Digital Image Post-Production Techniques #1

Image enhancement in the digital darkroom using Adobe CS4 and others - an introduction.   

This blog is #1 in a series that will explore some of the tips, tricks, techniques and skills in the "black art" of digital image post-production.  The term comes from the movie industry but also applies to music and other creative arts; it refers to the work that's done to an image after shooting but before publishing. The output from post-production is a finished image ready for publishing. The basic file for post-production work is the RAW image, or most usually, a tonally-balanced, colour-balanced, spotted and cleaned "raw" image, typically converted to a working format such as PSD or TIFF.

    A word or five about image formats. A RAW file is the actual output from the camera. JPEG is not a raw image, even although cameras will produce JPEG straight from the card. A sports pro shooting for press would produce JPEG for speed and upload.  Wedding, fashion, landscape and other pros usually shoot Camera Raw files and then enhance in post-production to get to their desired output. JPEG can be the output from post-production, but GIF and PNG can also be created if the intent is a web site. A RAW file contains the greatest scope for creativity - technically, its called dynamic range - so for pros working to a creative intent its the only choice as a start point.

   A final word on image formats and a personal one.   I shoot RAW but I convert to Digital Negative (DNG) on upload to the computer. I've done this now since I opted to use Adobe Lightroom as my Local Digital Asset Manager (LDAM) and image pre-processor about 4 - 5 years ago. Adobe have published DNG as an open format and any company is free to use it. A couple of camera manufacturers now produce DNG as their "RAW" file, but the mainstream manufacturers, Canon, Nikon and others don't, despite Adobe's pleadings. A key point for me in making that decision is DNG's absence of side-car files. Instead of using a separate side-car file (and all the hassle of making sure you've got them where they should be) all of the image content and metadata is held in the DNG format, so normal computing operations like copying and moving don't get more complicated than they should be. This is a personal opinion but it should be the default format for all serious digital photographers. If you're still using proprietry RAW then change over to DNG soon.

   Phew! I'm glad that's over.   For post-production work I use Adobe's CS4 suite and Adobe Lightroom 2.6 (soon to be LR3 which I'll upgrade to on release). CS5 is now available as I write this blog, but it'll be a few months yet before I decided to upgrade to it.  CS4 includes Photoshop Extended, Acrobat Pro and  Flash pro, all of which I use, but not every time. Lightroom (LR) is my Local Asset Manager and also my image pre-processor. LR allows a substantial number of adjustments to be made to images without resorting to use Photoshop and, for most basic toning and balancing adjustments, its perfectly fine. Within LR I also use ImageMagik and a number of plugins, particularly LR/Mogrify (thanks to Tim Armes, at Photographers Toolbox). LR is a client-based DAM and its no use if you have a web presence and sell or distribute images using the web (well not much use in its LR2 guise, LR3 holds promise though). I also use Zenfolio as my internet Digital Asset Manager (iDAM) but LR has some nice plugins to enable me to upload output images straight to Zenfolio (thanks to Jeffrey Friedl, at Jeffrey Friedl's Lightroom Goodies). I also use Slideshow Pro Director  to serve my website portfolio galleries into Flash-based components embedded in my Blog pages and as Flash controls for visitor viewing. Jeffrey's plugins come in handy for that too... so, LR can function as my local, client-based DAM and I can push out images to a number of other specialist iDAM's as and when I create them, cutting down significantly on the work-flow effort. Jeffrey's tools also allow me to upload straight out of LR to my Flickr and Facebook sites.

   And now for Photoshop! In PS I use Totally Rad's TR1 Action Mix along with Guy Gowan's Retouch Workflow and Cosmetic Techniques as well as some of my own, home-brewed Actions and techniques.

So, if you were thinking of setting up a business in photography for yourself, or doing this at home so you can shoot your daughter's wedding, go right ahead. Oh, and you'll need quite a bit of photography experience, the same amount of digital knowledge, not to mention some serious IT kit and a deep understanding of web-sites, Flash, the internet and so on. If you're now still saying "Bring it on..." then dive in and enjoy!

   When I shot it I exposed for the highlights in Aisha's dress. I'm familiar with the location and know how the light plays at different times of the day on the stone-work and the architecture. The stone responds well to a whole range of processing techniques and features show up that surprise me every time I shoot here. It's natural stone, so it's weathered and cut as you'd expect after centuries of use and abuse. The light is from the north so it's soft, consistent and diffused by the buildings that surround the cloister. Aisha posed as directed and I'd taken up a slightly angled shooting position to create some depth in the image by using the diagonals of the architecture.  So, I had some idea what I was looking for, but needed to give myself some "wiggle room" in the image to play with. 

Tip:    You can crop away what's there but you can't add back what isn't there. 

I now had a vision of the image I wanted to create. I saw her glowing in the left-side but with interest in the walls visible but fading away from her up and right. She's beautifully lit and the shadows behind her create the depth that make her the centre of the shot even although she's not in centre of the image. Her pose is statuesque but she's flesh not stone, so she eclipses all that surrounds her and the walls radiate light as a subjection of her beauty over their permanence. OK, so how to create that look??!

   The intermediate image now shows all of the elemental features of the final image, but without the WOW! Enter on cue, from stage-left... Photoshop. 

   I have to use PS for the next bit because LR doesn't have the same feature-rich action sets that PS has. LR allows the image to be exported into PS and will manage it once it comes back from PS. The tone and levels were pretty much as I wanted them but I wanted to emphasise Aisha and not have the image look as flat. Totally Rad's actions tend to re-balance the tone so its important to know what's going to happen to the image tone after the action has run. I used two actions in a recipe;  Super Fun Happy and Troy and tweaked the tone a bit to counter some of the unwanted (for this image) effects of the actions. SFH desaturates the red channel by default and this gave Aisha a pasty look, but it was close to what I expected. Open up the SFH layers and switch the desat-red layer off. Result! Well, nearly.

   The colour balance still wasn't quite right and I wanted to take the remaining colour out of the wall, but keep Aisha in colour so it was back to LR. I could have made the adjustments in PS but LR is easier to change if I can't quite get the effect I'm looking for. Also, I wanted to vignette the image and LR's vignette tool is much better than PS. In LR I adjusted tone to place Aisha in the focus point and adjusted the red, orange and yellow colour saturation and luminence values to get what I wanted. Finally, it was a simple case of vignetting and tweaking the values to get "the look". Job done.

A bit of a marathon explanation... it was quicker to do it than describe it... but the effect on the image is substantial. I'd say this would be near impossible to achieve in film and without a computer, so digital has allowed us to create new and exciting features. Check back for more next time.

See ya!

Brian, 14 April 2010