Formats, Sensors & Aspect Ratios
Formats
Formats once described the size of the film a particular camera
used, now it describes the sensor size. There are many formats including:
|
Canon 5D
24mm X 36mm
(A.K.A. Traditional or Full Frame) |
|
Canon 30D
15mm X 23mm
(APS "Advanced Photo System" size) |
|
Olympus C4000 zoom
13mm X 17mm
(larger 4:3 size) |
|
Pentax Optio S60
7mm X 9mm |
The size of the sensor directly relates to the quality of the image. The smaller
the sensor, the more the information has to be amplified, resulting in lower
quality images (think of your cell phone.) Generally, the larger the sensor,
the better the image quality.
Sensors
Sensors record the amount of reflected light off the subject and back to the
camera. They then translate the light into specific voltage values (called analog
to digital conversion). Because sensors don't see color, the light is recorded
as luminance and then filtered to determine color values. The two main sensors
used in digital cameras today are the CCD and the CMOS.
CCD
Charged Coupled Device or CCD is the most widely used sensor due to
its high dynamic range and it's smaller size. The CCD sensor doesn't perform
the analog to digital conversion on the sensor, but instead uses additional circuitry
to make the conversion. Nikon's SLRs and most point and shoot cameras use these.
CMOS
Complementary Metal Oxide Semidocnductor or CMOS is named for the components
used to create it. The main differences between the CMOS sensor and the CCD sensor
is that the CMOS has additional circuitry to make the analog to digital conversion
for each pixel on the actual chip. It also has the ability to transfer the data
in parallel, which means much faster transfer from the sensor to the camera circuitry.
Because of the circuitry on each pixel, the light capturing area for each pixel
is reduced thus requiring microlenses to amplify and focus the light. CMOS sensors
use less power by putting all these fuctions on to the sensor typically resulting
in better battery life. Canon's SLRs use these.
Foveon X3
This is a type of CMOS sensor in production that has stacked layers
of sensors that record different colors of light, this allows for a larger range
of color and more accurate color capture due to there not being any color interpolation.
The only cameras in production at this time that use the Foveon sensor are the
Sigma line of digital SLRs. Find out more about them here: http://www.sigmaphoto.com
For more information regarding sensors and how they work go to: http://www.creativepro.com
Aspect Ratios
If you have done any TV or computer monitor shopping lately, then this won't
be new to you. The aspect ratio is the proportions between the length and the
width of the image (or sensor). This means that the 4:3 sensor is making an image
that is 4 parts across to 3 parts high.
Image Size vs. Image Quality
Resolution
Resolution is
a way to describe detail in digital imaging. It more specifically refers to
the number of pixels (short for picture elements) used to make up an image.
It can also be used to describe the image quality of a monitor, screen or display.
More pixels yields a higher resolution and ultimately better image clarity.
Images are usually given a single number to describe their resolution (such
as 72 ppi), this is describing the amount of pixels per linear inch (the number
of pixels that can fit into an inch) in that image or device.
Cameras are described slightly differently, they use a megapixel count (such
as 6 megapixel camera). Essentially they are describing the same information
by multiplying the amount of pixels used in the width by the amount of the pixels
used in the height and then dividing by a million.
Example:
Canon Powershot SD600 -
highest resolution = 2816 x 2112 = 5,947,392 pixels = 5.9 megapixels
All photographic images are made up of pixels arranged
in a grid-like pattern; this is called a bit-mapped image because they are mapped
out where the pixels go and what color they will be. This type of imagery is
also known as "continuous tone" because of its ability to make smooth
transitions between colors.
Image Size
Image Size is the same as the physical dimensions that the image will
have when printed out, even if it will never be printed. The size can be in any
delineation such as inch, cm, mm, foot, picas, and even pixels.
The size is directly proportional to the resolution because the overall amount
of information (the megapixels a.k.a. pixel dimensions) doesn't change, so if
you increase either the size or the resolution, the other will decrease, and
vica versa.
|
So in the above example taken with the EOS 300D (original digital
rebel 6 megapixel camera) the 2048 X 3072 pixel image will print as an
8.5 X 12.8 inch image at a resolution of 240 pixels per linear inch. This
image's pixel dimensions are 18 megabytes of information (total data).
|
 |
The above example shows that changing the image's physical size is possible but
only by lowering the resolution. In order to keep the same pixel dimensions (amount
of data) we can redistribute it between either size or resolution as long as
the amount of data stays consistent. If you "resample" the
image, then you can change the overall file size bigger or smaller (pixel dimensions)
however you either throw away information or add information that was never there
to begin with; this usually results in a less than desireable image. So what
resolution you want really depends on what it's final output is going to be.
Use the following to determine your resolution needs:
72-96
ppi |
Web/Email.
Most screens fall in the 72-96 ppi range, though the
popularity of wide screen monitors and other flat panel
formats make this more of a suggestion these days. |
150-212
ppi |
Photo Ink Jet.
Beyond this and you usually won't see any increase
in print quality. |
250
ppi |
Online Printers.
A normal resolution for online printers for high quality
prints. |
300
ppi |
Minimum resolution for standard magazine printing.
A good archive resolution. |
600
ppi |
Extremely high resolution used
for high quality book or magazine reproductions. |
Physical size also depends on what you are going to do with it. If emailing the
image to a friend across the country is your intention, then making it small
enough to send in an email message is probably your first consideration, not
the quality so much.
By making it small enough to send quickly, you compromise image quality because
you are essentially throwing away information and compressing the file down.
The rule of thumb is that you can always go smaller (throw away info) but you
can't go larger (add info out of thin air). So the idea is to make the image
small enough for the job and yet still look good. Below is a chart showing relative
sizes in print, pixels and data at both 150 ppi and 300 ppi.
| print size |
height by width (pixels)
@ 150 ppi |
height by width (pixels)
@ 300 ppi |
5X7 |
750X1050 = 787500 (2.25M) |
1500X2100 =3150000 (9.01M) |
8X10 |
1200X1500 = 1800000 (5.15M) |
2400X3000 = 7200000 (20.6M) |
11X14 |
1650X2100 = 3465000 (9.91M) |
3300X4200 = 13860000 (39.7M) |
13X19 |
1950X2850 = 5557500 (15.9M) |
3900X5700 = 22230000 (63.6M) |
16X20 |
2400X3000 = 7200000 (20.6M) |
4800X6000 = 28800000 (82.4M) |
24X36 |
3600X5400 = 19440000 (55.6M) |
7200X10800 = 77760000 (222.5M) |
The thing to note is how large the files get as you increase both the physical
size and the resolution. If you divide 3 into the Pixel Dimensions size you will
see what megapixel camera is needed to capture and print that image at that resolution;
Example:
11X14 |
1650 X 2100 = 3465000 (9.91M) |
9.91/3 = requires a 4 megapixel camera |
If using an uncompressed setting (RAW or TIFF) setting in the camera, your sensor
records information for all three color channels (Red, Green,
and Blue) at the max resolution. For example; the
original Digital Rebel is a 6 megapixel camera meaning it captures 6 megapixels
in red, green and blue each,
resulting in an 18 megabyte file (3x6mp=18MB).
Setting Resolution
Your camera should have some way of setting the resolution as well as the quality,
usually via the function or menu button, some are even on the back for quick
access.
The resolution will be written as width and height in pixels or it may be in
a relative size such as Small, Medium or Large; it may also be represented as
Stars or other icons, just remember more or bigger = better.
The above shows the image quality, size and image format. Image quality determines
how much the image will be processed after capture. Usually this means how much
the camera will compress the
image down. In the above example the image is being saved at the highest setting
using a RAW format as well as saving a second image at the highest quality (L=large
size, semicircle=high quality by low compression) jpeg image.
To get the best image quality always set your resolution and quality settings
to their highest values. If you have an uncompressed format
such as RAW or TIFF, use that. The down side is that these will be much larger
files meaning you won't be able to save as many images to your memory card and
that they will take much longer to transfer; but the quality will be much better.
You will also have to process those images after the fact.
Think of it this way; you spent all that money getting the super-mongo megapixel
camera, why not use all of those megapixels instead the same amount that your
phone has?
Image File Formats
JPEG
Jpeg (.jpg)
stands for joint photographic experts group which is the
group that created it. It is considered a "common
file format" meaning that no specific program or
device creates it and that it can be used with many programs
and devices.
It was designed as a photographic format that could be sent relatively quickly
over the internet. The main goal of a jpeg is speed with acceptable quality.
Notice I didn't say best quality? This is because jpegs are a lossy
format, meaning they utilize compression (throw away info) to make images
smaller. The other problem with jpegs are that they only capture using 8 bits
of information resulting in a much smaller range of colors that is possible with
other formats. Most cameras only use jpegs because the general public is more
concerned with speed and how many images they can get on one card, and because
most consumers don't know any better.
RAW
RAW is
the format used by most pros because it contains very minimally processed information
(if any)and captures a much wider color range. Basically, all the information
the camera captures is written to the card without any post processing of color,
size, white balance, etc.
RAW formats are considered Device specific due to the camera that created it
(canon: .crw, cr2, nikon: .nef), so they require special image processing software
(photoshop updates their RAW panel as new cameras are released) since at this
time many operating systems cannot view the RAW thumbnails without jpeg previews.
RAW formats are based on the TIFF format, but have only 12 bits instead of the
native 16 bits that TIFFs have. There is no consumer level camera with a 16 bit
sensor, yet.
The main benefit of shooting in RAW is that the camera is not only utilizing
its entire capabilities, but that post processing doesn't affect the original
image, instead a copy is made and the changes are applied to that copy. This
is very similar to making a print from a negative and in fact Adobe Systems is
trying to gain acceptance of a common RAW format called the Digital
Negative Format (DNG).
TIFF
TIFF stands for
tagged image file format. Tiffs were created in the 1980's (well before digital
cameras were affordable) in an effort to get scanners to scan in the same format.
The biggest benefit was that the format could include much more info including
color information, inclusion of clipping paths, and the ability to contain 16
bits of color information.
Because of the native 16 bits, most tiffs are much larger than the RAW formats,
but in cameras, the RAW format captures to the capacity of the sensor, so if
you have both RAW and TIFF formats, then you are probably only capturing 12 bits
of information and sticking it into a 16 bit container making it larger than
necessary; probably why you don't see both offered on cameras very often. One
of the best things about TIFFs is that they can utilize LZW (lossless)
compression.
You only have JPEG!
So, if you only have the option of Jpegs, don't sweat it, chances are you will
be fine. Just keep in mind that every time you save an image in a photo program,
you tend to loose some information because of the jpeg compression which goes
in and looks at 8x8 pixels to determine how necessary the information is and
then tosses out the "unimportant" info.
The compression does this to
all the pixels (8x8 at a time) every time you save in jpeg format. Depending
on your quality setting in the compression box will determine how aggressive
it is in deciding what goes and what stays. Here's a good link with some additional
info: Jpeg, Tiff, Raw...
Other Image Formats:
.pdf |
Portable
document format. Created by Adobe Systems as a common file format for
word documents. |
.gif |
Graphics
Interchange Format. Created by compuserve for web use. |
.png |
Portable
Network Graphics. Lossless compression. |
.psd |
Photoshop
Document. Native file format for Adobe Photoshop |
|
