What we say here is not carved in stone. You can
change what we say according to circunstances. All we can say is that the
methods described are the optimum ways for us at the present time. They
are based on our experience to date.
This can be thought of
in a number of steps. We have put down the following in a particular order
which we consider the best.
1.
Saving.
2.
Sizing.
3.
Cleaning Up the Image.
4.
Manipulation.
5.
Preparing for Printing.
6.
Printing.
1.
Saving. The first step is to
save the scanned image. This must be done first just in case there is a
power failure for instance. When you save an image you will find that there
are several file formats from which to choose. The imaging programme that
use will have its own favourite format. So, for instance, Photoshop will
use .PSD format if you allow it to do. These formats will usually take
up the most room on your hard disk or other storage means. However, they
will save more too. That is, you can save all details of Layers, etc..
We, therefore, recommend
that the ideal format is the TIFF format. This can also save details such
as Layers and has three advantages which are very useful. The first is
that it compresses the file to take up less space on the hard disk. The
second is that this format can be read by virtually all other types of
computer and program. This may not seem important to you as yet, but as
time goes by you may find it useful for file exchanging. The third advantage
is that TIFF is a lossless format, unlike the JPEG format. In turn this
means that if you save your file repeatedly as you manipulate the image
(as you should just in case...) then there will be no degradation of the
image. If you do this with JPEG the image will get worse each time.
When downloading from a
digital camera the image will most likely be in the JPEG format. So the
trick here is to use the SAVE AS method and then pick out the TIFF format
and then save.
2.
Sizing. The second step is to
set the final image size. This is a cunning ploy especially if using small
image files from, say, a digital camera. These files are of the order of
700kb JPEG compressed size and as such will only produce a photographic
quality print of about 8x6 inches. However, if resized to, say, 12x9 inches
at this stage then by the time you have finished manipuilating the picture
any strange effects caused by up-sizing will be masked out.
Resolution should be changed
to that required for printing the final image. This we set between 200
to 300 dpi (dots per inch). There is no reason to any higher than this
because you will not see any difference in the print. If you less than
about 140 dpi then you could see dots in the print.
In addition why not crop
the image to the final shape or thereabouts? This will save disk space
still more.
At this point save your
image file again (just in case there is a power cut).
Also look at page 23 for a fuller account of sizing.
3. Cleaning
Up the Image. [See another
idea]
The third important step is to remove any defects from the image. For
a scanned 35 mm picture this is often because of dust particles that have
inadvertently scanned in as well.
The best method for doing
this is by cloning. We do not recommend use the filter known as 'Dust and
Scratches'. All it seems to do is to make the picture more blurred.
In Photoshop cloning is
known curiously as 'Stamping'. No doubt other imaging programs cloning
has other names.
How
does cloning work?
Generally cloning is merely
copying a small part of an image to another place on the same image in
order to cover up a defect. Bearing this in mind we must first find a defect,
such a dust speck, then select an area we wish to copy over the top. If
you are covering a defect then the area you select to copy must be such
that when cloned across you are unable to see anything peculiar. This implies
that when trying to remove spots and specks then the area from which clone
must be a similar size to the defect. Imaging programs have the means to
change the size of this area by altering the brush size which is measured
in pixels. The brushes available can also come in different flavours. Usually
a very soft brush is the best for this operation. This can lead to smoothing
out of textured parts; so take great care and only clone the minimum you
can get away with.
Size
of Image on the screen
When a picture is loaded
onto your monitor screen it can be almost any size. For cloning it is a
good idea to ensure that the magnification is set to 100%. In Photoshop
the quick way to do this is select either the Hand or the Magnifier tool
and then choose 'Actual pixels'. This will automatically set the magnification
to 100%. Now the picture will most probably be larger than your screen.
Navigate your way to, say, the top left of the screen as a reference point
to start cloning. As you finish working on a screenful you will need to
move onwards. Rather than move using the Hand tool or the Navigator try
clicking on the grey area to the right of the slider control at the bottom
of the screen in order to move exactly one screens width from your starting
position. When you get to the end of the top row you use the same technique
on the side bar. By doing this you can be sure that you have examined and
corrected every part of the image. By using 100% magnification you are
also sure that you can see every spot. If you were to use, say, 50% there
is a chance that you could miss a spot that is only one pixel wide because
at 50% half of all the available pixels are missing from the display.
Aligned
or Non-Aligned that is the Question.
There is another choice
when cloning. Normally you would choose the 'Aligned' mode. This
simply means that the area you choose to clone from and the area you clone
to remain the same distance apart wherever you try cloning on the image.
If, for instance, you choose to clone from an area just to the right of
the area to clone to this relationship stays constant. Now if you lift
your finger from the left mouse key or lift the pen tool (ie Wacom) and
go to another part of the image the area you start cloning will be just
to the right as previously.
For the 'non-aligned'mode,
on the other hand, the area you would clone would be selected from the
area you first sampled which could be a long way from your present position.
So what! If you have a image that has only a small part from which to clone
onto many other areas then using this 'non-aligned' mode enabled you to
clone bit by bit without the sampling area straying into inappropriate
areas. Trying it out is easier than writing it down. Because the clone
stamp tool samples the entire image, this option is useful for applying
multiple copies of the same part of an image to different images.
Position the pointer on
the part of any open image you want to sample, and Alt-click (Windows)
or Option-click (Mac OS). This sample point is the location from which
the image is duplicated as you paint. Drag to paint with the tool.
There are many other options
also available such as whether to use all Layers or the type of Blending
mode. These make life complicated and are best left alone at present.
Now save your image again
before doing anything else.
4.
Manipulation.
This is a vast topic and
is too big for these notes. Each Image Program will use different ways
to achieve the same ends which makes describing them fairly futile. Also
each program will have different filters and so on. Additionally there
are Plug-Ins for Paint Shop Pro or Photoshop which are made by independent
firms who have been allowed to make their extra filters able to blend seamlessly
into the main imaging program. Sometimes they cost extra; sometimes they
can be free depending on their complexity. They can extend the original
program significantly.
In fact this process is
changing all the time and is essentially infinite in extent. Play with
the controls on the program to your hearts content. As long as you do not
save you can always get back to your original picture.
So let us outline the basics
that you will need.
a)
Making Colour and Tonal Changes
Basic steps for correcting images (notes taken from
Photoshop help files).
4.1. Calibrate your monitor.
In preparation for adjusting images, use Adobe
Gamma or another monitor-profiling utility to characterize and calibrate
your monitor to a colour-display standard suited to your working needs.
Otherwise, the image on your monitor may look very different from the same
image when printed or when viewed on another monitor.
4.2. Check the scan quality and tonal range.
Before making adjustments, look at the image's
histogram to evaluate whether the image has sufficient detail to produce
high-quality output. The greater the range of values in the histogram,
the greater the detail. Poor scans and photographs without much detail
can be difficult if not impossible to correct. Too many colour corrections
can also result in a loss of pixel values and too little detail.
The histogram also displays the overall distribution
of shadows, midtones, and highlights to help you determine which tonal
corrections are needed
4.3. Adjust the tonal range
.
Begin tonal corrections by adjusting the values
of the extreme highlight and shadow pixels in the image, setting an overall
tonal range that allows for the sharpest detail possible throughout the
image. This process is known as setting the highlights and shadows or
setting the white and black points.
Setting the highlights and shadows typically
redistributes the midtone pixels appropriately. When pixel values are concentrated
at either end of the tonal range, however, you may need to adjust your
midtones manually. It is not usually necessary to adjust midtones in images
that already have a concentrated amount of midtone detail.
There are several different ways to set an image's
tonal range:
You can drag sliders along the histogram in the
Levels dialog box. You can adjust the shape of the graph in the Curves
dialog box. This method lets you adjust any point along a 0-255 tonal scale
and provides the greatest control over an image's tonal quality. You can
assign target values to the highlight and shadow pixels using either the
Levels or Curves dialog box. This can be a useful method for images intended
for printing on a press.
4.4. Adjust the colour balance.
Begin tonal corrections by adjusting the values
of the extreme highlight and shadow pixels in the image, setting an overall
tonal range that allows for the sharpest detail possible throughout the
image. This process is known as setting the highlights and shadows or
setting the white and black points.
Setting the highlights and shadows typically
redistributes the midtone pixels appropriately. When pixel values are concentrated
at either end of the tonal range, however, you may need to adjust your
midtones manually. It is not usually necessary to adjust midtones in images
that already have a concentrated amount of midtone detail.
There are several different ways to set an image's
tonal range:
You can drag sliders along the histogram in the
Levels dialog box. You can adjust the shape of the graph in the Curves
dialog box. This method lets you adjust any point along a 0-255 tonal scale
and provides the greatest control over an image's tonal quality. You can
assign target
values to the highlight and shadow pixels using either the
Levels or Curves dialog box. This can be a useful method for images intended
for printing on a press
4.5. Make other special colour adjustments.
After correcting the tonal range, you can adjust the image's colour
balance to remove unwanted colour casts or to correct oversaturated or
undersaturated colours. Examine your image with reference to the colour
wheel to determine which colour adjustments you need to make. You can choose
from the following colour adjustment methods:
The colour Balance command changes the overall mixture of colours in
an image. The Hue/Saturation command adjusts the hue, saturation, and lightness
values of the entire image or of individual colour components. The Replace
colour command replaces specified colours in an image with new colour values.
The Selective colour command is a high-end colour-correction method that
adjusts the amount of process colours in individual colour components.
The Levels and Curves dialog boxes let you adjust colour balance by setting
the pixel distribution for individual colour channels. The technique of
blending colours from different channels can also produce colour adjustments.
To best preserve the original
detail in your image as you make colour adjustments, convert the image
to 16 bits per channel. When you finish making colour adjustments, convert
it back to an 8-bit-per-channel image.
4.6. Sharpen the edges of the image.
As a final step, use the Unsharp Mask filter
to sharpen the clarity of edges in the image. This step helps restore focus
to images that have undergone resampling as a result of tonal adjustments.