Of course, today's telescopes no longer use photographic plates. Instead, a charge-coupled device, or CCD for short, is used. These enable direct connections between an object's incoming photons and its image on a computer. Here's how they work.
CCDs are based on a principle called the photoelectric effect. If a photon with sufficient energy hits an electron in the outer shell of an atom, The transfer of energy to the electron can be enough to free it from the atom altogether. This is a fundamental component of quantum mechanics, first analyzed by Albert Einstein in 1905. CCDs use a thin wafer of silicon to produce electrons from photons, because silicon easily releases electrons with visible light. A tiny positively charged capacitor is attached to the silicon wafer in order to collect the freed electrons. If we get one electron for each photon in the range, we'd have 100% quantum efficiency.
The highest quality CCDs can achieve up to 90% quantum efficiency. It's interesting to note that the quantum efficiency of the human eyes, rods and cones is only 1%. The photons start producing electrons as soon as the shutter is opened. The capacitor collects the freed electrons until the shutter is closed. At that point, the voltage across the capacitor represents the number of electrons the capacitor collected.
This information is sent to the computer. All of this is miniaturized into an integrated circuit and represents one pixel. CCDs are made of thousands or even millions of these configured as an array.
The CCD on Hubble's Wide Field Camera 2 has 2K by 4K arrays for an 8 megapixel CCD. And as before, for color, we simply repeat observations with filters. For example, here's Hubble's photograph of the planetary nebula.
MYCN18, 8,000 light years away. This picture has been composed from three separate images taken with a blue filter to identify light from oxygen, a green filter to identify light from hydrogen, and a red filter to identify light from nitrogen.