MAKING SENSE OF IMAGE SIZES

This short tutorial is designed to remove some of the confusion about the sizes of images and image files.

The screen shots in this tutorial are taken on an Apple computer running Photoshop CS3 so they may look a little different on your computer.

This information was prepared by Roy Killen and is copyright. It is part of the supplementary training materials for members of the Belmont 16 Ft Photographic Club and is not intended as a complete, stand-alone instruction package.

Comments and suggestions on this tutorial are welcome - they should be emailed to Roy Killen.

Making sense of the camera sensor

Digital camers have a SENSOR that takes the place of film in a film camera.

The diagram at the right represents the surface of the sensor. It is divided into small, isolated areas called PIXELS. Each coloured square on the diagram represents one PIXEL. (Some cameras do not have square pixels - but that's another story.)

Each pixel in the sensor responds to just one colour of light - red, green or blue - and produces an electrical signal that depends on the intensity of light falling on it.

The green pixels measure the green light, the red pixels measure the red light and the blue pixels measure the blue light.

The digital sensor in the majority of cameras is called a BAYER PATTERN sensor. This name comes from the arrangement of red, green and blue light-sensitive areas, as shown in the diagram.

There are 2 green sensitive pixels for each red and blue pixel. (This is because the human eye is more sensitive to green and it is an attempt to get the overall response of the sensor to be similar to that of a human eye.)

Most medium to high quality cameras have a 12-bit sensor. This means that each pixel can distinguish between 2096 different levels of light intensity.

Some high quality cameras have 14-bit sensors. This means that each pixel can distinguish between 16384 different levels of light intensity.

SENSOR SIZE - MEGAPIXELS

The "size" of the sensor is often quoted as the total number of pixels. For example, 8 megapixels (8MP) or 12 megapixels.

It is a common marketing ploy to claim that "more is better" - advertisers want you to beleive that a 12 megapixel camera must be better than an 8 megapixel camera (so that you will pay more for it). But this is not necessarily so.

If you compare two cameras that have the same physical size sensors (e.g., 24mm x 16 mm) then an 8 megapixels camera will have larger pixels than a 12 megapixel camera.

If you compare two cameras that have the same number of pixels on their sensors (e.g., 12 MP) then a camera with a physically larger sensor (e.g., full frame) will have larger pixels than a camera with a smaller (e.g., DX) sensor.

The physical size of the individual pixels can make a difference to the quality of image produced by the camera. As a general rule, cameras with larger pixels are more sensitive to light and tend to produce less noise (grain) in images - particularly at high ISO settings. So, you need to consider both the number of pixels on the sensor and the physical size of the sensor before deciding which camera will best suit your needs.

FROM LIGHT TO DIGITAL DATA

The electrical signal produced by the sensor represents the intensity of the light falling on each pixel.

The electronics in the camera convert this information into a digital file that will later be used to reconstruct the image on a computer screen or a printer.

The "reconstructed" image (what you see on the computer screen) is also made up of pixels - individual areas of colour arranged in rows and columns. However, the screen pixels are quite different from the camera sensor pixels. (More about that in a later tutorial.)

If your camera is saving JPG or TIFF files then the electronics in your camera uses the information from each sensor pixel (and the information from surrounding pixels) to produce a digital image file that can later be used to reconstruct the image on a screen.

If your camera is saving RAW files then most of the process of converting the sensor data to an image file takes place in your computer (not in the camera).

In either case, each pixel in the converted image now has three parameters: red intensity, blue intensity and green intensity. That is why it is called an RGB image. This means that each pixel in the image contains informatoin that allows it to produce a colour that is somewhere between total black (no red, no green, no blue) and total white (maximum red, maximum green and maximum blue). This information is used to reconstruct the image on screen or on a printer.

FILE SIZES - FROM THE CAMERA

The size of the files (in megabytes) that your camera saves to the memory card depends on the following:

1. The number of pixels on the sensor - for example, a 12MB sensor produces more data and bigger files than an 8MB sensor.

2. The "bit depth" of the sensor - for example, a 14-bit sensor produces more data from each pixel than a 12-bit sensor.

3. The type of files you are saving - JPG files will be smaller than TIFF files or RAW files (because the JPEG files are compressed).

4. The way in which your camera processes the files before saving them. This will vary according to such things as the "quality" setting you have chosen on the camera if you are saving JPG files.

5. For JPG files, the information in the original scene can change the file size. Scenes with many colours and tonal variations tend to produce slightly larger files than scenes with few colour and tonal variations. If there are many lines, details or textures in the image the JPG file will also be larger.

Some cameras save all JPGs with almost identical file size. This is because the camera is programmed to change the quality of the saved image to fit a predetermined size.

Other cameras are programmed to keep the quality of the JPEG images constant and let the file size vary to suit each image.

Some cameras allow you to choose (through the menu) whether you want the JPG files to be a constant size or a constant quality - if your camera gives you this choice, it is best to choose constant (high) quality.

NOTE: The size of the files produced by the camera (in megabytes) will NOT be the same as the sensor size in megapixels. For example, a 12 megapixel camera might produce 18MB RAW files and 5MB JPG files.

NOTE: As a general guide - the larger the size of the files saved by your camera the more information they contain and, therefore, the better your chances of being able to reproduce (on the screen or on paper) something that approaches the colours, tonal variation and detail in the original scene you photographed. So, save the largest files that you can on your camera.

IMAGE SIZES

When your camera saves an image file it will have certain dimensions in pixels. For example, it might be 4256 x 2832 pixels.

If you are saving RAW files these pixel dimensions are the same as the pixel dimensions of the sensor.

If you are savng JPEG files, the pixel dimensions of the image will depend on the "JPEG quality" setting you have chosen on your camera.

If you set the camera to lower "JPEG quality" settings the image will have fewer pixels (and the file size will be smaller). However, if you want good quality photographs there is no point in setting your camera to save anything less than the highest quality JPEG images. It is even better to save RAW files because they have not been compressesed (so no captured information is lost).

The size of the image (in pixels) determines how large the image will appear on a screen (at 100% size). The number of pixels in an image also determines how large you can print it and still get an image that looks good at normal viewing distances.

IMAGE SIZES - IN PHOTOSHOP (or similar image editing programs)

When you open an image in Photoshop you can check its size by selecting Image>Image Size from the menu.

This will open a dialogue box similar to the one on the right.

Notice that the size is shown in two different ways:

1. The PIXEL DIMENSIONS of the image. This is the size of the image produced by the camera (assuming you have not cropped it).

2. The DOCUMENT SIZE. This is the size that the image would be if printed at the given resolution (72 pixels/inch in this example).

If you deliberately want to change the pixel dimensions of the image you must first put a tick in the "Resample Image" box. You then get the option of selecting the dimensions in pixels or percentage.

You can type the new pixel dimensions that you want in either the "width" or "height" boxes.

For example . . .

Here I have typed "1500" as the desired width of the image and the height has automatically changed to 1004 pixels because the dimensions are "linked" (as indicated by the small link symbol to the right of the pixel dimensions).

Notice that the "document size" has changed because I have reduced the number of pixels in the height and width but I have not changed the resolution (it is still at 300 pixels/inch).

If you uncheck the "Constrain Proportions" box

1. The "link" symbol disappears, and

2. You can independenetly change the height and width of the image.

Here I have changed the height to 800 pixels.

DANGER! Doing this will distort your image - it will no longer have the same proportions (aspect ratio) as it had before. When editing photographs you will very rarely want to make this type of change.

THERE'S MORE . . .

Once you understand these basic ideas about image and file sizes you are ready to start resizing images for particular purposes - for example getting them ready to enter competitions where you will typically be required to have images that do not exceed 1024x768 pixels.

There is more informaton about how to do this in the next tutorial.

All images on this site and all documents and tutorials linked to this site are copyright.

This page last updated 25th August 2009 (RK)