Digital cameras operate in much the same way as conventional film cameras, except that light-sensitive film is replaced by a reusable sensor. This sensor, which may for example be the product of charge-coupled device (CCD), complementary metal-oxide semiconductor (CMOS) or Foveon technologies, captures data from millions of tiny picture elements (pixels) and sends it to a processor and a memory. The various sensor designs work in different ways to achieve a similar result but essentially capture only brightness information. When using a CCD, for example, the image is passed through red, green and blue filters to obtain colour data. Other technologies work in a somewhat different manner.

The various technologies count the total number of pixels in different ways, making direct comparisons difficult. Another consideration is the difference between the numbers quoted for actual and effective pixels. Some pixels around the edge of a sensor may not be used to record image data, so the effective pixel count is normally the most important. When the number of effective pixels exceeds the number of actual pixels, perhaps by a ratio of two-to-one, interpolation is being used. This means that the camera's processor is using a mathematical algorithm to insert an additional pixel between each measured value. This might double the pixel count in a reasonably successful manner, but can never equal the performance of systems systems that capture the equivalent number of pixels as raw data.