3. Raster Images (1/4)

Search in book...
Toggle Font Controls
Create new playlist

Name your new playlist

Playlist description (optional)
Sign In

Email address

Password

Forgot Password?

or

Continue with Facebook

Continue with Google
Sign Up

Full Name

Email address

Confirm Email Address

Password

or

Continue with Facebook

Continue with Google

Raster Images

Most computer graphics images are presented to the user on some kind of raster

display. Raster displays show images as rectangular arrays of pixels. A common

Pixel

is short for “picture el-

ement.”

example is a ﬂat-panel computer display or television, which has a rectangular

array of small light-emitting pixels that can individually be set to different colors

to create any desired image. Different colors are achieved by mixing varying

intensities of red, green, and blue light. Most printers, such as laser printers and

ink-jet printers, are also raster devices. They are based on scanning: there is no

physical grid of pixels, but the image is laid down sequentially by depositing ink

at selected points on a grid.

Color in printers is more

complicated, involving mix-

tures of at least four pig-

ments.

Rasters are also prevalent in input devices for images. A digital camera con-

tains an image sensor comprising a grid of light-sensitive pixels, each of which

records the color and intensity of light falling on it. A desktop scanner contains a

linear array of pixels that is swept across the page being scanned, making many

measurements per second to produce a grid of pixels.

Because rasters are so prevalent in devices, raster images are the most com-

Or maybe it’s because

raster images are so con-

venient that raster devices

are prevalent.

mon way to store and process images. A raster image is simply a 2D array that

stores the pixel value for each pixel—usually a color stored as three numbers, for

red, green, and blue. A raster image stored in memory can be displayed by using

each pixel in the stored image to control the color of one pixel of the display.

But we don’t always want to display an image this way. We might want to

change the size or orientation of the image, correct the colors, or even show the

image pasted on a moving three-dimensional surface. Even in televisions, the dis-

play rarely has the same number of pixels as the image being displayed. Consid-

54 3. Raster Images

erations like these break the direct link between image pixels and display pixels.

It’s best to think of a raster image as a device-independent description of the im-

age to be displayed, and the display device as a way of approximating that ideal

image.

There are other ways of describing images besides using arrays of pixels.

A vector image is described by storing descriptions of shapes—areas of color

bounded by lines or curves—with no reference to any particular pixel grid. In

essence this amounts to storing the instructions for displaying the image rather

than the pixels needed to display it. The main advantage of vector images is that

they are resolution independent and can be displayed well on very high resolution

devices. The corresponding disadvantage is that they must be rasterized before

they can be displayed. Vector images are often used for text, diagrams, mechani-

cal drawings, and other applications where crispness and precision are important

and photographic images and complex shading aren’t needed.

In this chapter, we discuss the basics of raster images and displays, paying

particular attention to the nonlinearities of standard displays. The details of how

Or: you have to know what

those numbers in your im-

age actually mean.

pixel values relate to light intensities are important to have in mind when we

discuss computing images in later chapters.

3.1 Raster Devices

Before discussing raster images in the abstract, it is instructive to look at the basic

operation of some speciﬁc devices that use these images. A few familiar raster

devices can be categorized into a simple hierarchy:

• Output

– Display

∗ Transmissive: liquid crystal display (LCD)

∗ Emissive: light emitting diode (LED) display

– Hardcopy

∗ Binary: ink-jet printer

∗ Continuous tone: dye sublimation printer

• Input

– 2D array sensor: digital camera

– 1D array sensor: ﬂatbed scanner

3.1. Raster Devices 55

3.1.1 Displays

Current displays, including televisions and digital cinematic projectors as well as

displays and projectors for computers, are nearly universally based on ﬁxed arrays

of pixels. They can be separated into emissive displays, which use pixels that

ON OFF ON

–

anode

cathodes

LEDs

Figure 3.1. The opera-

tion of a light-emitting diode

(LED) display.

directly emit controllable amounts of light, and transmissive displays,in which

the pixels themselves don’t emit light but instead vary the amount of light that

they allow to pass through them. Transmissive displays require a light source to

illuminate them: in a direct-viewed display this is a backlight behind the array;

in a projector it is a lamp that emits light that is projected onto the screen after

passing through the array. An emissive display is its own light source.

pixel subpixels

Figure 3.2. The red,

green, and blue subpixels

within a pixel of a ﬂat-panel

display.

Light-emitting diode (LED) displays are an example of the emissive type.

Each pixel is composed of one or more LEDs, which are semiconductor devices

(based on inorganic or organic semiconductors) that emit light with intensity de-

pending on the electrical current passing through them (see Figure 3.1).

The pixels in a color display are divided into three independently controlled

subpixels—one red, one green, and one blue—each with its own LED made us-

ing different materials so that they emit light of different colors (Figure 3.2).

Grooved

alignment layers

Horizontal

polarization filter

Liquid crystals

Vertical

polarization filter

Polarized light

Unpolarized

backlight

Grooved

alignment layers

Horizontal

polarization filter

Liquid crystals

Vertical

polarization filter

Unpolarized

backlight

Figure 3.3. One pixel of an LCD display in the off state (bottom), in which the front polarizer

blocks all the light that passes the back polarizer, and the on state (top), in which the liquid

crystal cell rotates the polarization of the light so that it can pass through the front polarizer.

Figure courtesy Erik Reinhard

(Reinhard et al., 2008).

56 3. Raster Images

When the display is viewed from a distance, the eye can’t separate the individual

subpixels, and the perceived color is a mixture of red, green, and blue.

Liquid crystal displays (LCDs) are an example of the transmissive type. A

liquid crystal is a material whose molecular structure enables it to rotate the po-

larization of light that passes through it, and the degree of rotation can be adjusted

by an applied voltage. An LCD pixel (Figure 3.3) has a layer of polarizing ﬁlm

behind it, so that it is illuminated by polarized light—let’s assume it is polarized

horizontally.

A second layer of polarizing ﬁlm in front of the pixel is oriented to trans-

mit only vertically polarized light. If the applied voltage is set so that the liquid

crystal layer in between does not change the polarization, all light is blocked and

the pixel is in the “off” (minimum intensity) state. If the voltage is set so that

ON OFF ON

–

polarizer

backlight

liquid

crystal

Figure 3.4. The opera-

tion of a liquid crystal dis-

play (LCD).

the liquid crystal rotates the polarization by 90 degrees, then all the light that en-

tered through the back of the pixel will escape through the front, and the pixel

is fully “on”—it has its maximum intensity. Intermediate voltages will partly

rotate the polarization so that the front polarizer partly blocks the light, result-

ing in intensities between the minimum and maximum (Figure 3.4). Like color

LED displays, color LCDs have red, green, and blue subpixels within each pixel,

which are three independent pixels with red, green, and blue color ﬁlters over

them.

Any type of display with a ﬁxed pixel grid, including these and other tech-

nologies, has a fundamentally ﬁxed resolution determined by the size of the grid.

For displays and images, resolution simply means the dimensions of the pixel

The resolution of a dis-

play is sometimes called

its “native resolution” since

most displays can handle

images of other resolutions,

via built-in conversion.

grid: if a desktop monitor has a resolution of 1920 × 1200 pixels, this means that

it has 2,304,000 pixels arranged in 1920 columns and 1200 rows.

An image of a different resolution, to ﬁll the screen, must be converted into a

1920 × 1200 image using the methods of Chapter 9.

3.1.2 Hardcopy Devices

The process of recording images permanently on paper has very different con-

straints from showing images transiently on a display. In printing, pigments are

distributed on paper or another medium so that when light reﬂects from the pa-

per it forms the desired image. Printers are raster devices like displays, but many

printers can only print binary images—pigment is either deposited or not at each

grid position, with no intermediate amounts possible.

An ink-jet printer (Figure 3.5) is an example of a device that forms a raster

image by scanning. An ink-jet print head contains liquid ink carrying pigment,

paper

inkjet

head

ink

droplet

Figure 3.5. The operation

of an ink-jet printer.

which can be sprayed in very small drops under electronic control. The head

3.1. Raster Devices 57

moves across the paper, and drops are emitted as it passes grid positions that

should receive ink; no ink is emitted in areas intended to remain blank. After

each sweep the paper is advanced slightly, and then the next row of the grid is laid

down. Color prints are made by using several print heads, each spraying ink with a

There are also continuous

ink-jet printers that print in

a continuous helical path

on paper wrapped around a

spinning drum, rather than

moving the head back and

forth.

different pigment, so that each grid position can receive any combination of differ-

ent colored drops. Because all drops are the same, an ink-jet printer prints binary

images: at each grid point there is a drop or no drop; there are no intermediate

shades.

An ink-jet printer has no physical array of pixels; the resolution is deter-

mined by how small the drops can be made and how far the paper is advanced

after each sweep. Many ink-jet printers have multiple nozzles in the print head,

enabling several sweeps to be made in one pass, but it is the paper advance,

not the nozzle spacing, that ultimately determines the spacing of the rows.

The thermal dye transfer process is an example of a continuous tone printing

process, meaning that varying amounts of dye can be deposited at each pixel—it

is not all-or-nothing like an ink-jet printer (Figure 3.6). A donor ribbon contain-

ing colored dye is pressed between the paper, or dye receiver,andaprint head

containing a linear array of heating elements, one for each column of pixels in the

image. As the paper and ribbon move past the head, the heating elements switch

donor

ribbon

dye

receiver

linear array

thermal

print head

Figure 3.6. The opera-

tion of a thermal dye trans-

fer printer.

on and off to heat the ribbon in areas where dye is desired, causing the dye to dif-

fuse from the ribbon to the paper. This process is repeated for each of several dye

colors. Since higher temperatures cause more dye to be transferred, the amount of

each dye deposited at each grid position can be controlled, allowing a continuous

range of colors to be produced. The number of heating elements in the print head

establishes a ﬁxed resolution in the direction across the page, but the resolution

along the page is determined by the rate of heating and cooling compared to the

speed of the paper.

Unlike displays, the resolution of printers is described in terms of the pixel

density instead of the total count of pixels. So a thermal dye transfer printer that

has elements spaced 300 per inch across its print head has a resolution of 300

pixels per inch (ppi) across the page. If the resolution along the page is chosen

to be the same we can simply say the printer’s resolution is 300 ppi. An ink-jet

The term “dpi” is all too of-

tenusedtomean“pixels

per inch,” but dpi should

be used in reference to bi-

nary devices and ppi in ref-

erence to continuous-tone

devices.

printer that places dots on a grid with 1200 grid points per inch is described as

having a resolution of 1200 dots per inch (dpi). Because the ink-jet printer is a

binary device, it requires a much ﬁner grid for at least two reasons. Because edges

are abrupt black/white boundaries, very high resolution is required to avoid stair-

stepping, or aliasing, from appearing (see Section 8.3). When continuous-tone

images are printed, the high resolution is required to simulate intermediate colors

by printing varying-density dot patterns called halftones.

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.

Table of Contents for 3. Raster Images (1/4)

Create new playlist

Sign In

Sign Up

Table of Contents for
3. Raster Images (1/4)