[ Team LiB ] Previous Section Next Section

Imaging Sensors

The heart of any digital camera is its image sensor, which actually captures the image, registering it in a form that can later be used. It detects photons and translates their energy into a minute electrical current that computer circuits can sense, amplify, digitize, and store.

Many people call image sensors CCDs, the name of the technology used by most digital camera image-sensing devices (charge coupled devices). Although CCDs are used in most cameras, some companies have adopted a similar technology that's usually termed CMOS. Yes, it's the same name as the memory used for system-configuration data in your computer, because the image sensors and the memory use the same semiconductor technology—Complimentary Metal-Oxide Semiconductor technology. Because of the use of garden-variety circuit technology, CMOS sensors tend to be less expensive, although they typically are not as sensitive as CCDs.

Although the term CCD usually is used in the singular, the actual image sensor of a digital camera requires an array of CCDs. The mere presence of light is not as important as its pattern, which makes up the actual image. A single CCD (or CMOS sensor) element registers only a single point. Consequently, image sensors are an array of individual CCD elements. In video cameras, the CCD elements get arranged as a matrix. Camera circuitry samples each element in turn to scan an image frame.


Image sensors come in various sizes. Typically they measure one-quarter, one-third, or one-half inch (diagonally). All else equal, the larger the CCD, the greater the number of elements that can be packed into the array. The number of elements determines the resolution of the image signal produced by the CCD. This number is related to, but is not the same as, the number of pixels in an image.

The size of the sensor affects other elements of the camera design. For example, the coverage of the lens usually is tailored to match the sensor size. (That's one reason why camera-makers cannot simply put CCDs inside ordinary 35 mm cameras—today's practical sensors are substantially smaller than 35 mm film and would only register a fraction of the image from the lens.) Because most image sensors are smaller than the film used in ordinary cameras, interchangeable lenses often act "longer" on digital cameras. That is, a 50 mm lens from a standard film-based SLR might give the view of a 65 mm lens when twisted onto a digital camera. The sensor sees only the central part of the image from the lens. With many of today's high-end digital cameras with interchangeable lenses, 35 mm lenses act as if their focal length were 1.3 times longer than on a film camera.


The maximum possible resolution is set by the number of pixels that the CCD can sense. The keyword in digital photography currently is megapixel—that is, a million pixels. At one time photographers used the million-pixel mark as distinguishing good cameras from bad. Today the best consumer-model digital cameras have about five megapixels of resolution, which results in images comparable to 35 mm film.

The number of total pixels is directly related to the resolution of the image sensor. If you know the number of elements in an image sensor (video cameras are sometimes described by this figure), you can determine its highest possible resolution using these formulae:

Horizontal resolution = 4 * sqr(number of pixels/12)

Vertical resolution = 3 * sqr(number of pixels/12)

Unlike the measure of resolution of other computer peripherals, where each pixel requires an individual sensor for each color (three sensor elements equal one pixel), digital camera makers exaggerate their resolutions by using the total number of pixels for all colors to come up with a resolution figure. Most digital cameras use an array of sensors that alternate green with each of the other two colors. Using interpolation algorithms, the camera calculates three individual color values at each element position, regardless of its color sensitivity; in effect, creating three virtual pixels (one of each color) for each image sensor element. A megapixel camera thus has only one million sensing elements.


Image sensors vary in their sensitivity to light. The best can detect a single photon. Those in digital cameras require dozens of photons to make a detected signal. It is similar to the rating of film speed. In fact, to give film photographers a better feel for their digital cameras, many digital cameras allow you to adjust their sensitivities to the equivalent of an ASA or ISO film speed rating (typically the ASA 100 or ASA 400 rating of the most popular films).

The inherent sensitivity of the image sensor is not, however, directly relevant to practical photography. Other aspects of the design of the camera overwhelm concerns about the native sensitivity of image sensors. The most important of these is the lens.

    [ Team LiB ] Previous Section Next Section