What Exactly is Digital Video? 33
WHAT ABOUT GRAY?
Images made by a black-and-white video camera contain more
than simple black or white dots. Black-and-white video images are
comprised of black, white, and many shades of gray—as seen in
Figure 2-10.
In order to convert shades of gray into numerical values for
digitizing an image, we must defi ne the shades of gray between
black and white. Earlier in chapter one, Figure 1-6 portrayed a
10-step grayscale chart and the video waveform that results from
one scan line of the chart by a video camera. It should be clear that
Figure 2-10 Shades of Gray
34 Digital CCTV
10 shades of gray are not enough to reproduce the many shades
of gray in an image like the one in Figure 2-10. How many shades
are enough? Figure 2-11 illustrates the answer to this question.
64 steps of gray look almost good enough to represent con-
tinuous shading from white to black. 128 steps looks like what we
need to produce continuous shading. Any more steps would
probably be wasted. 256 steps are far more than what is needed
to represent all of the shades of gray that the human eye can
perceive.
The important piece of information about the number of
shades necessary to make acceptable continuous shading is the
number of bits needed to defi ne the number of shades of gray.
Table 2-6 illustrates the number of bits (and bytes) needed for 64
shade, 128 shade, and 256 shade grayscale choices.
After the digital images reach their fi nal destination, the
digital signal often has to be reverted back to an analog signal.
This is achieved with a digital-to-analog converter (DAC).
Basically, digital-to-analog conversion is the opposite of analog-
to-digital conversion.
Figure 2-11 Shades of Gray in Steps
Table 2-6 Number of Bits Needed To Make Each Picture
Fig. 3-2 Fig. 3-3 Fig. 3-4 Fig. 3-1
6 dpi 12 dpi 24 dpi 300 dpi
Dots per row 16 32 64 800
Number of rows 16 32 64 800
Total dots 256 1024 4096 639,997
Number of bits 256 1024 4096 639,997
What Exactly is Digital Video? 35
When you digitize video, you decide:
1. Frame rate—how many frames per second to capture.
2. Frame size—what size the frames should be.
Even though the NTSC standard dictates 29.97 frames per second
to achieve an acceptable video image, you can actually slow that
rate down to 15 or 16 frames per second without diminishing the
illusion of movement. The level of video resolution is also directly
related to the size of the screen and the viewing distance.
Although it is possible to capture video at a size that will fi ll
your screen—typically 640 pixels by 480 pixels—it is advisable to
specify a smaller frame size in your video capture software. For
example, on a monitor displaying 640 × 480 screen resolution a
frame size of 160 pixels by 120 pixels would fi ll 1/16 of the screen.
Frame sizes always maintain an aspect ratio of 4 : 3 to refl ect the
resolution of computer monitors and televisions.
A video image the size of an average 640 × 480 frame size
with a resolution of 24 bits per pixel (thousands of colors) and a
standard NTSC frame rate of 30 frames per second represents a
little over 26 MB of data per second of video, not counting audio.
This means a 1 GB hard disk could only hold about 38 seconds of
video. By reducing frame size, frames per second, and bits per
pixel you can make the video more manageable (i.e., make changes
that would reduce the amount of information to store or transmit)
at the expense of the image quality.
To calculate storage needs for digital video from black-and-
white cameras, multiply the horizontal camera resolution by the
Table 2-7 Continuous Shading
Pixels per row 640
Number of rows 480
Total pixels 307,200
# of Bits/Pixel Total Bits Total Bytes
64 shades—6 bits 1,843,200 230,400
128 shades—7 bits 2,150,400 268,800
256 shades—8 bits 2,457,600 307,200

Get Digital CCTV now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.