computer graphics
[kəm′pyüd·ər ′graf·iks]Computer graphics
A branch of computer science that deals with the theory and techniques of computer image synthesis. Computers produce images by analyzing a collection of dots, or pixels (picture elements). Computer graphics is used to enhance the transfer and understanding of information in science, engineering, medicine, education, and business by facilitating the generation, production, and display of synthetic images of natural objects with realism almost indistinguishable from photographs. Computer graphics facilitates the production of images that range in complexity from simple line drawings to three-dimensional reconstructions of data obtained from computerized axial tomography (CAT) scans in medical applications. User interaction can be increased through animation, which conveys large amounts of information by seemingly bringing to life multiple related images. Animation is widely used in entertainment, education, industry, flight simulators, scientific research, and heads-up displays (devices which allow users to interact with a virtual world). Virtual-reality applications permit users to interact with a three-dimensional world, for example, by “grabbing” objects and manipulating objects in the world. Digital image processing is a companion field to computer graphics. However, image processing, unlike computer graphics, generally begins with some image in image space, and performs operations on the components (pixels) to produce new images.
Computers are equipped with special hardware to display images. Several types of image presentation or output devices convert digitally represented images into visually perceptible pictures. They include pen-and-ink plotters, dot-matrix plotters, electrostatic or laser-printer plotters, storage tubes, liquid-crystal displays (LCDs), active matrix panels, plasma panels, and cathode-ray-tube (CRT) displays. Images can be displayed by a computer on a cathode-ray tube in two different ways: raster scan and random (vector) scan. See Cathode-ray tube, Computer peripheral devices
Interaction with the object takes place via devices attached to the computer, starting with the keyboard and the mouse. Each type of device can be programmed to deliver various types of functionality. The quality and ease of use of the user interface often determines whether users enjoy a system and whether the system is successful. Interactive graphics aids the user in the creation and modification of graphical objects and the response to these objects in real-time. The most commonly used input device is the mouse. Other kinds of interaction devices include the joystick, trackball, light pen, and data tablet. Some of these two-dimensional (2D) devices can be modified to extend to three dimensions (3D). The data glove is a device capable of recording hand movements. The data glove is capable of a simple gesture recognition and general tracking of hand orientation.
In the production of a computer-generated image, the designer has to specify the objects in the image and their shapes, positions, orientations, and surface colors or textures. Further, the viewer's position and direction of view (camera orientation) must be specified. The software should calculate the parts of all objects that can be seen by the viewer (camera). Only the visible portions of the objects should be displayed (captured on the film). (This requirement is referred to as the hidden-surface problem.) The rendering software is then applied to compute the amount and color of light reaching the viewer eye (film) at any point in the image, and then to display that point. Some modern graphics work stations have special hardware to implement projections, hidden-surface elimination, and direct illumination. Everything else in image generation is done in software.
Solid modeling is a technique used to represent three-dimensional shapes in a computer. The importance of solid modeling in computer-aided design and manufacturing (CAD/CAM) systems has been increasing. Engineering applications ranging from drafting to the numerical control of machine tools increasingly rely on solid modeling techniques. Solid modeling uses three-dimensional solid primitives (the cube, sphere, cone, cylinder, and ellipsoid) to represent three-dimensional objects. Complex objects can be constructed by combining the primitives. See Computer-aided design and manufacturing, Computer-aided engineering
The creation of images by simulating a model of light propagation is often called image synthesis. The goal of image synthesis is often stated as photorealism, that is, the criterion that the image look as good as a photograph. Rendering is a term used for methods or techniques that are used to display realistic-looking three-dimensional images on a two-dimensional medium such as the cathode-ray-tube screen (see illustration). The display of a wire-frame image is one way of rendering the object. The most common method of rendering is shading. Generally, rendering includes addition of texture, shadows, and the color of light that reaches the observer's eye from any point in the image.
)
Computer-generated images are used extensively in the entertainment world and other areas. Realistic images have become essential tools in research and education. Conveying realism in these images may depend on the convincing generation of natural phenomena. A fundamental difficulty is the complexity of the real world. Existing models are based on physical or biological concepts. The behavior of objects can be determined by physical properties or chemical and microphysical properties.
graphics
(1) The computer's display system. See display adapter, discrete graphics, integrated graphics and graphics pipeline.(2) The creation and manipulation of picture images. All graphics terms in this encyclopedia refer to "computer graphics." A fast computer is required for graphics work, and although mice are widely used to create illustrations, a drawing tablet is also used for precise illustrations (see digitizer tablet and CAD).
Vector and Bitmapped Graphics
The major categories of digital graphics structures are vector graphics and bitmapped graphics. To understand this field, knowing how these two architectures are used and intertwine is necessary.
 |
| Drawing, Scanning and Painting |
|---|
| When pictures are "drawn" (top), they become vector graphics. When they are photographed or scanned, they become bitmaps (a matrix of pixels). Images can also be "painted" (see paint program). |
 |
| Vector vs. Bitmap |
|---|
| Intricate illustrations can be made with drawing (vector) programs. However, nothing can provide the realism of a photograph or scan (bottom). (Images courtesy of Adobe Systems, Inc.) |
Vector Graphics for CAD and Drawing
Vector graphics are a series of points, lines, arcs and other geometric shapes. They are created in computer-aided design (CAD) and drawing programs. As you draw, each line is stored as a vector, which is two end points on an x-y matrix. For example, a square becomes four vectors, one for each side. A circle is turned into dozens or hundreds of tiny straight lines, the number of which is determined by the resolution of the drawing. The image is stored in the computer as a list of vectors.
Vector Graphics Are Easily Scaled
Each vector element maintains its own identity and can be selected and manipulated independently of the others. Any element can be independently scaled (resized) smaller or larger.
From Vectors to Bitmaps
Monitors display pixels, and printers print dots, both of which are bitmaps. Therefore, vector images must be converted into bitmaps ("rasterized") for the screen and printer. The rasterization process is performed by the operating system and printer language (see page description language).
3D Graphics
3D images are vector graphics, but 3D CAD and drawing programs are significantly different than 2D programs. Objects are created in 3D form in a 3-dimensional workspace. They can be viewed at any angle by simply rotating them, whereas in 2D programs, the object would have to be redrawn entirely. 3D programs can render the drawing with lights and shadows, and camera angles and light sources are used to depict the objects as real-world elements.
 |
| The 3D Stage |
|---|
| In 3D graphics, objects are created on a 3-dimensional stage where the current view is derived from the camera angle and light sources, similar to the real world. (Image courtesy of Intergraph Computer Systems.) |
Bitmapped Graphics for Imaging and Painting
Bitmapped graphics, also known as "raster graphics," are made up of dots like TV images. Each image is divided into horizontal rows, with each row divided into "pixels" (dots). There can be millions of pixels in a single image (see megapixel).
Bitmapped graphics are created manually in image editor and paint programs. They can also be scanned from paper documents, photographed by digital cameras, recorded by video cameras, as well as extracted from the computer screen (see screen capture).
Many Formats, All Digital
Unlike TV, which uses one family of formats for the country, there are dozens of different vector and bitmapped graphics standards. See graphics formats and DTV.
Bits Per Pixel
When an image is scanned into or "painted" on a computer, the bitmap is created in a reserved area of RAM with some number of bits corresponding to each pixel. The simplest monochrome bitmap uses one bit (on/off) per pixel. Gray scale bitmaps store a number for each pixel corresponding to a shade of gray; for example, 8 bits holds 254 gray levels plus black and white.
Color bitmaps require three times as much storage in order to represent the shades of red, green and blue. Since colors are designated with numbers, changing red to green is a process of searching for the red number and replacing it with the green number. See color depth.
Although often compressed further to save space, bitmapped image files are typically larger than their vector counterpart. Storage for each pixel is required whether part of the object or the background. A small object in a vector image requires storage of only a few vectors.
From Bitmaps to Bitmaps
Although bitmapped images are already in a raster format, they typically have to be combined with other bitmaps, vector and text elements and be "rasterized" into a bitmap of the screen's resolution or the printed page.
 |
| Drawing vs. Painting |
|---|
| Although more painting tools are added to drawing programs and more drawing tools are added to paint programs, their inherent structure is different. Drawing programs (vector graphics) allow for the creation of objects that can be manipulated independently. Paint programs (bitmapped graphics) provide a canvas that can be covered with electronic paint. |
 |
| Canvas Specializes in Both |
|---|
| Deneba Software's Canvas combines extensive drawing and imaging tools in one program. The PC drawing on top is a vector graphics rendering and the "first mouse" underneath is a bitmap. The open menu shows the image editing tools. |
 |
| Getting Closer All the Time |
|---|
| A major goal is to create virtual people who look real, and it took 19 rendering passes in 2004 to create this lovely lady. Also, this JPEG image is 1/74th the size of the original 2.3MB TIFF file. It was compressed to 32KB to save space in this encyclopedia (see JPEG). (Image courtesy of NVIDIA Corporation.) |