Fundamentals of Digital Image Processing - Control of Color Imaging Systems: Analysis and Design - page 67

1

0.9

1.5

0.8

0.7

1

0.6

0.5

0.5

0.4

0.3

0

0.2

0.5 1

0.5

0.1

0

0

-0.5

F y

-0.5

F x

-1

-1

FIGURE 2.30 Frequency response of the design filter.

J 1 4

p

" ! #

p

n 1 þn 2

n 1 þn 2

1
8

p

h(n 1 ,n 2 ) ¼h d (n 1 ,n 2 )w(n 1 ,n 2 ) ¼ p

n 1 þn 2

0 : 54 þ cos

28

n 1 , n 2 ¼ 28, 27, ... ,0,1, ... ,28

The frequency responses of the designed filter using uniform and hamming win-

dows are shown in Figure 2.30.

2.7 IMAGE RESIZING
Inmany imaging applications, it is sometimes necessary to create a low-resolution image
from a high-resolution image and vice versa for purposes of viewing, transmitting, or
printing. An example for such a situation is when an image is initially scanned at a high
resolution (for example: 3072 2048 pixels). This image is suitable for high quality

printing. However, for the purpose of viewing this image on a high de

nition television
(HDTV) monitor, a lower resolution image is required. Similarly, an even lower
resolution image is needed for national television system committe (NTSC) television
viewing. Image resizing is basically performed by the process of sampling rate conver-
sion. Sampling rate conversion decouples the spatial resolution of the image source from
the spatial resolution requirement of the display or printing device. Thus, the same image
source can be viewed on different display systems with different resolutions [18,19].

2.7.1 D EFINITION OF S AMPLING R ATE C ONVERSION
Given an image sampled at a rate R 1 , create a new image at a spatial rate R 2 , where
the two rates are related by a factor of

a , that is,

R 2 ¼ a R 1 ( 2 : 103 )