Signal Processing Stack Exchange is a question and answer site for practitioners of the art and science of signal, image and video processing. It only takes a minute to sign up.
Sign up to join this community
This is usually a two-part, trichroic prism that divides the light into separate red, green, and blue beams, each of which is detected by a separate CCD or CMOS image sensor. A circuit (3) mathematically synchronizes and combines the outputs from the red, green, and blue image sensors to make a single video signal based on components called luminance and chrominance (loosely, the brightness and color of each part of the image).
I feel this as tearing paper and sticking it back with a gum.
This is the process of using dichroic prisms for color separation in high end camcorders and digital cameras. The other approach used is with color filters directly on a single CCD photosensor in a grid pattern used to create a Bayer filter. The prisms can offer better light separation and less loss than the surface filters, and for a given CCD sensor, all the pixels in each sensor can be used for each color, resulting in higher resolution. The Bayer filter approach also creates mosaic artifacts that can be avoided with prism color separation and separated CCD sensors.
This link further details the differences and demonstrates the advantage of a 3CCD system over a Bayer system:
It is done to send colour TV transmissions in a way that is compatible with grayscale TV receiver sets, and to enable the colour TV receiver sets compatible with grayscale TV transmissions.
Before colour TV was invented, TV transmissions were not in colour, but in "grayscale", where the image is just sent as voltage level how brightly each "pixel" is lit (except there was no pixels as it was analogue, but you likely get the idea).
So TV camera had only one sensors and it only detected image brightness.
When colour information is needed for colour transmission, you need basically a TV camera with three sensors, one for each colour. The light is split up into mirrors into three parts which can have a distinct colour filter, so you can have three brightness sensors, one for red light, one for green light, and one for blue light.
However, to make the colour TV transmissions compatible with the greyscale TVs, it was necessary to send a compatible greyscale signal. The voltages are summed up (with appropriate scale factors) to end up with a brightness signal identical to what a greyscale TV camera would send.
This is the luminance part of the signal, called Y.
But as colour transmission needs three signals, and the brightness is already calculated from the colour signals, two other signals (called U and V) are calculated in a similar way to end up with colour difference signals which are then combined with a colour carrier (basicaly using quadature amplitude modulation) which is the chroma signal. The chroma signal is added to the luminance, in order to transmit the colour composite video signal.
This signal can be received by greyscale and colour TVs. Greyscale TVs can't decode the colour and will show the luminance signal Y only. Colour TVs can detect the colour signal, decode it to U and V signals, and then using the Y, U and V signals, they can be calculated back into red, green and blue signals, as the TV CRT tube has three electron guns, red, green and blue.
