The simplest possible photo detector works thusly. You basically have a diode with reverse bias applied to it. The current can't flow, because there are no carriers (electrons or "holes") in the depletion region. (Look up diode for details.) However, if a photon strikes one of the electrons and excites it into the conduction band, you end up with a free electron and a free hole, which will immediately flow in opposite directions. If you have enough photons striking the depletion region, you'll have a current flowing. The current will depend primarily on the intensity of light, allowing you to measure it. I'm sure the rest is self-evident.
Oh, the idea here is that so long as the energy is sufficient to knock out an electron, a higher energy doesn't make much difference. The low threshold for cameras is usually in near-IR region (about 900nm), so all visible light produces roughly equal effect. To get color, there is a filter mosaic overlaid on the sensor.