Lossless And Lossy Digital Photo File Compression

By Author: Jack Church
Total Articles: 158
Comment this article

Digital photos are stored as digital files on electronic media. Digital photos are comprised of pixels each one with a unique color and intensity. While digital photos do not have to be compressed in most cases they are. Compression serves one goal make files smaller and save storage space.

A digital photo is a built of pixels. A pixel is a single dot in the photo that has a value. A pixel's value represents its color and intensity. Usually each pixel is represented by an RGB value (Red, Green and Blue that combined together create the pixel's color) that occupies 3 bytes. Digital photo files can be very big. For example if you shoot a photo using an 8 mega pixels digital camera the photo will have 8000000 pixels each one occupying 3 bytes. The total file size would be 8000000*3=24000000 or 24Mbytes. However ff compressed using the right technology this file could be much smaller - about 3Mbytes would be common.

There are two main types of compression technologies: lossy and lossless:

Lossless compression: Lossless compression means that if you take a file compress it and then decompress ...
... it - the decompressed file would be the exact copy of the original file. With lossless compression no data is lost in the compression process - the compression software uses better representation of the data in the file but it does not remove any data from it.

Lossy compression: Lossy compression means that if you take a file, compress it and then decompress it - the decompressed file would be slightly different than the original one. The compression software not only represents the file's data more efficiently but it also removes data that it analyzed as being minor or not important. Such data can be removed without hardly any noticeable degradation in the compressed file quality. The differences between the original file and the decompressed file are minor and negligible to the user.

Lossless compression is usually applied to text and other data where all data is equally important. For example when compressing the text in this article and later on decompressing it you would want to get the exact original article without any words or sentences dropped by the compression software that decided they were not important.

Lossy compression on the other hand is usually applied to digital photos and graphic files. Such files include data that the viewer would not be able to notice if removed. For example small changes to color in a photo might not be noticeable. The decompressed file is not exactly the same as the original one but when viewing both the original and decompressed photos side by side the viewer can not tell the difference. Lossy compression software can be set to different compression levels based on how much data is allowed to be lost. At some point losing too much data is noticeable and degrades the digital photo quality. Many digital cameras allow you to set the level of compression from low to high where high compression means smaller files but less quality and low compression means bigger files but no noticeable quality degradation.

Digital photos are almost solely compressed using lossy technologies. The reason is that due to the nature of a photo (it includes noise, very minor changes that are hard to compress but are not important to the view and more) lossless compression technologies do not perform well on it. On the other hand Lossy compression technologies can be very effective in reducing digital photo file sizes - sometimes an order of magnitude or more. Jack Church explains at digital prints