Computer graphics comes in two flavors, which are referred to by the same words I used earlier to differentiate graphical video displays: vector and raster. Vector graphics involves creating images algorithmically using straight lines, curves, and filled areas. This is the province of the computer-assisted drawing (or CAD) program. Vector graphics finds its most important application in engineering and architectural design. A vector graphics image can be stored in a file in a format referred to as a metafile. A metafile is simply a collection of vector graphics drawing commands usually encoded in binary form.
The use of lines, curves, and filled areas of vector graphics is entirely appropriate when you're designing a bridge but hopelessly inadequate when you want to show what the actual constructed bridge looks like. That bridge is a real-world image. It's simply too complex to be represented by vector graphics.
Raster graphics (also known as bitmap graphics) comes to the rescue. A bitmap encodes an image as a rectangular array of bits that correspond to the pixels of an output device. Just like a video display, a bitmap has a spatial dimension (or resolution), which is the width and height of the image in pixels. Bitmaps also have a color dimension (or color resolution, or color depth), which is the number of bits associated with each pixel. Each pixel in a bitmap has the same number of bits.
Although a bitmap image is two dimensional, the bitmap itself is just a single stream of bytes — usually the top row of pixels, followed by the second row, followed by the third row, and so on.