Article text

File naming and backup solution - Part 1


The process of safely getting your pictures off the camera and storing them is, let's face it, quite basic. However, doing this in a control manner, in terms of immediately backing pictures up and naming them strategically, is a bit more in depth and requires some planning. In this first of the series of articles around file management and backup I'll explain how I approach it both in terms of my business and personally.

I use Nikon cameras for my DSLRs, and a FujiFilm point and shoot camera where I don't want to carry around all the gear required for a DSLR. However, the process that I use to extract the images really isn't reliant upon the particular camera, although, I'll highlight some exceptions though a bit later.

I use a card reader to extract the pictures from my memory cards, personally I have a card reader built as part of my floppy drive, which takes SD cards as well as compact flash and I use "Nikon Transfer 2" to extract the images. This is where the caveat comes in, Nikon Transfer 2 will happily extract any jpg files from any of the cards as well as Nikon RAW files. However, I don't know if this will work for other brand (eg, Canon) RAW files. I would suggest to test it out, or understand the principles behind my strategy and then apply to your own software.

Moving on, with the Nikon Transfer software, and I imagine the Canon counterpart, you have a number of options as the transfer is in progress. Just to be clear though, the transfer software has no image viewer / processor as part of it, it's just about getting the pictures from the memory cards onto the computer. Back to the options as you transfer the files, these, in the Nikon software can be broken down into 3 elements, filename, location (to save) and embedded metadata. We'll now go into each of these in a bit more detail.

File renaming

Most camera's have an internal numbering system that uniquely names each of the files, often with a combination of letters as well as numbers. For example DCS_1234.jpg In this case, the 1234 is unique, at least within 10000 images and as such is quite suitable for our needs. One point to note is that on many cameras there is an option to reset the numbering each time you format the cards; personally, I would suggest not to do this, you want the numbering to be as unique as possible. The problem with the filename of the camera is that it really doesn't "tell" you anything about the photo, so, the solution that I use is to add the time and date (the photo was taken) into the filename.

Let me digress for a minute here, if you are using the capture time and date as part of the filename, then this is actually something written in to the metadata of the image by the camera at the time the image is saved onto the card (by the camera). With this in mind you need to ensure that the time and date is correct on the camera before you use it. Two key points here are, if you're changing time zones, make sure that you change the time to keep it local otherwise the image time won't really make sense in relation to the photographs. Second (crucial) point is that if you use multiple cameras then make sure that they are all in sync, not roughly, but exactly, I do this to the second. The reasoning behind this is then once you pull the photographs off the cameras and rename them the photo's will appear in chronological order no matter which camera was used at the time.

Back to the main point, renaming the files. I would suggest to use the following format yyyymmdd_hhmmss_originalFilename

By this I mean setup the software to add the capture date and time, in reverse order to the filename that is on the camera. The logic behind this is in a number of parts. Firstly, the reverse format notation for the year (4 digits!) month and day, along with the hour minute and second means that you'll get every picture in the correct order no matter which camera was used (see above about sync'ing the camera internal clocks). I use an underscore between the date and the time to make it easier to read, nothing more. Then we have an underscore as a separator and finally the original filename.

This may seem as a bit of an overkill but with camera able to shoot muliple frames per second, the time and date alone will not be unique. Most transfer software applications get around this by appending numbers, (eg _01) on the end of the filename, but then your formatting isn't consistent which can later cause issues. It's much easier to just use the "unique" filename from the camera, and as long as you're not really unlucky (2 cameras, both on the same internal picture count, take 2 images in the same second....) it will not only be unique but also consistent with all other photographs.

The other advantage is, if you're asking clients / friends for which photos they like, they can often just provide you the last 4 digits (unique camera reference) to identify the image without having to worry about the full filename.

Advanced Renaming

It's worth mentioning that I actually slightly change the "original" photo names sometimes, based upon camera model, to make it not only consistent from the same camera, but consistent, in structure, across all cameras. By this I mean that I will always have the following format yyyymmdd_hhmmss_ABC_1234 as the filename (not including extension) where ABC refers to the camera either directly as saved or changed to fit in this format.

The reasoning behind this is we sometimes need to change, search, or update filenames in bulk. If we know that ALL our photographs are stored in exactly the same format then this is easy, if however, some, for example, didn't have the "_" between the ABC and the 1234 then this makes the format inconsistent and not so easy to deal with later. Moral of the story is to do things right up front, get consistency (you don't have to copy my strategy exactly, whatever works for you) and most importantly is to get the date and timestamp in reverse order at the front of the filename.

Download Locations

Nikon Transfer 2 isn't anything particularly special but does have a nice facility to extract the pictures and save them to 2 places at once. All of the other options, filename and metadata is the same on each image, but having the option to save to two locations provides your first level of instant backup.

My architecture is to save the first (main copy if you like) on my desktop machine that will be used for any processing later on. Like a lot of people I have the OS (Windows 7 in my case) on one drive in the computer, and then my data, including images is saved on another. In reality both of these, OS drive and data drive, are actually mirrored disks themselves in order to reduce the chance of losing data through disk crashes. So, that's 4 drives in total in use at the moment.

On my data drive I have a single directory for photographs (as we'll talk about later this helps with Adobe Lightroom sync'ing directories) and then within this single directory, lets call it "workflow" I have a directory named "0-RAW" This is the root directory for all my extracted photographs, even if they are not RAW images. It is named with a "0" at the front to ensure that directories are displayed in order (again later in Lightroom this helps).

Nikon Transfer 2 software has the facility, as I'm sure a lot of extract applications do, to create directories at the point of extraction. I use this facility, again with a yyyymmdd_hhmmss format to create a directory within the "0-RAW" dir. The result is then you have something like d:\workflow\0-RAW\20141231_134210 as a directory structure for any files extracted from the memory cards at 13:42 on 31st December 2014. Within this directory all the files will then be named in line with the above comments, ensuring that the filename represents the time the photos were actually taken.


Back to the idea of having 2 locations that can be written to at the same time; the above gives us a working copy (nice and fast as it's local to the software to be used). The second (backup) location, then in my case is on a network drive, a Synology Network Attached Storage (NAS) box to be precise. I'll cover details of my reasoning behind using Synology in a different article, but at this point it doesn't really matter as long as it's a safe location, ideally with RAID capability (set to provide some redundancy).

On this backup location, I have 2 different "root" directories, one for personal work and the other commercial such as my weddings. The reason for the split (and we'll talk about this later) is I use slightly different backup systems for personal images to those taken for clients; split things at this stage and we don't have to worry about it later. Irrespective of the directory used on the backup location, the remainder of the settings is exactly the same as the "main" copy written to the computer. A reverse notation date and time stamped directory that relates to the extract time, and then the images are all saved as above.

Nikon Transfer 2 does simplify this by having a "use the same settings as primary transfer" type checkbox, so all you need to do is select the backup location and everything else is the same. The result, you've now got 2 identical copies of every image taken, one to work with and the other as a backup to refer to if required.


Our final element is any metadata that you want to embed in the images as you extract them. Personally I only embed copyright data at this stage on any personal images, and my company information as well as copyright on any commercial work. Additionally on any commercial work I add a "category" keyword, so that could be Wedding for example. The key thing is to keep this category consistent throughout and I would suggest quite broad, so only have a max of 5 options I would suggest. The category is used to help later if you're looking for photo, so making it quite broad and consistent means you'll know instinctively what you would have called the photos you're searching for and although you'll have a lot to search through it helps (me at least)


We've taken the images from possibly multiple cameras, all that have the internal clocks sync'd and saved them along with basic metadata into two different locations. One local for our review and editing process, the second ideally on external storage, split depending on if this is personal or commercial work. In the transfer process we renamed the images to ensure we have all the photos in a chronological order and yet unique and consistent through keeping the original filename.

I hope this review of the first step in my file management workflow has been of interest. If you found this helpful it would be appreciated to show your support by "Liking" my Facebook page TonyHalfordPhotography do so easily by clicking on the Like button, top right.

All the best!

 Comments (click to expand)

Loading comments...

Add a comment (click to expand)