- RAW files are 12, 14 or 16 bit
- in a 16 bit raw file each pixel can have an RGB value of 65,536
- total number of shades for each pixel is 281 trillion
- JPEG files are 8 bit
- each pixel can have an RGB value of 255
- total number of shades for each pixel is 16.8 million
- When looking at a 281 trillion vs. 16.8 million factor it is easy to see how a JPEG reproduction is losing a lot of information.
JPEG is a “lossy” file format that averages out data as it compresses it. In practical terms this means that a typical photo whose pixels have an RGB colour profile will have different values to those actually being photographed. A camera shooting in JPEG mode, or photos reproduced as JPEGs, are actually only sketches of factual data.
A sketch of factual data is fine for non evidential photos but when in court evidence may well need to be presented as fact. In this instance a JPEG will not be suitable for evidential work as it can easily be proven that a JPEG is not an accurate factual representation of what has been photographed.
- can it be proven that shades of colour shown in a photograph are an accurate representation of what was originally in the frame?
- will small areas of photographs need to be blown up/expanded to be used as factual evidence?
- does photographic evidence contain reflections?
- are fingerprint scans used where small inaccuracies and details could render the data useless as evidence?
JPEG reproductions of photographic data are useful as they can speed up workflow as they require less system resources but an efficient system to manage evidential photographic data should, for example, also shoot original photos in RAW format and retain and manage this data in an evidential way for analysis and use where factual evidence is required.