0

As we know that RGB monitors requires separates signals for red, green, and blue components of the image but television monitors uses single composite signals. For this composite signal use YIQ color model.

My question is why we can say YIQ is composite signal and RGB is 3 different signal?

N. B. - I am computer science student, I want to understand just intuition rather than details because much using technical term it may creates unnecessary confusion. So please explain easily.

3
  • 3
    I'm afraid you are just confusing two things - the information content of the signal with its physical transmission characteristics.
    – Brian H
    Nov 29 '21 at 20:40
  • 3
    I am afraid the poor grammar makes it unclear what you are actually asking about. Nov 29 '21 at 20:48
  • 2
    Regardless, it's not retroocmputing. Nov 29 '21 at 23:17
6

Not so sure what that question is about, as global terms like this can have different meanings, depending on context. Also it seems as if it mixes physical definitions of signal transmission with colour models. Most basic:

  • About Signals:

    • Separate signals are independent signals transferred over independent lines.
    • Composite is one signal with multiple components transferred over a single line
  • About Colour Models:

    • RGB is a colour model describing colours to be produced via three levels of Red, Green and Blue, thus each colour as its own by its luminosity
    • YIQ (or YUV) is a colour model describing colours to be produced with three components combined in two separate values as colour coordinates (I/Q) and an over all luminosity (Y) (*1)

Colour models do not tell anything about composite or not. For example, NTSC is YIQ over a single signal, while S-Video is the same, but using two lines, one for lumina (Y) and one for colour (IQ). Although some models are more easy to be turned into composite signals then others - in case of YIQ, it's already made as relative encoding, so combination is way less effort than would be needed for a composite RGB signal.


*1 - Making luminosity a dedicated component is what made it colour TV compatible with B&W TV, as now the basic B&W signal could be encoded like before, while adding the new colour component in a way old devices would not detect.

16
  • I am asking reason of these 2 points..and why we use YIQ model inspite of already have an RGB color model?
    – Alok Maity
    Nov 29 '21 at 20:53
  • @Stack What reasons? one is three signals, independent of each other, the other is one signal
    – Raffzahn
    Nov 29 '21 at 20:54
  • why we use YIQ model inspite of already have an RGB color model in TV?
    – Alok Maity
    Nov 29 '21 at 20:56
  • @Stack Because There each TV channel is ... well .. a single channel. Thus all signals for a colour picture need to be 'composed' into a single signal that can be send thru this single canal. Think of it like having only one wire to be plugged int the TV - vs. RGB needs three.
    – Raffzahn
    Nov 29 '21 at 20:59
  • @Stack composite video does not use the RGB colour model.
    – Justme
    Nov 29 '21 at 21:00
7

YIQ and related luminance (Y) plus chrominance (some other two channels) relate to the history of broadcast television — originally monochrome, with colour added in a backwards-compatible fashion — but that stems from the nature of human vision.

Eyes have a high density of rods, which are comparatively quite sensitive and detect luminance only, and a much lower and less even density of cones, which act to detect frequency and therefore colour.

As a result humans see a high resolution of brightness and overlay that with a low resolution of colour. Which means that usually video is encoded with a high resolution of luminance and a low resolution of chrominance — usually half or a quarter as mich information.

Video engineers and televisions therefore think most natively in terms of 1d luminance plus 2d chrominance, and that is partly expressed in the connectivity standards, including original colour composite, S-Video and modern component.

So: the reason is partly continuous evolution of the standards, partly keeping things in terms of what most closely correlates with perception.

2
  • While there are fare more rods than cones, multiple rods connect to a single neuron which averages their inputs, while each cone has its own neuron, so the resolution difference is at least not as great as the difference in number of cells suggests. Also it is not clear to me whether rod cells have any major role in photopic (daytime) vision at all. I assume the lower resolution of color perception compared to brightness reception has more to do with brain structures than with the structure of the eye.
    – TeaRex
    Nov 30 '21 at 9:06
  • @TeaRex Also cones are concentrated at the center of the eye where vision is sharper, while rods are denser in the periphery (which is why you can see an object in the dark better by not looking directly at it). But the different 'color' cones are distributed similarly to the triad dots on a CRT, so each color has lower resolution than the combination of all colors. There are also horizontal cells between rods and cones that enhance edge sharpness but don't discriminate between colors. Dec 3 '21 at 4:24
5

As television signal was originally a single brightness channel, there was a need to add colour in a way that is compatible with the older standards. This way it is possible to watch colour transmissions on a B&W TV, and B&W transmissions on a colour TV.

Three components are required for transmitting colour, but the brigthness signal is needed for compatibility. So RGB is translated to three other signals, one that is equal to the B&W brightness Y, and two other signals that are called the colour difference signals, U and V.

In NTSC colour transmission, the two U and V are slightly processed (scaled and rotated by 33 degrees) to end up with two similar signals called I and Q.

Basically U&V, or I&Q, is a sine (or cosine) wave at the color carrier frequency (3.579545 MHz) which encodes colour in the amplitude of the sine wave (indicating saturation, how much colour there is) and in the phase of the sine wave (indicating hue, which colour there is). The phase reference, the colourburst, is sent on each line before video.

So in short, two colour signals are quadrature amplitude modulated at the colour carrier frequency and added to the brightness signal with reference to end up with the composite signal.

Thus the single composite signal does not contain the three YIQ/YUV signals directly, but in encoded form. So composite must be decoded into the three YIQ/YUV signals before converting back to RGB for display.

15
  • what is the meaning of B&W ?
    – Alok Maity
    Nov 29 '21 at 21:03
  • @Stack black&white, monochrome TV, before TV had colour.
    – Justme
    Nov 29 '21 at 21:10
  • frequency is proportional to intensity?
    – Alok Maity
    Nov 29 '21 at 21:12
  • 1
    @Stack yes and no. Frequency is an encoding detail when it's on the air. again a different part. TV is very simple. Basically. But this simplicity is pat of very elaborate schemes with many different aspects depending on the point one looks at.
    – Raffzahn
    Nov 29 '21 at 21:18
  • 1
    @Stack Not sure how to say this friendly enough. You're touching an enormous wide with lots of analogue technology. This covers many layers of interconnected technology each filling books. Maybe start by reading wiki articles about how radio transmission works, how a TV picture is structured, what a colour space is, how different encodings work, how quadrature modulation works, how multiple signals are mixed into a channel and how channels are working in general. as all of this (and much more) goes into defining why certain decisions about how to create signals, how to compose them were made.
    – Raffzahn
    Nov 29 '21 at 22:26
3

A lot of detail is in the other answers. I try to focus on the basic concept. For a color image to be displayed on a color CRT, you need three components. They can be red, green, blue (RGB) or a general brightness (Y) combined with information about the cyanness (or orangeness, which is just negative cyanness) an well as magentaness (or greenness, which is just negative magentaness).

Now we need to talk about how the three components get transmitted to the monitor. On the one hand, look at cable radio: You have one cable, but by tuning your receiver, you can receive different programs from the same cable. On the other hand, look at an analog PBX telephony system: If you have three phones in one office, you have three (pairs of) wires running from the PBX to the office, each one only carrying one call. The composite signal is like cable radio (single cable, three programs), whereas RGB is like the phone lines (three cables, three programs).

There is nothing inherent to the YIQ (or the related YUV) color models preventing them to be tranmitted on three different cables. Essentially, that's exactly what "component video" used for analog HD transmission is doing.

6
  • what does mean of component video?
    – Alok Maity
    Nov 29 '21 at 21:48
  • en.wikipedia.org/wiki/Component_video Three cables. One for Y (brightness), two for color information (Pb, Pr in this case). The way of representing color is very similar to YIQ, but the method of transmission (three cables) is similar to classisc analog RGB transmission, as in VGA cables. Nov 29 '21 at 21:55
  • What does it mean of this statement!!"There is nothing inherent to the YIQ (or the related YUV) color models preventing them to be tranmitted on three different cables. "--- you mean if I use three cables of RGB which will work same when I use single cable of YIQ?
    – Alok Maity
    Nov 29 '21 at 22:06
  • composite video transmits Y, I and Q combined through one cable. component video transmits (something like) Y, I and Q over three cables. Both is possible. Nov 29 '21 at 22:12
  • but when use 3 cable it is similar to RGB which isn't feasible?
    – Alok Maity
    Nov 29 '21 at 22:15

Not the answer you're looking for? Browse other questions tagged or ask your own question.