Photoshop 6 for Windows Bible- P32 pot

However, as demonstrated in the middle and right examples inthe figures, you can darken the effects of the Custom filter to create high-contrastimages by entering a negative value into t

Chapter A ✦ Constructing Homemade Effects

Figure A-13: To blur more subtly, increase the central matrix

value and the Scale value by equal amounts

Edge-detection

Many of you are probably beginning to get the idea by now, but just in case you’re

the kind of person who believes that friends don’t let friends do math, I’ll breeze

through it one more time in the venue of edge-detection If you really want to see

those edges, enter 1s and 2s into the neighboring options in the matrix and then

enter a CMV just small enough — it’s a negative value, after all — to make the sum 1.

Examples appear in Figure A-14 for your viewing pleasure

Figure A-14: To create severe edge-detection effects, enter

a negative CMV just small enough to compensate for the

positive values in the matrix

To lighten the edges and bring out the blur, raise the CMV and enter the resultingsum into the Scale option box The first example in Figure A-15 pushes the bound-aries between edge-detection and a straight blur.

Figure A-15: To blur the edges, increase the central matrix

value and then enter the sum into the Scale value

Non-1 variations

Every image shown in Figures A-7 through A-15 is the result of manipulating matrixvalues and using the Scale option to produce a sum total of 1 Earlier in this chap-ter, I showed you what can happen if you go below 1 (black images) or above 1(white images) But I haven’t shown you how you can use non-1 totals to produceinteresting, if somewhat washed-out, effects

The key is to raise the Offset value, thereby adding a specified brightness value toeach pixel in the image By doing this, you can offset the lightening or darkeningcaused by the matrix values to create an image that has half a chance of printing

Lightening overly dark effects

The first image in Figure A-16 uses nearly the exact same values used to create theextreme sharpening effect in the last image of Figure A-8 The only difference is thatthe CMV is 1 lower (8, down from 9), which in turn lowers the sum total from 1 to 0.The result is an extremely dark image with hints of brightness at points of high con-trast The image looks okay on screen — actually, it looks pretty cool because of allthose little star-like sprinkles in it — but it’s likely to fill in during the printing pro-cess If the first image in Figure A-16 looks like anything but a vague blob of black-ness, it’s a miracle Most people would kiss this image good-bye, and rightly so It’stoo darn dark

Chapter A ✦ Constructing Homemade Effects

Figure A-16: Three examples of sharpening effects with

sum totals of 0 I lightened the images incrementally by

entering positive values into the Offset option box

To prevent the image from filling in, lighten the image using the Offset value

Photo-shop adds the value to the brightness level of each selected pixel A brightness

value of 255 equals solid white, so you don’t need to go too high As illustrated by

the last example in Figure A-16, an Offset value of 100 is enough to raise most pixels

in the image to a medium gray Figure A-17 shows the results of lightening an overly

dark edge-detection effect using the Offset value

Figure A-17: Three examples of edge-detection effects with

sum totals of 0, lightened incrementally by using progressively

higher Offset values

Darkening overly light effects

You also can use the Offset value to darken filtering effects with sum totals greaterthan 1 The images in Figures A-18 and A-19 show sharpening and edge-detectioneffects whose matrix totals amount to 2 On their own, these filters produce effectsthat are too light However, as demonstrated in the middle and right examples inthe figures, you can darken the effects of the Custom filter to create high-contrastimages by entering a negative value into the Offset option box

Figure A-18: Three examples of sharpening effects with

sum totals of 2 I darkened the images incrementally by entering negative values into the Offset option box

Figure A-19: Three examples of edge-detection effects with

sum totals of 2, darkened incrementally with progressively lower Offset values

Chapter A ✦ Constructing Homemade Effects

Using extreme offsets

If a brightness value of 255 produces solid white and a brightness value of 0 is solid

black, why in blue blazes does the Offset value permit any number between negative

and positive 9,999, a number 40 times greater than solid white? The answer lies in the

fact that the matrix options can force the Custom filter to calculate brightness values

much darker than black and much lighter than white Therefore, you can use a very

high or very low Offset value to boost the brightness of an image in which all pixels

are well below black or diminish the brightness when all pixels are way beyond white

Figure A-20 shows exaggerated versions of the sharpening, blurring, and

edge-detection effects The sum totals of the matrixes are –42, 54, and 42, respectively

Without some help from the Offset value, each of these filters would turn every

pixel in the image black (in the case of the sharpening effect) or white (blurring and

edge-detection) But as demonstrated in the figure, using enormous Offset numbers

brings out those few brightness values that remain The images are so polarized

that there’s hardly any difference between the three effects, except that the first

image is an inverted version of the other two The difference is even less noticeable

if you lower the opacity of the effect using Filter ➪ Fade, as demonstrated in the

sec-ond row of examples in Figure A-20

Figure A-20: You can

create high-contrasteffects by exaggeratingall values in the matrixand then compensating

by entering a very high

or very low Offset value(top row) When youback off the effect usingFilter ➪ Fade (bottom row), these dramaticeffects are barelydiscernible

Figure A-21: Enter positive matrix values in a horizontal

formation (left) or vertical formation (middle) to create slight motion blurs By positioning positive values in opposite corners of the matrix, you create a vibrating effect (right)

Directional sharpening

To selectively sharpen edges in an image based on the angles of the edges, you canorganize negative and positive matrix values into rows or columns For example, tosharpen only the horizontal edges in an image, fill the middle row of matrix optionswith positive values and the rows immediately above and below with negative val-ues, as demonstrated in the left example in Figure A-22 Similarly, you can sharpenonly the vertical edges by entering positive values in the middle column and flank-ing the column on left and right with negative values, as shown in the middle exam-ple in the figure In the last example, I arranged the positive values along a diagonalaxis to sharpen only the diagonal edges

You even can combine directional sharpening with directional blurring Figure A-23shows the first example from Figure A-22 blurred both horizontally and vertically

To blur the image horizontally, as in the middle example of Figure A-23, I added itive values to the extreme ends of the middle row, thereby extending the range ofthe filter and creating a sort of horizontal jumbling effect To blur the image verti-cally, as in the final example of the figure, I added positive values to the ends of themiddle column

Chapter A ✦ Constructing Homemade Effects

Figure A-22: Arrange positive values in a row (left), column

(middle), or along a diagonal axis (right) to sharpen horizontal,

vertical, and diagonal edges exclusively

Figure A-23: The image from Figure A-22 (left) blurred horizontally (middle)

and vertically (right)

So far, we aren’t going very nuts, are we? Despite their unusual formations, thematrix values in Figures A-21 through A-23 still manage to maintain symmetry Well,now it’s time to lose the symmetry, which typically results in an embossing effect.Figure A-24 shows three variations on embossing, all of which involve positive andnegative matrix values positioned on opposite sides of the CMV (The CMV happens

to be positive merely to maintain a sum total of 1.)

Figure A-24: You can create embossing effects by distributing positive and

negative values on opposite sides of the central matrix value

This type of embossing has no hard and fast light source, but you might imaginethat the light comes from the general direction of the positive values Therefore,when I swapped the positive and negative values throughout the matrix (all exceptthe CMV), I approximated an underlighting effect, as demonstrated by the images

in Figure A-25

In truth, it’s not so much a lighting difference as a difference in edge enhancement.White pixels collect on the side of an edge represented by positive values in thematrix; black pixels collect on the negative-value side So when I swapped the loca-tions of positive and negative values between Figures A-24 and A-25, I changed thedistribution of white and black pixels in the filtered images

Chapter A ✦ Constructing Homemade Effects

Figure A-25: Change the location of positive and negative

matrix values to change the general direction of the light source

Embossing is the loosest of the Custom filter effects As long as you position

posi-tive and negaposi-tive values on opposite sides of the CMV, you can distribute the values

in almost any way you see fit Figure A-26 demonstrates three entirely arbitrary

arrangements of values in the Custom matrix Figure A-27 shows those same effects

downplayed by raising the CMV and entering the sum of the matrix values into the

Scale option box

Figure A-26: You can create whole libraries of embossing

effects by experimenting with different combinations of

positive and negative values

Figure A-27: To emboss more subtly, increase the central

matrix value and the Scale values by equal amounts

Incidentally, the main advantage of using the Custom filter rather than using Filter ➪ Stylize ➪ Emboss to produce embossing effects is that Custom preservesthe colors in an image while Emboss sacrifices color and changes low-contrast portions of an image to gray Figure A-28 shows the matrix values from the firstexample of Figure A-26 applied to a color image It also shows examples of otherCustom effects, including variations on sharpening and edge-detection

If you become enthralled with convolution kernels and you feel compelled toexplore them in more detail, you may want to check out a $200 plug-in called KPT Convolver from MetaCreations Although steeply priced, it permits you

to explore convolutions using a more intuitive interface and without using any math Convolver doesn’t do anything that you can’t already do with the Custom filter, but it can simultaneously apply different convolutions to separatecolor channels, and it permits you to investigate filtering possibilities that youaren’t likely to discover on your own

Chapter A ✦ Constructing Homemade Effects

Figure A-28: Color versions of four Custom filter effects, including (clockwise

from upper left) mild sharpening, offset sharpening, edge-detection, and full-color

embossing

Displacing Pixels in an Image

Photoshop’s second custom effects filter is Filter ➪ Distort ➪ Displace, which

enables you to distort and add texture to an image by moving the colors of certain

pixels in a selection You specify the direction and distance that the Displace filter

moves colors by creating a second image called a displacement map, or dmap

(pro-nounced dee-map) for short The brightness values in the displacement map tell

Photoshop which pixels to affect and how far to move the colors of those pixels:

✦ Black: The black areas of the displacement map move the colors of

corre-sponding pixels in the selection a maximum prescribed distance to the right

and/or down Lighter values between black and medium gray move colors a

shorter distance in the same direction

✦ White: The white areas move the colors of corresponding pixels a maximum

distance to the left and/or up Darker values between white and medium graymove colors a shorter distance in the same direction

✦ Medium gray: A 50 percent brightness value, such as medium gray, ensures

that the colors of corresponding pixels remain unmoved

Suppose that I create a new image window the same size as the scan of the Egyptiantemple carving that I’ve used about sixty times now in this chapter This new imagewill serve as the displacement map I divide the image into four quadrants As shown

in the middle example of Figure A-29, I fill the upper left quadrant with black, thelower right quadrant with white, and the other two quadrants with medium gray.(The arrows indicate the direction in which the quadrants will move colors in theaffected image They do not actually appear in the dmap.)

When finished, I save the dmap to disk in the native Photoshop format so that theDisplace filter can access it I then return to the Egyptian carving image, chooseFilter ➪ Distort ➪ Displace, edit the settings as desired, and open the dmap fromdisk The result is the image shown in the last example of Figure A-29 In keepingwith the distribution of brightness values in the dmap, the colors of the pixels inthe upper left quadrant of the carving image move rightward, the colors of the pix-els in the lower right quadrant move to the left, and the colors in the upper rightand lower left quadrant remain intact

Figure A-29: The Displace filter moves colors in an image

(left) according to the brightness values in a separate image, called the displacement map (middle) The arrows indicate the direction that the map moves colors in the image (right)

A dmap must be a color or grayscale image, and you must save the dmap as a tened image in the native Photoshop file format (If you want to be absolutely surethere are no layers, use the Photoshop 2.0 format.) The Displace command doesnot recognize PICT, TIFF, or any of the other nonnative file formats Who knowswhy? Those programmers move in mysterious ways

flat-Note

Chapter A ✦ Constructing Homemade Effects

At this point, you likely have two questions: How do you use the Displace filter,

and why in the name of all that is good would you possibly want to? The hows of

the Displace filter are covered in the following section To discover some whys —

which should in turn help you dream up some whys of your own — read the “Using

Displacement Maps” section later in this chapter

Displacement theory

Like any custom filtering effect worth its weight in table salt — an asset that has

taken something of a nose dive in the recent millennium — you need a certain

degree of mathematical reasoning skills to predict the outcome of the Displace

fil-ter Though I was a math major in college (well, actually, I double-majored in math

and fine arts, and I must admit to paying the lion’s share of attention to the latter),

I frankly was befuddled by the results of my first few experiments with the Displace

command Don’t be surprised if you are as well With some time and a modicum of

effort, however, you can learn to anticipate the approximate effects of this filter

Direction of displacement

Earlier, I mentioned — and I quote — “The black areas of the displacement map

move colors to the right and/or down the white areas move colors

to the left and/or up.” Yikes, talk about your fragmented quotations I think I’ll sue!

Anyway, the point is, you may have wondered to yourself what all this “and/or” guff

was all about “Is it right or is it down?” you may have puzzled, and rightly so

The truth is that the direction of a displacement can go either way It’s up to you

If you like right, go with it If you like down, don’t let me stop you If you like both

together, by all means, have at it

Beginning to understand? No? Well, it works like this: A dmap can contain one or

more color channels If the dmap is a grayscale image with one color channel only,

the Displace filter moves colors that correspond to black areas in the dmap both to

the right and down, depending on your specifications in the Displace dialog box.

The filter moves colors that correspond to white areas in the dmap both to the

left and up

Figure A-30 shows two examples of an image displaced using a single-channel dmap,

which appears on the left side of the figure (Again, the arrows illustrate the

direc-tions in which different brightness values move colors in the affected image They

are not part of the dmap file.) I displaced the middle image at 10 percent and the

right image at 20 percent Therefore, the colors in the right image travel twice the

distance as those in the middle image, but all colors travel the same direction (The

upcoming section “The Displace dialog box” explains exactly how the percentage

values work.)

Figure A-30: The results of applying a single-channel

displacement map (left) to an image at 10 percent (middle) and 20 percent (right)

However, if the dmap contains more than one channel — whether it’s a color image

or a grayscale image with an independent mask channel — the first channel cates horizontal displacement, and the second channel indicates vertical displace-ment All other channels are ignored Therefore, the Displace filter moves colorsthat correspond to black areas in the first channel of the dmap to the right and col-ors that correspond to white areas to the left (Again, this depends on your specifi-cations in the Displace dialog box.) The filter then moves colors that correspond tothe black areas in the second channel downward and colors that correspond towhite areas upward

indi-Figure A-31 shows the effect of a two-channel dmap on our friend the pharaoh Thetop row shows the appearance and effect of the first channel on the image at 10 per-cent and 20 percent The bottom row shows the appearance and effect of the sec-ond channel

Brightness value transitions

If you study Figure A-31 for any length of time, you’ll notice a marked stretchingeffect around the edges of the image, particularly around the two right images This

is an effect you want to avoid

The cause of the effect is twofold: First, the transition from gray to black and gray

to white pixels around the perimeter of the dmap is relatively quick, especially pared with the gradual transitions in the central portion of the image Second, tran-sitions — reading from left to right, or top to bottom — produce a more noticeableeffect when they progress from light to dark than when they progress from dark tolight The reason for this is that these transitions follow the direction of Photo-

com-shop’s displacement algorithm (I know, when I throw in a word like algorithm,

everybody’s eyes glaze over, but try to stick with me.)

Chapter A ✦ Constructing Homemade Effects

Figure A-31: The horizontal (top row) and vertical (bottom

row) results of applying a two-channel displacement map

(left column) to an image at 10 percent (middle) and

20 percent (right)

For example, in the light-to-dark transition on the left side of the first-channel

dmap in Figure A-31, one gray value nudges selected colors slightly to the right,

the next darker value nudges them an extra pixel, the next darker value another

pixel, and so on, resulting in a machine-gun displacement effect that creates

a continuous stream of the same colors over and over again Hence, the big

stretch

Get it? Well, if not, the important part is this: To avoid stretching an image,

make your dmap transitions slow when progressing from light to dark and

quick when progressing from dark to light For example, in the revised dmap

channels shown in the left column of Figure A-32, the gray values progress

slowly from gray to black, abruptly from black to gray to white, and then slowly

again from white to gray Slow light to dark, fast dark to light The results are

smoother image distortions, as demonstrated in the middle and right columns

of the figure