Introduction
Light rays from the scene are focused by the imaging lens onto a small electronic chip called the image sensor [1]. The sensor measures this light and converts it to a digital image. Figure 1 shows Bigshot's image sensor. The sensor works much like the retina inside the eye [3]. On the retina, tiny cells convert incoming light to electric signals. The signals are carried to the brain, where they are interpreted as images. Similarly, the image sensor has a grid of detectors called picture elements or pixels for short. Hover over Figure 1 to see how an image sensor's pixels look under a microscope. When exposed to light, each pixel generates an electric charge that depends on the intensity of light. The charge is converted to a digital number and stored in memory. The digital numbers collected from all the pixels forms a digital image. While pixels can measure the brightness of the light falling on them, they cannot measure the color (wavelength [4]) of the light on their own - they are completely color blind. To measure color, we can place color filters on top of each pixel. Most of the colors that we see can be measured using just three types of filters - red, green, and blue. Each pixel measures just one of these three colors, but the rest of the colors can be "guessed" by comparing groups of nearby pixels. Besides capturing images, Bigshot's image sensor can perform an impressive amount of image processing. It not only enhances the brightness and color of the captured image, but also "compresses" the image so that less memory is needed for storage. All of this, and more, is achieved by a chip that is just 9 mm X 9 mm X 1 mm in size - smaller than a penny!
References
|
