1. The number of color bits per pixel. Images captured from digital cameras or scanners are recorded as pixels (picture elements), which are tiny square blocks of light that are merged by the human eye to form the image. Each pixel contains information about the colors recorded. The bit depth refers to the number of bits within each pixel and the corresponding number of colors within those bits. When more data can be recorded or captured, the bit depth is higher, which results in a greater range of color. If, for example, 24 bits of color are recorded, each image pixel will have 8 bits of data for each of the red, green and blue color channels. The higher the bits, the better the color reproduction, and consequently, the larger the overall file size.
The minimum bit depth is two colors, which is a 1-bit color depth. This means that every dot or pixel that makes up the image is represented by 1-bit of data. The color palette for the 1-bit image determines which two colors are displayed, out of the millions of possibilities. Bit depth increases exponentially. A 2-bit color depth contains four colors (2 x 2), a 3-bit color depth contains 8 colors (2 x 2 x 2), an 8-bit color depth contains 256 colors (2 to the 8th power), and a 24-bit color depth contains 16,777,216 colors (2 to the 24th power). Another way to express 24-bit color is that each of the three additive primary color channels used to reproduce color on a computer monitor (red, green, and blue), can each display up to 256 different intensity values, resulting in 16,777,216 colors (256 x 256 x 256). All of these colors give the display a photographic quality.
2. Bit depth also refers to the ability of a scanner or digital camera sensor to record the spectrum of colors in the image.
(The image below has a bit depth of 8, resulting in 256 colors (2 to the 8th power) that are contained in the image. The image at the bottom shows the palette of 256 colors that were used.)