Photographs are an integral part
of human life today. People take
photos for many different reasons
or even without any reason, just
for sake of hobby. With today's
point & shoot cameras, taking
a photograph has become a child's
play. We don't have to bother
about processing of the photos
either, as such a service is
available around every street
corner. In the black&white
era every photographer used to
have his own darkroom for
developing and printing his
films. All the studios were
called photo 'art' studio.
Photography then was an artist's
job. The studio photographer used
to give his own "touch"
to the photo and sign at the
bottom. This artistic speciality
of "finishing" the
photo was possible only because
he had his own darkroom
techniques. Then started the era
of colour photography. Colour
films and paper processing being
relatively difficult to
understand and process, not all
could convert their darkrooms for
colour processing. This made the
way for the development of
automatic colour processing
machines, and resulted in huge
growth of the photographic
market. This mechanised
processing however, took away the
freedom of the photographer over
the composition and he had to be
satisfied with whatever quality a
photo lab offered him.
Now, with the recent
developments in digital
technology it is possible to take
a photo without a film roll!!
This technology has created a
paradigm shift in the science of
photography. It has eliminated
the entire chemistry from the
darkroom and brought photography
into broad "daylight".
That chemistry, which was at the
heart of conventional photo
processing is replaced now by
"software". Many
photographers still do not
believe in the fact that a
permanent change in the way we
look at photography is just round
the corner. Let us discuss in
this article the advantages and
possibilities offered by latest
developments in software
technology used for digital
photography.
With the advent of
the personal computer during the
'80s, printing and publishing
industry started using them for
page composing. Then came photo
editing software to help the
publishing industry to process
photos and make illustrations to
be put into digital page making.
This was when there were no
inkjet printers and digital
cameras. So the development of
photo editing software was done
keeping in mind the requirements
of an artist, who worked for the
printing and publishing industry.
The main emphasis was on
replacing the real drawing tools
like canvas, brushes, pens etc.
with suitable software tools to
make a digital artwork. This
'artwork' then was used for 4
colour separation which is
required for colour printing.
This made a tremendous impact on
the quality of the printed output
as the facilities these software
provided were very much useful to
easily achieve greater quality.
When photo quality
inkjet printers appeared in the
market around end of 1999,
artists started using them with
photo editing software for
restoration of old photos. This
could be done successfully
because all the drawing tools
were available in the software
and a photoquality output could
be taken on the inkjet printer.
Later, when digital camera came
into use people started using the
same photo editing software to
print photos digitally. But as
the requirements of a
photographer are very much
different than that of an artist,
these photo editing software were
of not much use to photographers.
Because photographers require
multiple copies of a photo at a
time and photos are required
strictly in particular sizes.
Changing the size of a photo with
photo editing software is tedious
and mostly results in loss of
details.
Advancements in
digital camera technology and
printers have reached to a level
that we can print photos sitting
at home without a darkroom and
without any chemicals. Thanks to
today's higher end megapixel
digital cameras and photoquality
2400 or 2880 dpi inkjet printers
which give us enough resolution
to make a life size enlargement
of a photo using an appropriate
software.
Let us see what
advantages a higher megapixel
camera offers over a lower one. A
digital camera captures the image
on a CCD(charge coupled device)
or a similar capture device
instead of using a film roll and
stores that image in digital
format. The "Megapixel"
word refers to the number of
pixels the CCD can capture. Thus
a 4 megapixel camera can capture
more details in a scene than a 2
megapixel camera. This image is
stored as a matrix of pixels. The
typical 4 megapixel image has
2448 X 1632 matrix of pixels. So,
obviously a photo taken through a
higher megapixel camera with high
quality setting will be much more
better than the same photo taken
with a lower resolution camera.
Similar is the case
with an inkjet printer, where if
we use a 300 dpi printer to print
a photo, it will result in a
grainy print. Smaller the dot a
printer can print, finer is the
quality of the output. So it is
necessary to have a printer which
offers minimum 2400 dpi
resolution.
With all these
digital devices now in place, we
will see how we can achieve best
quality printing of photos, which
will give justice to the higher
mega pixels of a digital camera
and the printers capacity to
print small details. To do this
we need to understand the
concepts of a pixel, dpi, ppi
etc. We also need to understand
that there is a close relation
between pixels, dpi, file size
and the final printed size.
Pixel :
A pixel is the smallest well
defined piece of information in a
photo file. A pixel carries the
information of colour for a given
area. A pixel, square in size,
can be seen by a human eye on a
final printed output if the dpi
of that file is lower than say
100. Smaller the pixel in a
print, the higher or finer the
quality we get. Optimally the
pixels(dots) per inch should be
one fourth of the photo quality
dpi of the printer, i.e. if your
printer's dpi is 2400 then your
picture should have 600 pixels
per inch and so on. But this is
only for the rule.
DPI :
The full form of dpi is (1) dots
per inch when we refer to a
printer and (2) pixels per inch
(also called PPI) when we refer
to a digital photo file. The
printer always prints, say, 2400
dots per inch in photoquality
mode irrespective of the photo
file dpi. The photo file dpi
information is used by most photo
editing software to let the
printer know to which size the
photo is to be printed. This
means that if a photo of 1200
pixels height X 900 pixels width
has a dpi of 300, then the photo
editing software will tell the
printer driver to print 300
pixels of this photo per linear
inch (90,000 pixels per square
inch) with its 2400 dots per inch
resolution (roughly 2400/300 = 8
dots per pixels) and the final
size after printing will be
1200/300=4 inches height and
900/300=3 inches width. A digital
camera photo file does not
specifically have any dpi
mentioned.
File Size :
This is the memory space required
on a storage device. For a photo
file this varies with the colour
depth, dpi, file type etc. For
example a photo file of 3 inches
by 4 inches with 300 dpi will be
smaller than a photo file of the
same dimensions but with a dpi of
600. Similarly the file size will
increase with more colour depth
per pixel. A pure black &
which picture without any gray in
between can be saved with a
colour depth of just 1bit. For a
photo with all its continuous
colour tonal range the colour
depth has to be 24 bits i.e. 8
bits per colour (Red, Green,
Blue) per pixel. Considering this
a photo file of 1200 x 900 pixels
will have a size of 139 KB at 1
bit colour depth, whereas the
same file will have a file size
of 3,240 KB at 24 bit colour
depth.
File Type :
The information of each pixel in
a photo file is organised in
different ways in each file type
(format). There are many formats
like .bmp, .jpg, .pcx, .tif etc.
The .bmp format is Microsoft
Windows own format. This file
format can store photo
information at 24 bit colour
depth which is a must for the
true to life reproduction of a
photo. But this format at 24 bit
colour depth does not support
compression of information
(data). The file format of .jpg
also stores photo information at
a colour depth of 24 bit and it
also supports compression of
information although with a loss
of information depending on the
percentage of compression. The
compression algorithms used in
.jpg file are very complex. User
can decide between higher quality
- large file size and lower
quality - small file size. The
above mentioned photo file of
1200 x 900 pixels would be
approximately 247 KB at a
compression of 95% for .jpg after
loosing a little bit of quality.
From the above
information we can assume that
for a photo to look like a photo,
we need higher "pixels per
inch" printed with higher
"dots per inch". We can
get higher number of pixels when
we choose a higher megapixel
camera and take a picture with
"high quality" settings
in the camera. To get higher dots
per inch while printing we have
to use a photo quality printer
which has a capacity to print
2400 or more dots per inch.
Now, in between
taking a snap on a camera and
printing that photo on an inkjet
printer, there comes the role of
a suitable software. Because, it
is the software that processes
the information in a digital
photo file and tells the printer
how to print it. There are many
software available today which
can handle digital photo files.
The choice of a particular
software depends on what actually
we want to do with that photo.
There are two distinct things we
normally do with a digital photo
- 1] we edit them for some
purpose like removing or changing
the background, removing red-eye
etc. or 2] we print them on photo
paper. Until recently, the
available software in the market
were only suitable for the
purpose of photo editing, even
though they have the ability to
somehow print the photo. With
these software you have to
"resize" the original
photo to the desired print size
before printing. We will see what
is exactly meant by
"resizing" and what is
the effect of it on the original
photo information.
Let us consider that
the original digital photo has
1600 X 1200 pixels (1.92
megapixel) and we want to print
that to a passport size of
45mm(1.771")x
35mm(1.377"). The aspect
ratio (height : width) of
original photo is (width divided
by height) 1200/1600 = 1 : 0.75,
that is if we scale down height
to 1 the proportionate width
without distorting the photo will
be 0.75. Now the aspect ratio of
passport size photo is different
(35/45 = 1 : 0.777) , so to print
the photo to passport size we
will also need to crop some 44
pixels from the width to bring
the original photo to the exact
aspect ratio of passport size,
otherwise the photo will look
elongated if we forcefully fit
those extra pixels within the
final width. This means that we
have to do two things - first, we
have to cut off some portion from
the width (by changing the paper
size of the image) and secondly,
we have to resample the whole
photo file to the passport size
with a certain dpi which is a
must, otherwise the printer will
not know to which size the photo
is to be printed. Generally a dpi
of 300 is used for printing image
files. So now, in our case if we
resample our 1600x1200 pixel
photo to passport size with a dpi
of 300, we will have a photo of
531 X 411 pixels (300X1.77=531,
300X1.37=411). This means that
the original 1600 pixels have
become 531 pixels and 1200 pixels
have become 411 pixels. In this
process we have lost lot of
information from the original
file. The total number of pixels
we have now is only 218,241 in
place of 1,920,000. We may have
lost some important details like
a few highlights, fine lines of
hair etc. That is why 'resizing'
a photo always results in loss of
original details!!! So it will
not make any difference whether
we use a lower megapixel camera
or a higher one.
Now with the
advancements in software design
with focus on digital photography
a new technology called DPI
independent printing is emerging.
With this we can meet both the
ends in one package. We can print
photos with sharp details in
smaller sizes and also bigger and
grainless enlargements. We do not
need to resize photo information
for different sizes of prints.
Let us see how it works.
The details in a
photo depend directly on the
capacity of the CCD of a digital
camera and of course the lens.
All the information of pixels
captured by the digital camera is
fit into whatever size we want
the photo to be printed. So the
print quality of photo goes on
increasing as we use a higher
megapixel camera. We don't need
to crop or resize the photo. We
have to just select the size and
then the photo to be printed.
Rest all is done by the software.
Let us see how it is
done. A high quality photo taken
through a 3.1 megapixel camera
contains 2160 X 1440 (3,110,400)
pixels. The software uses all
these pixels to print the photo.
If we print this photo in
passport size of 1.771 inches by
1.377 inches the resultant dpi
will be more than 1000 because
all the 1440 pixels of the width
are fit into the width of 1.377
inches (1440/1.377 = 1045 pixels
per inch) and of all the pixels
in the height of 2160 pixels only
1852 pixels will be used for
printing within the height of
1.771 inches (this is because the
software automatically calculates
the aspect ratio of 3.5 : 4.5 for
the passport size and crops the
extra pixels). The result - all
the details are printed as they
are in the photo. This shows that
this new technology does the
"resizing" without
"resampling" the
original information in a photo.
This is very important, otherwise
there will be no point in
purchasing a higher megapixel
camera if all the photos are
going to be printed with a fixed
resolution. Here we will be able
to see the difference in quality
and sharpness of a photo taken
through a webcam with a
resolution of just 640X480 pixels
(0.3 megapixel) and that of a 3.1
megapixel camera giving 2160 x
1440 pixels. This dpi independent
printing gives justice to the
current capacities of megapixel
cameras and photo quality
printers.
Now let us see what
happens when we enlarge a photo.
Digital information can be
manipulated with a suitable
software algorithm. So that more
and more relative information can
be added to original information.
Let us understand this with an
example. We have an image of 1600
X 1200 pixels, which we want to
print at 8"X10". As
mentioned earlier, we can see
individual pixels when they are
larger than about 1/100th of an
inch. By experimenting we can
conclude that for a photo to look
like a photo the software should
send atleast 180 pixels per inch
to the printer. In our case we
have 1200 pixels which are going
to print in a width of 8 inches.
So, we will get 1200/8 = 150
pixels per inch, which is not
suitable for a photo and has to
be increased to 180 pixels per
inch. This is automatically taken
care by the dpi independent
printing software. We do not need
to calculate whether the pixels
are less or more. We get
grainless enlargements with this.
In the future, the
CCD's may be replaced by some
innovative technology which will
capture more and more details.
Also we may have printers which
would deliver more than double
the resolution of currently
available printers. So, remember
if you want to print excellent
prints from your megapixel camera
and see what the camera
resolution is really worth, you
must use a software which uses
dpi independent printing
technology.
Go to the top
