Photo Editing

When it comes to post shoot processing of photos I have reservation in relation to the amount of post processing applied.

In my opinion you have either photos that server as documentation for a point of view in point of time or you have images – an artistic work based on photos.

It is hard to draw a precise line between the 2 concepts, because some tools are not considered to disturb the documentation purity when applied a little or moderate, but when applied more intense it disturbed the validity of the documentation. E.g. if I take a photo in dark sorrounding (using high ISO, large aperture and slow exposure) I can get a photo that looks somewhat lighter that what was actual perceived by those attending the scene. If I at home uses my post process to increase the light massively and get to an image that resembles daylight the validity of the documentation is disturbed.

Other tools is more obvious e.g cloning tool. I take a photo of a city landscape with beautifull buildings and ugly building cranes. At home I use the clone tool to remove the cranes. In my perspective we are no longer talking about a photo – with documentary validity – but about a work of art (an image).

I have chosen to use RAW, instead of JPEG, because the RAW file will provide 12 bit (my camera can deliver 14 bit) color space which gives a larger dynamic room for the picture. When exporting to JPEG it will become 8 bit, but now I have a better opportunity to control what part of the dynamic room that will be removed.

RAW files (*.NEF as I use Nikon) requires post processing in order to provide the finished product. Nikon did provide a tool, but it will only run on Mac or Windows, not Unix, Linux or BSD. As all my computers run Linux I cannot use the Nikon tool, but must use what is available on the open source platform.

You are not left in the dark on a Linux desktop when it comes to RAW photo editing. Darktable, UFRAW, Rawtherapee, DigiKam, LightZone, Shootwell are examples of available applications.

For my work-flow I have chosen Darktable. Darktable is not alone a conversion and editing solution, it’s also an archive for the RAW photos.

Prior to choosing Darktable I performed some tests. My primary criteria’s was:
– quality of finished product
– ease of work
– speed of conversion

When I first tested Darktable I actually decided not to use it because it did not handle noise in a pleasant way, all the output did have more noise than what I got from camera based JPEGs. But the situation  changed dramatically in later versions. Especially after the introduction of profile based noise reduction, where the application includes a profile for each camera and uses it for eliminating noise. To apply the right profile is just a click, but the amount of noise reduction can be adjusted.

A nice feature in Darktable is that data is stored both in a SQLite database and in sidecar files. So if you crash the database you can recreate it by importing the photos and their sidecar files – remember to check the tick mark in settings – or you can take the photo and it’s edit to another computer.

When I arrive home from a photo shoot the first thing I’ll do is to get rid of the photos that is obviously not good and that I don’t think can be saved through editing. Then I apply post processing to the rest. Typically I will apply the following to a photo
– lense correction
– correct exposure
– noise reduction
– sharpness
– align horizon
– crop

For some photos I will work more on high- and low lights in order to dampen over exposed areas or to enhance under exposed area, while avoiding burn out in other areas.

What I do is mostly comparable with the works provided in the traditional dark room

For rare occasions I do not stop at this level. I have used the cloning and spot removal tool to wipe a disturbing building crane from the scene.

For creation of different sizes – for e.g. web publishing – I’ll go back to the photo in Darktable. Do eventual new cropping and perform an export in the relevant resolution.

For years I used JPEGs as my primary format. When I need to edit these images I cannot use Darktable, but revert to other tools.

To perform a quick resize of pictures I’ll run them through ImageMagick – which is available on Linux, Windows and Mac – with a script that applies the needed size, quality, naming etc. This is a very efficient way to handle a large number of pictures with little effort.

When the pictures arrives at the web platform (Coppermine or WordPress with NextGen Gallery) creation of thumbnails, rotation, annotation are handled by PHP GD Libary. The GD Libary can also be used locally with a local installation of PHP, by that you can build a function to fit almost whatever custom function your script needs.

When I just need to edit a single picture or if I should chose to manipulate an image further, I’ll go to the GIMP. The GIMP can be found to Linux, Windows and Mac. It delivers approximately the same functions as e.g. Adobe Photoshop Elements and on top it has a build in scripting engine, that allows you to build work-flows or automated functions.