A digital camera uses a sensor array of millions of tiny pixels to produce the final image. When you press the shutter of his camera and starts the show, each pixel is a “photosite” to be discovered to collect and store photons in a cavity. After the exposure, the camera closes in each of the photosites, and then try to estimate the number of photons fell into each. The relative amount of photons in each cavity are classified into different levels of intensity, the accuracy is determined by the bit depth (0 – 255 for an 8-bit image).

Each cavity is unable to distinguish the amount of each color has fallen, so the example above would be able to create shades of gray. To capture color images, each cavity must have a filter placed on it, which prevents the penetration of a certain color of light. Virtually all digital cameras today can capture one of the three primary colors in each cavity, so it will take about 2 / 3 of the incoming light. As a result, the camera has to bring the two primary colors to obtain information about the three colors at each pixel. The most common type of color filter array is called a “Bayer matrix”, shown below.

A Bayer pattern consists of alternating rows of red-green filters and blue-green. Notice how the Bayer array contains twice as many green sensors as red or blue. Each primary color does not receive the same fraction of the total area because the human eye is most sensitive to green light than red light and blue. Redundancy with green pixels produces an image that seems to be less noisy and has finer detail than could be achieved if each color were treated equally. This also explains why the noise in the green channel is much lower than for the other two primary colors

Bayer “demosaicing” is the translation process, Bayer array of primary colors in the final image, which includes full-color information for each pixel. How is possible if the camera is not able to directly measure the color? One way to understand this is to think instead of 2×2 each set of red, green and blue as one of the four colors of the cavity.

It would be a good job, but most cameras take additional steps to extract the information of the image in the color table even more. If the camera handles all the colors in each 2×2 box as it landed in the same place, then be able to achieve half the resolution in the horizontal and vertical. On the other hand, when a color camera derived from the superposition of several arrays of 2×2, so he could get a higher resolution than is possible with a single set of 2×2 tables. The following combination of overlapping 2×2 arrays can be used to extract information about the image.

Notice how the information was not calculated image at the very edge of the range, since we assume the continuous image in each direction. If you really were the edges of the die cavity, the calculations here, would be less accurate, since the number of pixels and are not everywhere. This is not a problem because the same information at the edges of the image can be clipped to the cameras with millions of pixels.

There are other demosaicing algorithms that can extract a little more resolution to produce images that are less noisy, or adapt to a better approximation of the image at each location.

You may need the articles below :

Tagged with: • • • • • • • • •

Filed under: Concept

Like this post? Subscribe to my RSS feed and get loads more!