Landscape photography: To Photoshop or not to Photoshop?
Posted by Sam Harrison on January 22, 2015 at 15:12.
carnedds devils kitchen glyders idwal ogwen pen yr ole wen photography photoshop snowdonia
I got a polarising filter and remote control for my camera as a Christmas present, and so with the forecast set for good weather and with plenty of snow on the ground, I headed to Snowdonia on the Monday after Christmas to test them out. I was up in the hills both for sunrise and sunset, and it was genuinely one of the most enjoyable days out I've had. The lighting was stunning for most of the day and I managed to snap a fair few photos that I was proud of, one of them in particular of Llyn Idwal with Pen yr Ole Wen as the backdrop. I posted some of the photos on the usual array of social media sites and to my surprise, one of my tweets received a fair amount of attention:
Whilst 37 retweets and 43 favourites isn't exactly setting records, it's more credit than I've got for a photo before, and I was encouraged by a few replies from people saying how much they liked the photo (and even a chap called Gareth setting it as his desktop background; I feel honoured!). However, one of the replies wasn't so positive:
Wynn raises a valid point and one that I've argued with myself over for many years, and indeed his tweet made me revisit those arguments. Unsurprisingly, I came to the same conclusion, that a bit of gentle Photoshopping is a good thing (or at least, isn't a bad thing), but I wanted to explain this to Wynn and convince him that there is nothing "artificial" about my photograph, nor have there been any "additives". Unfortunately, the 140 characters Twitter allows wasn't nearly enough to get my point across, so I thought a blog post was the best bet. Hopefully, it will also convince some others that are thinking the same thing as Wynn but have kept their disapproval quiet.
You can definitely go over the top, as I said earlier all you have to do is Google "HDR photography" for some examples, but I think subtly using the technique to make the scene closer to what you actually saw is fine. Which brings me onto my first rule above; "the final image must be a true representation of the scene I saw". What do you actually see?
Why? Because a camera takes a snapshot; letting a certain amount of light in and composing the image based on this information. But when we look at a scene, our eyes scan around, constantly adjusting to differing lighting levels and taking the equivalent of thousands of photos, all compiled seemlessly by our brain into the image that we "see", indeed, the lasting memory that we take away. There's plenty of scientific research to back this up, summarised nicely by the introduction to a 2011 paper entitled "Eye Movements Help Link Different Views in Scene-Selective Cortex" [1]:
[1] Julie D. Golomb, Alice R. Albrecht, Soojin Park and Marvin M. Chun, "Eye Movements Help Link Different Views in Scene-Selective Cortex", Cerebral Cortex 21, 2094 (2011).
Another one from yesterday, of Llyn #Ogwen and Pen yr Ole Wen from Devil's Kitchen. #Snowdonia #snow #Wales pic.twitter.com/X0wG3Fr0DQ
— Sam Harrison (@samharrison7) December 31, 2014
Whilst 37 retweets and 43 favourites isn't exactly setting records, it's more credit than I've got for a photo before, and I was encouraged by a few replies from people saying how much they liked the photo (and even a chap called Gareth setting it as his desktop background; I feel honoured!). However, one of the replies wasn't so positive:
@samharrison7 @kate_llyn photoshoped again, the landscape has its own beauty there is no need for artificial additives
— Wynn Davies (@Torgoch) January 5, 2015
Wynn raises a valid point and one that I've argued with myself over for many years, and indeed his tweet made me revisit those arguments. Unsurprisingly, I came to the same conclusion, that a bit of gentle Photoshopping is a good thing (or at least, isn't a bad thing), but I wanted to explain this to Wynn and convince him that there is nothing "artificial" about my photograph, nor have there been any "additives". Unfortunately, the 140 characters Twitter allows wasn't nearly enough to get my point across, so I thought a blog post was the best bet. Hopefully, it will also convince some others that are thinking the same thing as Wynn but have kept their disapproval quiet.
My Rules
The scope of what it means to "Photoshop" an image is very broad, from simply tweaking the brightness/contrast, to altering colours, to removing or adding entire elements of a photograph. When I talk about myself Photoshopping images, I'm not talking about painting the sky orange or removing stray power lines, but simply enhancing what is already there, and to do so I stick to a few simple rules:- The final image must be a true representation of the scene I saw. Adjusting contrast, bringing out highlights, lightening shadows and removing colour casts is all okay because that's exactly what your eyes do when they look at a scene anyway (more on that later).
- No painting colours. Sure, I'll enhance colours and bring out the elements I think were most striking on the day, but all the colour information I need to make the image true to life should already have been captured by my camera (see the bit later on RAW files). Artificially painting a photo different colours is a big no.
- Cropping and straightening is okay. Just the same as I would if I stuck the photos in a physical photo album, I think it's perfectly okay to crop images to make them look their best (we all know that magic "rule of thirds"!).
- No adding or removing entire elements. Don't like a particular tree where it is, or got a lamp post I wish wasn't there? Tough, that's the scene and I'd be going against my first rule if I did anything about it! The only exception I'll make is if, for example, a small fly decides to pass by the lens just as I take the shot.
- Blending multiple images is okay. Taking a few different photos to capture the dynamicity of a scene and then blending them is fine, after all (as I hinted earlier and which I'll talk more about later) that's exactly what your eyes do. But be careful not to go over the top; stick to rule number one! If you want some examples of images with far too high a dynamic range, just Google Image search "HDR photography"!
Raw images files; "digital negatives"
Most (if not all) digital SLR cameras these days give you the option of saving your photo as a "raw" image file. These files are often referred to as "digital negatives", because in the same way that you have to process a negative to get a photographic print, you must process a raw image file to turn it into a viewable format (such as a JPEG). They contain all the information captured by your camera's sensor, which is much more information than the final print could ever show, information such as light intensities and colour balances. In the same way that processing negatives in different ways will result in a different image, some more desirable than others, the digital processing of raw files enables one to select the most desirable balance of lighting and colours available from the camera's sensor. If I wanted, I could optimise these settings on my camera and get it to save the files as JPEGs, but I don't see how choosing to do this at a later date in the comfort of your own home, instead of out in the mountains when it's blowing a hooly and you're beginning to get frost nip, is cheating at all.Blending images
"High dynamic range" photography is all the rage at the moment, and it basically means taking multiple exposures (some lighter, some darker) of exactly the same view and blending them together to get a high "dynamic range"; that is, the ratio of light to dark in the photo. There's a good introduction to it over at Lifehacker. In its simplest form, this could be taking a "normal" exposure to capture the foreground of a scene, and a darker exposure that makes the foreground seem underexposed but bring out the sky nicely (which was overexposed in the normal exposure), and then blending them together. Smartphones like the iPhone and Android devices have been doing this for years, possibly without you even noticing.You can definitely go over the top, as I said earlier all you have to do is Google "HDR photography" for some examples, but I think subtly using the technique to make the scene closer to what you actually saw is fine. Which brings me onto my first rule above; "the final image must be a true representation of the scene I saw". What do you actually see?
How your eyes view a scene
Physically taking a photo is as artificial as altering that photo later on. Even the best sensors can't hope to capture as much information as your eyes do, and it is for this fundamental reason more than any that I think that post-processing photos to get them as true to life as possible is completely justified. Sure, sensors are calibrated to default to taking life-like photos (and generally speaking the more expensive the camera the better they are at doing this), but they'll never see what you do.Why? Because a camera takes a snapshot; letting a certain amount of light in and composing the image based on this information. But when we look at a scene, our eyes scan around, constantly adjusting to differing lighting levels and taking the equivalent of thousands of photos, all compiled seemlessly by our brain into the image that we "see", indeed, the lasting memory that we take away. There's plenty of scientific research to back this up, summarised nicely by the introduction to a 2011 paper entitled "Eye Movements Help Link Different Views in Scene-Selective Cortex" [1]:
No modern-day camera can even come close to replicating what the eyes see, so in my opinion a little help from Photoshop in getting the final image as true to life as possible is fine.In order to construct stable images of the world, we must
integrate visual information over time. Although researchers
debate whether visual information is literally integrated over
time in the mind (Irwin 1991; O’Regan 1992; Henderson 1997;
Cavanagh et al. 2010), no one doubts the challenge posed by eye
movements, namely the rapid changes in visual input received
by the retina and transmitted to the brain. For instance, our
perception of a large visual scene is comprised of a series of
snapshot views; when we move our eyes to actively explore the
scene, even small eye movements can make the local foveated
elements change completely, yet from these discrete successive
views, we perceive a smoothly continuous visual experience.
Art
There's a point where photography and art overlap. There are some beautiful photographs that I've seen that have been post-processed so that they definitely aren't a true representation of the original scene, but are nonetheless still beautiful. Indeed, a lot of landscape photography competitions are won by photos that unashamedly fit into this category. And whilst I don't tend towards this approach, at the end of the day, what is wrong with turning your photo into a masterpiece of art, as long as it pleases the eye to look at? Though I think some line between what is a piece of artwork and what is a photograph is needed, as after all when I'm looking at photographs, I'm doing so to see a real-life scene.The photo in question
So that all begs the question, what did I "do" to that initial photo that sparked this blog post?- It was a single exposure (taken with a Nikon D40 and a Nikon 18-55mm lens, with a Hoya HD circular polariser).
- I tweaked certain values from the raw image file (i.e. what the camera captured) to optimise the foreground (Llyn Idwal and Pen yr Ole Wen).
- I then copied the photo and re-tweaked it to optimise the sky (basically, making it darker).
- Finally, I blended these two versions of the same image together to give a dramatic sky with colourful highlights where the sun is shining through, whilst keeping Llyn Idwal suitably exposed so you can see the fantastic colours that were reflected in it. I'm happy that the final is true to what I saw on the day.
[1] Julie D. Golomb, Alice R. Albrecht, Soojin Park and Marvin M. Chun, "Eye Movements Help Link Different Views in Scene-Selective Cortex", Cerebral Cortex 21, 2094 (2011).