When Spectrums Collide: Selective Processing With Infra Red
by Lee Mandrell
A Simple Infrared Selective Color Project
As they so often are, it was another perfect day in the Smoky Mountains. I’m an avid color landscape shooter, but I am always on the lookout for infrared shots as well as anything I think might separate my shots from the norm. I happened upon this scene at the end of ‘The Roaring Fork Motor Trail’, just at the edge of Gatlinburg, TN. My wife and I asked the shop owner if we could shoot the old dodge truck that resides on the property. She informed us that we could take pictures, but we had to stay on the outside of the fence, and also to let her know if we felt we got anything worthwhile. To me this meant shooting what has been shot thousands of times before me.
Oh well, you accept the conditions you’re given and make the most of it and be happy about it, right? You find ways to make the location, light and weather work in your favor, right? So I took this as a challenge and went out to see what I could come up with. For the sake of this article however, I will leave out all the scientific details as they relate to the visible and invisible spectrums.
As I was taking my round of color photos, the thought crossed my mind, “Wow, this could be a great selective color shot, and even greater if it was an infrared selective color, but how in the world could I pull that off?” Color infrared conversions have never looked that good to me, and no way to get it from just the infrared shot. The wheels began turning. Had this technique been done before by anyone else? Maybe, but I was really unsure, and knew I was going to have to give it a try. This wasn’t going to be a channel swap photo produced from a single infrared file by any means. I made the decision to try and produce this photo not only from two different shots, but from two different cameras as well. Crazy? Maybe, but I like to push myself. I shot the color photo with a Minolta Maxxum 7D (I have since switched to a Sony A77), and the infrared with a Canon G5 dedicated infrared camera. I figured, I have nothing to lose and at the very least, I would have a pretty cool infrared shot even if I came away with nothing else.
I set up for the color shot first. Having taken a few shots and looking over them, I decided on what would be my final composition for the infrared. I snapped a series of 3 brackets. I then handheld the infrared camera directly on top of the tripod mounted color camera and also shot a series of 3 brackets and hoped that luck would be on my side. Nothing complicated for sure. The nice thing about a converted infrared camera is that you can hand hold them for the exposures with no problem. Now that my shots are taken, it all goes into the digital darkroom where the real magic happens. Once into post editing, I work all of my files in Adobe Camera RAW. For color shots, I tend to work most of them generally the same way.
I add the usual contrast, shadows / highlights adjustments, color temperature, a touch of vibrance and clarity while always keeping a close eye on the levels. I make every effort not to over work or over saturate my shots. I use these settings more or less for almost every shot I work on, but each shot does vary. In this instance though, my focus and concern is only with the truck itself, so this photo is an exception to my normal work flow. The rest will be weeded out in the final composite.
I also find that many times I have to tone down the greens or desaturate them right out of the camera. Even Fuji Velvia film was never so vibrant! Otherwise, they look over saturated and unrealistic.
I then open the file into Photoshop. Now, I come back around and grab the infrared version. Again, I work it in ACR, much the same way as the color file and I’m still keeping a close eye on the levels.
Once both are opened in Photoshop, I make a copy of the color shot and paste it onto the infrared version. I reduce the opacity of the ‘color’ layer so that I can see how far off the alignment is and what all I will need to weed out and delete.
To my absolute delight, the overall alignment was actually very close. Not perfect, but close. This would be very doable with a minimal amount of work digitally speaking.
From here I scaled the image until I felt it was “close enough” to work without making myself crazy or over complicating the editing process. I used the marquee tool to make quick work of deleting large chunks of unwanted image.
Once I am satisfied with what has been taken away, now comes the fun part. I make a duplicate layer and turn it off in the event I make a mistake during the next phase. This saves time should I feel I need to start over. You could also rely on the history palette, but my work flow is to always create duplicate layers as a safety net. It has saved me many headaches in the past. To do good retouch work or compositing, it’s always best to be zoomed in on your subject.
I set my lasso feather to 2 pixels for this type of work.
This gives me a nice blended edge without being either to hard or too soft looking and maintaining a high degree of a realistic, seamless look. It also saves from having to use the blur tool to blend your image. In my opinion, this has a more realistic look to the finished piece. I work my way around in sections, deleting as I go. You could also use the eraser tool, but my feeling is that I can be more precise with the lasso. It’s merely a personal preference is all. Use whatever works best for you, whatever you are most comfortable with. Once done deleting, I set the opacity back to 100% in the ‘color’ layer, then turn off the infrared layer. In doing this, I can see any color pieces I may have missed while deleting and I take steps to get rid of them at this point. If you don’t perform this step, you can possibly have stray artifacts show up in a printed piece. It’s best to be sure you wont have any surprises on an expensive print, particularly if a client is the one purchasing it.
I turn the infrared layer back on, assess the image as a whole, save a psd version in case I should ever need or want to do something down the road, then also save a high resolution jpeg of the composited image for web posting. If I decide I want to make a print, I will come back to the psd and save it as a tiff at that time. I save disc space by waiting to create the tiff only when it’s needed. Again, this is a personal choice.
As you can see, there are no real secrets and no real tricks here and something anyone could do. This can be done with basically any type of file (even one you may have converted), and even a minimal skill set if you have barely used Photoshop, Elements, Lightroom, or even Gimp for that matter. The results are well worth the effort and it doesn’t take that much time. I spent less than 30 minutes on this one. The real trick is all in stepping outside your comfort zone and trying. You just never know what you will achieve and what great photo you will end up with. It turned out to be a simple technique. Needless to say, I did email the woman a link to the final composite, thanked her for letting us shoot there and she was delighted with the results! Enjoy!
To see More of Lee’s work, visit his website.