image quality is affected by varying N and k?

Three typical images used to test the quality of images with different combinations of N and k.


Isopreference curve depends on image


Zooming and shrinking (interpolation and decimation)

Commonly used interpolation schemes: 1) Nearest neighbor repetition

2) Bilinear interpolation

3) Bicubic interpolation

dcxybyaxyxv +++=),(


Neighbors, Adjacency, Connectivity, Region, Boundary

Neighbors of a pixel 1) 4-neighbors of p, N4(p):

2) 4 diagonal neighbors of p, ND(p):

3) 8 neighbors of p, N8(p):

N8(p)=N4(p)+ND(p)


Mathematical tools for DIP


Arithmetic operations in DIP



Scaling:




Masking of ROI


Set operation


Set operation


Set operations


Intensity transformation


Neighborhood operations


Geometric/spatial transformation of image

For example,


Transformed images


Image registration


Vector product as a distance


DIP in transform domain


Transform-domain processed image


Images with different contrasts

Example 1 (problem. 2.7)

Suppose that a flat area with center ),( 00 yx is illuminated by a light source with intensity distribution

])()[( 2

02

0),( yyxxKeyxi −+−−= Assume for simplicity that the reflectance of the area is constant and equal to 1.0, and let K=255. If the resulting image is digitized with k bits of intensity resolution, and the eye can detect an abrupt change of eight shades of intensity between adjacent pixels, what value of k will cause visible false contouring?

Example 2 (problem. 2.10)

HDTV generates images with a resolution of 1125 horizontal TV lines interlaced (where every other line is painted on the tube face in each of two fields, each field being 1/60th of a second in duration). The width-to-height aspect ratio is 16:9. The fact that the horizontal lines are distinct fixes the vertical resolution of the images. A company has designed an image capture system that generates digital images from HDTV images. The resolution of each TV (horizontal line) in their system is in proportion to vertical resolution, with the proportion being the width-to-height ratio of the images. Each pixel in the colour image has 24 bits, 8 bits each for a red, a green, and a blue image. These three primary images form a colour image. How many bits would it take to store a 2-hour HDTV program.

Documents

Chapter 2: Digital Image Fundamentals - Concordia …users.encs.concordia.ca/~weiping/ELEC6641-W14/Chapter 2.pdf · Mathematical tools for DIP . Chapter 2: Digital Image Fundamentals