Resizing
When resizing an image it is important to know what and how you are changing the image information.

Because all image applications are different, the following are guidelines to follow.

Pixel Dimensions
This is the term photoshop uses, but it refers to the Quantity of Data in an image as defined by megapixels.

The "Rule of Thumb" is that however big your image is (for example 6.3 megabytes) you should never change the image so that the pixel dimensions get larger (such as changing a 6.3 megabyte image to 8 megabytes) without loosing image quality.

You can, however, redistribute those original 6.3 megabytes any way you want as long as you don't add any. Because an image is made of information, displayed as physical size (4 in. x 6 in.) and resolution (150 ppi.) you can manipulate one at the detriment of the other.

To keep your pixel dimensions consistent, turn off Re sampling in your image manipulation program if you are able. This locks your pixel dimensions so if you make a change to either print size or resolution, the program will alter the other accordingly without loosing or adding information.

Know your Program
Make sure you know what your program is doing when you hit the "crop" button. Most people have no idea what there program is doing. Almost all programs have some kind of "properties" or "preferences" menu item that will allow you to make some choices for resizing.

If you aren't sure, or if you don't have the ability to specify the exact size (both physical dimensions and image resolution), then change programs. Many programs strive for simplicity, but this is usually at the expense of image quality and user operation.

In addition, most programs that give you preset frame sizes as options for cropping (4x6, 5x7, 8x10) don't take into consideration the original file size or aspect ratio (width and height proportions). This usually results in cropping the image, squishing the image, or adding white area around the image.

In a program like Photoshop, you can maintain the aspect ratio by holding the shift key while dragging the crop tool.

Cropping
When you resize with re sampling off, you are redistributing information; not throwing it away. Any time you crop an image, you are throwing information away that you can't get back.

So if you crop a 6 megapixel image down to 3 megapixels, you can't resize it back to 6 megapixels without a serious loss in quality. Same thing with an 8 x 10 inch image that you cropped down to 3 x 5 inches; that information is gone.

If you try to make the image larger, the program is using an algorithm to make up the missing pixels. Unfortunately, without special software, the image will never have the same quality (sharpness and clarity) as the original.

Compression
Certain formats, such as Jpeg, will automatically use compression when you save them. Compression is utilizing software to throw away certain "unnecessary" information in order to reduce a files size (for web or email). The bias is toward file size, and not image quality.

Jpeg compression looks at 8 pixels high by 8 pixels wide throughout an entire image and then decides what is necessary and what is not. When it decides a pixel is not necessary, it throws the pixel away. This means you are loosing information every time you save an image as a jpeg in a program, which is why we call the jpeg format "Lossy".

You may have some control over the compression, such as the level of compression. This is telling the compression scheme how aggressive to be when it makes its decisions of what stays and what goes. Usually the scale will be from 1-12, 12 being the best quality and the least aggressive.

In a program such as Photoshop, you may have more control such as the "save for web" command which gives you compression settings from 1-100 as well as a visual preview between the proposed settings and the original.

There are formats that use other kinds of compression that are "loss-less" such as the TIFF format, but these are fairly large files, so the compression is not intended for web usage.