PIXCI® accepts S-Video input. This feature is a major benefit to applications requiring high quality luminance (intensity or grey level) and chrominance (color) registration. To understand the advantage of S-Video, it is useful to begin with a discussion of the signal characteristics of standard NTSC or PAL broadcast signals. The advantages of S-Video should be self-evident after NTSC & PAL are understood.
NTSC & PAL Signal Description
NTSC & PAL video standards were engineered to allow transmission of color television images in a format compatible with already existing RS-170 & CCIR monochrome. NTSC & PAL color is transmitted via a composite wave form with limited bandwidth. The luminance information is transmitted via the amplitude of the carrier wave while the chrominance information is transmitted via a frequency modulated color signal superimposed on the amplitude.
The simultaneous transmission of monochrome and color is the factor that constrains the quality of NTSC & PAL video. A receiver, a television set, VCR, or PIXCI, separates the luminance and chrominance signals by running the composite signal through filters that strip the color signal from the carrier wave. The process of decoding composite video with filters tends to smooth the transitions in the wave form. This smoothing process is exhibited as edges that aren't as sharp as originally recorded, and colors that aren't as distinct.
The purpose of broadcast television is to display reasonable approximations of reality for general viewing by the human eye, and small levels of distortion aren't significant. In contrast, many machine vision applications require edge detection to sub-pixel accuracy, and distortion isn't acceptable. Differences in shading can be critical to applications requiring precise color information, often required in the printing, medical, or cosmetics industries.
S-Video Eliminates Filtering
S-Video is a higher quality method for delivering luminance and chrominance signals. The quality is improved by eliminating the need for the filtering processes that degrade video output. The method is simple - don't combine the signals in the first place. If the signals are always separate, never combined, then no filtration is required. The resulting image, captured in the computer's memory, is a more precise representation of the image originally captured by the camera or recorded by the VCR.
There is only one significant disadvantage to S-Video - cost. S-Video equipment is generally more expensive than composite video equipment. Another disadvantage is that S-Video is not a broadcast standard. Video can only be moved through interconnecting cables.
Camera/Imaging Device Recommendations
For accurate, low cost edge detection applications, an RS-170/CCIR camera is recommended. A simple monochrome camera doesn't impose the complications of color. Edges are digitized with precision because filtering is not required.
For applications requiring color, and where sub-pixel accuracy is not needed, an NTSC/PAL color camera is fine.
For color applications requiring better registration of luminance
and chrominance information, the use of an S-Video camera is suggested.
EPIX Vision - March 1996 Newsletter