Tutorial: Exposure Bracketing and Photo Blending

One of the limitations of the digital camera vis a vis the human eye is it's inferior ability to discriminate between dark and light; the average human eye can distinguish approximately 2^20 - or approximately a 1,000,000x difference between darkest and lightest points. My high-end Olympus camera can only discriminate a 7,332x difference - 140-fold less. You've probably noticed when you take photos in high contrast light conditions such as a sunset or any scene where you have very bright parts and strong shadows that you either have a well exposed sky with extremely dark land or a well exposed land with a very pale sky. This lower discrimination power is why. But there are ways for overcoming this. In the days of analogue cameras, photographers would place a sheet of glass in front of the lens that was half smoky, half clear and use the smoky half to reduce the light in the sky. These days, most photographers achieve the same effect by taking multiple photos at different exposures (exposure bracketing) and then assemble them to a single image using software (photo blending). 

This is the final edit of one of my photos of the delightful Dolomitenhütte in the Lienzer Dolomiten that featured in my recent post Osttirol Revisited -  Part II.

The Dolomitenhütte at Sunset || Olympus 50 mm, f/5.6, 1/50 s, ISO 200

Exposure Bracketing

Some cameras, such as my Olympus OMD E-M1 Mk II, allow you to automatically take multiple exposures with varying levels of exposure compensation. One click of the button and the camera will rapidly take a series of three photos, one at the default setting, one with ¼ of the light ("2 stops under") and one with 4x the light ("2 stops over"; a "stop" in photographic terms is always half or double the light/exposure). If your camera doesn't offer this facility, you can almost certainly use exposure compensation (usually a dial labelled E.V. with a scale running from -3 to +3 in ⅓ steps). This is a magical dial that every photographer should be familiar with and using all the time, especially if you're shooting through an electronic viewfinder. Normally you need to do this with a tripod in order to be able to overlay the images 100%.

Exposure compensation scale (orange)

Photo Stacking/Blending

Most modern photo editing software allows you to compile multiple photos as layers and choose what parts of each layer should be visible in the final image. One prerequisite for the process is that the images be essentially compositionally identical, otherwise matching them up can be a pain. Like I say, normally this is assured by using a tripod to fix the camera in place, composing the photo carefully and then manually setting the focus. Normally. Unfortunately I tend to get a bit sloppy with these types of photo and shoot by hand and using auto-focus. 

The real art lies in blending the photos; defining which part of which photo is in the final image. Again, most modern programs offer the option of automated HDR image creation (HDR = high dynamic range, the art of compressing more levels of light into the final image than is possible with a single photo). Alternatively, the process can be done manually by opening multiple photos in layers. I prefer the latter process as it gives me a great deal more control over the final image. Here, the cleaner the edge defining the various layers the better, but invariably there are trees in the photo at the borders of the two layers and these are always a challenge to process. I've chosen a pretty straightforward image here that I manually masked for blending. I'll take you through the process as I do it in ON1 Photo Raw 2020, my photo editor of choice.

We'll start with the basic image, where you can see that the foreground is ok, if a bit dark, but the sky has no character whatsoever, a poor reflection of what we were seeing at the time. 

Native exposure - Olympus f/5.6, 1/15 s, ISO 200

The auto-bracketed shots are 2 stops under- and 2 stops over-exposed. In this example I didn't actually use the over-exposed image as it didn't fit the mood of the scene. As you can see from the under-exposed image, there's a lot more interesting stuff going on in the sky, but if you compare it to the final image, I've taken it even further. 

2 Stops Under Exposed - 1/60 s

2 Stops Over Exposed - 1/4 s

In the ON1 Photo Raw Browser module I select both images and then on the right click on the icon [Layers]. This generates a new image file in the Edit module where both images are simultaneously open. To start with you'll only see the top image, however, so don't be disconcerted. Any time you're working on the image from now on, you'll have to pay attention to which layer you're currently working on and you can click on them at the top of the right hand menu bar. The active layer has a turquoise highlight:

How the layers are displayed in ON1 Photo Raw, the blue circle indicates that the layer is selected (visible)

Masking

So how do you define which bit of which photo appears in the final image? Through an arcane process called masking. Masking defines which parts of a layer are visible (white in our case) and which parts of a layer are invisible (black). But the mask isn't limited to just black and just white, it can also be shades of grey, allowing nuances of layers.

So who tells the program which bits of each layer I want in the final image? Well, you do using masking brushes and gradients. You paint in (or out) the bits of the layers that you want. Clicking on the icon to the right of the layer thumbnail opens a panel giving you access to the masking tools. When you first open it, this panel will be completely white, indicating that all of this layer is visible.

Accessing the various Masking options

There are multiple ways to create a mask from the simple gradient tools, the brush as well as luminosity and colour range masks. Looking at the layer mask that I created for this image, I used the Masking Brush combined with the Perfect Brush option to get a hard edge on the hut itself, together with a slightly softer brush towards the bottom right corner:

Mask for the final image; the Native exposure image is on top, this black and white mask defines which parts of that image are visible over the darker underexposed layer

Ok, so apply that, and what do we get? Certainly a lot more appealing than either of the originals, but still a long way from the final image:

First Blend

So what else do we need to do to get from this to the final? There are a number of steps that I applied to each of the layers in order to really bring this image to life. In ON1 Photo Raw it generally doesn't matter what order you apply the steps in (there are exceptions, but they are too complex for this post. OK, I admit it. I don't know what they are, I just remember watching a YouTube video once where a bloke tried adding the effects in different orders and got different results. There. Happy now?). I don't generally adjust too much in the develop tab at the moment, concentrating almost exclusively on local adjustments

Background Layer (sky and left-hand foreground):

1. Cropped the image to give a more pleasing cut
2. Globally added the Effect Filter: Color Enhancer - Increase Color
3. Locally added a -1 EV Darken adjustment to the sky (gradient mask)
4. Removed the stray branch top left using the Retouch tool

Foreground Layer (cabin):

1. Locally added a +1 EV Lighten adjustment with +20 to the shadows and +30 Color Temperature to the area around the cabin lights

These I combined to form a New Stamped Layer (a new layer combining all of the visible parts of the image; right-click in one of the layers and choose this option).

Stamped Layer:

1. Globally added the Effect Filter: Vignette - Big Softy 50%
2. Globally added the Effect Filter: Color Enhancer - Warmer 50%
3. Locally added the Effect Filter: Tone Enhancer - Clarity on the cabin
4. Locally added the Effect Filter: Tone Enhancer - Tonal Contrast to the forest bottom left
5. Locally added a Vibrance adjustment to the sky
6. Other minor dodge and burn (local brightening and darkening) adjustments

And that was it. The first Color Enhancer added a touch of colour to the sky, amplified by the Darken adjustment. The Lighten adjustment with the colour warming added a pleasing glow to the lights in the cabin, even spilling over to the rocks outside. The Vignette focuses the viewer's vision on the central part of the image - I do this to almost all my images - it's part of my personal preset. The Clarity enhancer on the cabin brings out the texture of the wooden slats of the cabin and the Tonal Contrast enhancer brings a bit of definition to the dark forest bottom left - making it a bit less 'mushy'.

And that's it. A lot more editing that I'd normally employ, and certainly a lot more 'dodging and burning', but I think it's worth the effort. End-of-day photos are always a little more challenging due to the lighting conditions. If you have any thoughts or questions on the editing process, do leave them below.

- Dolomitenhütte at Sunset -

In Seach of the Milky Way - From Dusk 'Til Dawn in the Allgäu

At the end of May I scouted a possible location for an astro shoot in the Allgäu mountains come next new moon. A good milky way photos lives and dies by the foreground and I wanted somewhere with a good view over the alps. The middle of June arrived and the forecast for Friday and Saturday was awful; clouds, rain, the lot. Nothing doing. The next weather window was going to be Tuesday night. To make matters worse, the initiator of the whole idea - my friend Matthias - couldn't make it as he was stuck up at work in northern Bavaria. But an old school friend who happens to live in Munich was interested enough to bite when I announced my intentions on Facebook and so Martin and I ended up heading down to Fischen im Allgäu early Tuesday evening to see what we could see. Meteorologically we were on a rising tide - there were still some clouds around but the weather was definitely clearing from the north and so we were optimistic.

The Shoot

Waiting For The Moon To Set || Huawei P30 Pro

Dusk

Driving down there were still cloud remnants lurking in all the wrong places, but the northerly wind was driving them deeper into the alps and so we weren't too concerned. We set off from the car-park in the early evening light. Normally, like most people, I spend most of my time in the mountains during the day. Unless I'm staying in one of the mountain huts, I'm generally back down in the valley for dinner. This evening's experience may get me to revisit that schedule. There's a good reason why landscape photographers choose to shoot at the ends of the day; the light quality is very different. It's softer, gentler and less blue. Watching the pastel colours changing on the Vorarlberg mountains in the gathering dusk made me start rethinking how I plan my days in the Alps.

As the Sun Goes Down || Olympus f/7.1, 1/80 s, ISO 200

It was great to see that not all of the snow had disappeared in the intervening weeks; I find the last snowfields really make a photo like this and I don't shy back from exaggerating them in post-processing like  here (above).

The Pastel Colours of Evening || Olympus f/7.1, 1/160 s, ISO 200

All the Layers || Olympus f/7.1, 1/30 s, ISO 200

Moonset

I'd promised my school chum a relatively easy hike up to the first mountain top that I'd identified. Unfortunately we didn't make it all the way and ended up about 100 m below the summit on a slight shoulder. The location was optimal for the first spectacle of the evening - the fresh new moon was due to set behind the peak above us and our location appeared to be optimal for that. My smartphone app (Sky View) was showing me where the moon should be in the sky, but for the life of me I couldn't find it. I was beginning to fret that either (a) the app was playing up, or (b) there wasn't going to be enough moon to see. As the sky darkened and I still couldn't see it Martin spotted it - why we hadn't seen it before was a complete mystery as it was so clear at this point.

Well It's Obvious Now! || Olympus f/8, 1/5 s, ISO 200

Setting Moon || Olympus 100 mm f/5.6, 0.6 s, ISO 400

Before we turned our attention to the stars that were beginning to appear in twos and threes there was one more dusk photo to grab before the technical stuff began:

When Purple Comes Out To Play || Olympus f/5.6, 40 s, ISO 200

As fun as it was to shoot the setting moon, it cost me my first milky way shots as I wasn't set up to get a good foreground photo during the blue hour that I could use to blend into the final image in post. The shot I ended up getting was ok, probably one of my better photos of the Milky Way per se, but the foreground is lacklustre. I don't know whether it would have been any better 100 m higher up either, but the lesson is that you need to be in position early, you need to get your foreground shots in in good time and you can only recon on about two good astro shots per night, one after the sun goes down and one before it rises, unless you're shooting with multiple cameras.

Collision Imminent || Olympus 300 mm f/5.6, 0.6 s, ISO 1600, Composite Image

Milky Way

The challenge with astrophography is the low light levels. You need an open aperture and relatively high ISO with a wide angle lens to get enough light to the sensor before the earth's rotation turns the star points into lines - the so-called star trails. As a rule of thumb, the maximum exposure time before this happens is 500/focal length* (400 if you're being conservative). So for my 16 mm equivalent wide angle lens, I can shoot for 25 s. In order to further increase light-sensitivity, we need to increase the ISO. Small sensor cameras (I shoot with an Olympus micro four-thirds crop sensor, technically a small sensor camera) are notorious for being sensitive to high ISO, the images tend to have a lot of unsightly noise - a random phenomenon that causes an unsightly buzz in the image. But being random, there's a solution: If you take multiple shots and then take the average of those shots, the noise is diminished. The challenge is that the software performing the calculations needs to take into account the fact that the stars are moving - if it didn't you'd end up with the trails again. All in all quite a technical feat, especially when there's a foreground involved. Fortunately the free Windows software Sequator can do all of this - it's what it was written for. It's not the most user-friendly app, but if you know what you're doing (or in my case if you know someone who knows what they're doing), it's extremely powerful. The trick seems to be not using too many of the features, but more of that below.

*25 mm equivalent, so with my micro four-thirds sensor I need to double my focal length

First Starlight - Stacked but Unprocessed || Olympus 8 mm, f/2.8, 25 s, ISO 6400, Stacked

We could just about make out the trail of the Milky Way in the Allgäu sky. From our vantage point I tried several series of shots, with and without my Haida Clear Light filter and Tiffen Fog filter and with the new 7Artisans 7.5 mm fish-eye. In the end I liked the unfiltered images through the Panasonic Leica 8-18 mm the best. The above image is stacked to reduce noise but otherwise unprocessed. The sheer clarity of the image just blows me away. The detail in the Milky Way is amazing. But I don't think this is a good Milky Way photo. It's a good photo of the Milky Way, but like I said earlier, a good Milky Way photo stands or falls by the foreground and I wasn't satisfied with what I was getting here, so after a cup of tea  and a pasty (thanks Martin, those were life-savers) I lay down for a 45 min shut-eye before heading up to the ridge, where I knew just the spot.

With Haido Clear Sky Filter, Unprocessed

7Artisans 7.5 mm Fish-eye Lens, Unprocessed

One of the greatest challenges with astro photography is finding infinity on the focus dial. Looking through the viewfinder you're confronted with an almost completely black image - finding something to focus on can be extremely difficult. If there's a bright light a couple of miles off this works relatively well, as at this distance there's effectively no focal change between 2 miles and infinity. It was quite a challenge in the dark though, and I had to hunt through the magnified viewfinder on the Olympus to locate Jupiter (bright spot in the middle of the images above) and twiddle the focus ring of the lens to render the tiny fuzzy light as tight and small as possible. Then it's shoot and hope. You can examine the images on the camera display, but you're never entirely sure whether your photo is sharp or not until you get the image on the computer screen.

Despite being pitch dark, it was easy enough to find my way by the light of my head-torch, having scouted the area less than a month before. I located the second spot relatively easily and set up the tripod again. A few ultra-high-ISO shots helped me frame the photo correctly. I also tried some low ISO long exposures with a bit of light painting so that I would have a foreground to go with it. This is a separate image that is blended with the processed astro image in post processing (see last image). This final photo here has to be my favourite image of the year so far. I didn't get a huge number of finished shots that I was happy with, but at the end of the day, if you come away with at least one, that's a win.

Milky Way Over The Allgäu || Olympus 8 mm, f/2.8, 25 s ISO 6400, Stacked

I carried on shooting for a while after this into the blue light of pre-dawn before giving up and catching a few more minutes sleep lent against my rucksack. It was a mild night with little wind and the ground was comfortable and dry beneath me, not a sound to be heard.

Birdsong and a yipping fox in the valley below alerted me to the coming dawn. By the light of the red band of cloud to the east I could just make out a small peat pond nearby that I'd just avoided in the dark. In retrospect I should have taken a lower vantage point for this shot. This is two images blended in ON1 Photo Raw, one for the sky, one for the foreground. I took the liberty of brightening the tufts of cotton grass to emphasise them in the pre-dawn light.

Mountain Tarn at Dawn || Olympus f/5.6, 1/6 s, ISO 200

Then it was down to find Martin and head down to the car. I found him warming himself over his camping stove and ready to return to the valley. As I waited for him to pack his things there was time for one last shot of the sun rising on the mountain escarpment opposite us. 

The Red Light Of Dawn || Olympus f/5.6, 1/30 s, ISO 200

Processing the Images

If you're not interested in the nitty-gritty of processing astro photography you can probably stop reading here and scan to the last images. It's very easy to over-process Milky Way photos: I spent two evenings working on this last shot only to have Sharon (correctly) tell me that it was too speckled and she didn't like the foreground. I've left this technical bit to the end for my techie friends who'll be interested in this stuff (hi Matthias!). 

Before the images can be 'stacked' (averaged), they need to be in tiff format. ON1 Photo Raw, my post-processing software can export tiffs, but they're the wrong format for my stacking software, Sequator. The images have to be opened in Olympus Workspace and exported as tiffs from there. For each image I used 10 star images and two control images with the lens cap on (lights and darks respectively).

Crop of main image

Unprocessed image OOC	Stacked in Sequator 10 Star images, 2 Noise images (with lens cap on), Align stars, Freeze ground, Sky region: Partial, otherwise all default settings	Processed in ON1 Photo Raw 2020 Develop: Contrast +24, Highlights +18, Midtones +15, Shadows -20, Whites +45, Temperature +40, Tint -3 Effects: Sharpening, Noise Reduction (Moderate), Tone Enhancer (Shadows Lighter on foreground; Midtone Contrast Boost, 50%), LUTs (1983, 35%), Sunshine (Sunglow, 50%), Curves (mild S-curve), Dynamic contrast on Milky Way

I also tried a blend of my favourite image with the light-painted foreground in ON1 Photo Raw. It's not perfect as I seem to have adjusted the camera between frames and had to stretch the resulting images to fit. I'd be interested to hear which image you prefer, this one with more detail in the foreground trees, or the original "Milky Way Over The Allgäu" above. My jury is still out on this one.

Milky Way Over The Allgäu II || Olympus 8 mm, f/2.8, 25 s ISO 6400, Stacked and Blended

Location Scouting for Astrophotography

A couple of weeks ago, one of my photography pals texted me asking me if I wanted to join him on a Milky Way hunt some time since his summer plans had been scratched due to Corona (Matthias, you really need to set up a website so I can link to your photos!). He mentioned a couple of possible locations, which got me thinking; where would _I_ shoot the Milky Way from. I came up with an idea (that I won't share here yet, I haven't seen any other astro shots from here and I'd like to have a crack at it before someone else gets in first, so if you recognise any of the spots in this post, plesase keep them to yourself for now - hopefully in a month or so we'll be able to pull this off). I knew that I wanted to include some 'civilisation glow' in the shot if possible - the night glow from a human habitation and I opened up Google Earth to look for a possible mountainside that would give me the angle that I wanted.

Watching the Sun Go Down || Olympus f/16, 1/60, ISO 1250

It's not as easy as I'd thought to find a deforested slope facing the right direction in the lower alps. Despite my local knowledge, it took quite a bit of playing around before I found a peak that would work, and even then it was only on paper. At the end of May I headed down there for an evening to catch the sunset and the blue hour to see whether the angles matched up. Even a couple of days beforehand I wasn't too sure about the trip, no-one wanted to join me and I was a little wary of being on my own in the mountains at night - even in familiar territory there are plenty of things that can go wrong. And the forecast wasn't playing ball. Until Friday morning that is. All of a sudden the evening forecast was opening up - a green light.

It's a Sign! || Olympus f/8, 1/80 s, ISO 200

With sunset projected to be 21.05, I set off from home just after 18.00 to allow myself enough time to get down there, get up onto the ridge and find a couple of spots whilst the light was still good. Climbing up from the car park the light was gorgeous. A few high clouds might have been nice to complete the scene, but I wasn't going to complain. The few people still on the hills but coming down gave me some strange looks and I had a quick chat with one bloke who was interested to know what I was doing going up at that time of day before he got pulled to heal by his wife. And then I was on the ridge.

At the Ridge || Olympus f/8, 1/500 s, ISO 200

The views were beautiful, clear air with some mountain cloud hanging around the peaks. It was all I had hoped for from my virtual tour planning and more. There is some lovely rock around, and the peaks still have significant patches of snow on them despite the non-winter of 2019/20. From the ridge its a broad path leading to one of the peaks that had looked promising and so after taking a couple of smartphone shots I headed out, warm setting sun on my left, valley on my right. The light was catching the fir trees along the ridge, especially the skeletal branches of dead trees that had long since lost their bark to the elements. The orange light picked out the highlights nicely.

Catching the Last Rays || Olympus f/8, 1/80 s, ISO 250

Just before the potential spot that I'd looked out there's a broad shoulder that looked promising, the spot itself had a lot more trees and was a lot narrower than it had looked on the computer, pretty much ruling it out. By this time, the sky was beginning to turn orange in the west and I was desperate to find an open piece of hillside facing westwards. East wasn't a problem, I could get to a treeless spot there any time, but the other side was light forest. In the end I had to settle for a gap between the trees. I say settle, I think the silhouettes work quite nicely, but it wasn't the picture I had first envisioned. Dropping the aperture to f/16 helped with the sunburst.

And Down She Goes || Olympus f/16, 1/100 s, ISO 200

In the immediate aftermath of the sunset I headed back along the ridge to where I had first hit it to wait for the first stars. The path back offered a couple of nice shots too, and I particularly like this one with the roots on the path. This is a composite of two exposures, one for the foreground and one for the background. After mixed experience with ON1 Photo Raw's HDR assembling, I've gone over to using layers to assemble these shots, it gives me a deal more flexibility when I can control precisely which elements of the photo I use from each exposure. I'll write a separate blog entry on this some time.

The Path Goes On || Olympus f/6.3, 1/50 s, ISO 1600

Up until this point I'd had my go-to lens on the camera, the Zuiko f4 12-100 mm, getting back to the top of the ridge it was time to get the wide-angle zoom out, the Leica f2.8 8-18 mm. The wider aperture and wider field of view make this a great astro-landscape lens, although I wouldn't complain if there was a convenient lens with an even wider aperture for night shots. The optical quality is second to none though, provided I can focus it properly. During the daytime this isn't an issue, I just leave it on autofocus normally, but focusing on pin-pricks of light at night to get the infinity focus right is tricky, even with glasses.

Golden to Blue || Olympus f/4, 1/60 s, ISO 500

I set up the tripod with my Benro Geared Head to allow me good control over the level and direction and started shooting the blue hour. When hiking, I rarely wear more than t-shirt and shorts, but even setting out from the car I was wearing long sleeves and trousers. Despite my soft-shell jacket it was beginning to get quite cool in the late evening breeze. Wooly hat and gloves time, though putting the gloves on and off was a pain. I was glad of the tripod net that I'd bought for the occasion to store bits of camera in the low light, it made keeping kit under control in the dark a lot easier. 

Waiting for the Stars || Olympus f/2.8, 1/6 s, ISO 200

And so I sat and waited for the first stars to appear, the half-moon high in the sky away to my right, the dew beginning to fall. Taking a shot every five minutes or so, exposure bracketing just to be on the safe side. I was watching the clock too, I didn't want to be out too late on my own - there was still a trek down to the car in the dark. Even with the half-moon and my head-torch, stumbling around in the dark in the mountains shouldn't be taken lightly. And then there was the 90 min drive home.

Gibbous Moon over the Allgäu || Olympus f/3.3, 10 sec, ISO 200

The descent proved easier than I had feared and I was soon back on the tarmac. Definitely worth the evening's trek. Oh, and remember the SD card that went through the washing machine? It seems to have finally given up the ghost. Fortunately after I transferred the photos to the PC and not before. Sometimes you've gotta have a bit of luck!

End of the Day || Olympus f/4, 0.8 s, ISO 800

When What You See Is Not What You Get

or 'How to Suck Less at Photography'

 

So often I'll get back from a day's shoot and I'm disappointed by a greater or lesser portion of my photos. Things that looked great in the field simply look bleurgh on my computer screen and I'm not always sure why. A mountain vista, a forest scene, a great skyline all looked really impressive in real life, but for one reason or another the picture fails to live up to the reality. There are as many strategies for improving the quality of our images as there are photographers, and believe me, I've listened to a lot of them. In the field I've tried to concentrate on the conventional rules of composition, but sometimes something still just isn't right. With experience I've started recognising more and more frequently what will and what won't work - a photographer's spidey-sense. I've come to realise that part of the problem lies in the differences between what we perceive in the field and what the camera sees.

Sometimes It Does All Come Together

Of course there are technical mistakes we can make in the field too that result in images being sent straight to the recycle bin; too slow shutter speed resulting in motion blur, missed focus, over- and under-exposure, but those are rookie mistakes we don't make any more, aren't they? (I wish!)

Three Differences Between your Eyes and your Camera

There are (at least) three significant differences between our eyes and even the best camera that negatively affect the images we create. Knowing what they are can seriously improve our photography. Forewarned is forearmed, as they say, and if we know ahead of time what isn't going to work, we can either avoid those shots or we can try to compensate for our cameras' 'deficiencies'.

Vision or Gaze Focus

Let's start with focus. How broad is your vision focus? What do I mean? If we include our peripheral vision, most people can see approximately a 170° arc in front of us. Coming forward we pass from far- to mid- and near-peripheral vision before we reach centre gaze. There's a great little Wikipedia article about it. Not with me? Try this. Put your finger in the middle of a piece of text such as this page of writing. Focus on your finger and now try to read a word at the edge of the page without shifting your focus. Chances are, depending on the size of the font and the size of screen that you're reading this on, you'll only be able to read a couple of words either side of your finger before you have to start making educated guesses. 

Our focal range is extremely narrow and has to do with the concentration of light receptor cells in a narrow part of the retina in our eyes. It's actually extremely efficient to see like this - peripheral vision is our radar for detecting things of potential interest, but we don't need to be able to discern all the information in this part of our sight, especially when we're in a familiar environment. It would take up far too much of our attention. Instead, the ability to discern detail in a scene is limited to a few degrees directly in front of direction of gaze.

By way of contrast, our camera picks up all the information equally over the whole picture. There's no concentration of pixels at the centre of an image, we're presented with all of it at once. Of course, when we're looking at a photograph we can only focus on a narrow part of that image, but all the information is there.

Why is this important? Of all of the differences between our eyes and cameras, I think that this is the one that's most difficult to get our heads around. At least until we become aware of it, and even then it catches me out as often as not. Take this image of a mountain woodland scene taken on a recent excursion to the alps. Walking down the path I was struck by the glossy lime green beech tree left of centre. The photograph doesn't work though, there are too many details competing for your attention. It's one of those scenes that is almost impossible to capture well unless the conditions are on your side, but we'll look at that in a minute.

When Selective Focus Lets Us Down

3D

This is the obvious one. Stereoscopic vision affords us the ability to perceive our surroundings in three dimensions, particularly objects at close range. There are three mechanisms we use to estimate the distance of an object - stereoscopic vision, experience and motion. If we know how big an object is, we can estimate its distance depending on its size independent of stereoscopic vision, similarly, if we know the type of object we're looking at our brain is quite good at interpolating that object's speed to distance. Surprisingly, our 3D vision is only reliable to about 6 m (!). Nevertheless, the fact that we perceive in 3D is very different to the way a single lens camera depicts an image. We're afforded an instant snapshot of a scene. We can only discern the three dimensional context based on visual cues that we're familiar with.

Dynamic Range

Dynamic range is the difference in light intensity between the darkest and lightest parts of a scene. Photographers like to talk in terms of stops, a stop being a doubling or halving of the amount of light. A good camera can differentiate up to 14 stops of light at its native ISO (the standard ISO of the camera, not necessarily the lowest). That's a factor of 16,348 from the darkest point to the lightest point. Impressive, huh? Sure. Until you realise that the eye is able to distinguish 18-20 stops - up to a factor of just over a million. So the eye is 64x better at distinguishing between light and dark than your camera.

So How do we Overcome the Deficiencies of our Cameras?

To make good images, we have to compensate for these differences and bring the viewer's eye to the subject of the photo using some tricks. This is the art of composition, creating a strong image using the tools available to us. 

There are ways of overcoming our cameras' 'deficiencies' (actually I prefer to think about differences between eye and camera rather than deficiencies of the latter - our cameras are actually pretty advanced tools): In order to overcome issues associated with dynamic range, we can take multiple shots at different exposure settings and combine them either in camera or in post-production (high dynamic range or HDR images). There are numerous techniques for enhancing the 3D-feel of our images, for example by shooting low to the ground or using a wide angle lens. The aim of all these techniques is to lead the viewer's eye to the subject of our composition. There are also numerous composition tricks we can use to do this, structuring our photos using the rule of thirds or the golden rule, geometry, leading lines and the like. There's plenty of good content out there on how tame our cameras to compensate for these differences.

Vision focus isn't so easy though and requires that we think about our images slightly differently. One way of understanding how to bring the viewer's attention to the subject of our image is to think in terms of separation: How can I separate the subject of my photo from its environment in order to make it clear to the viewer that this is what the image is about? Of course, this pre-supposes that we know what the subject of our image is ourselves, which isn't always obvious, particularly when it isn't even a concrete object.

Achieving Separation

What do I mean by separation? It's the skill of highlighting the subject of your image in such a way that it's obvious to the viewer what the photo is about. This separation can be subtle or obvious, there are a number of tricks we can use. I want to mention five here. The list is not exhaustive and they can all be combined with one another for more or less effect. Sometimes using just one of them can lead to really strong images though.

Light

Light is the easiest of the five tools to use. The eye naturally falls on the brightest parts of an image. Spotlights pick out the members of a band at a concert to focus your attention on them rather than the mess of back stage. Parts of our image that are unintentionally bright distract the eyes and pull them away from what we want the viewer to look at. 

Of course, we can turn this on its head and use dark to highlight the subject of the image, but the default setting is that light attracts. By using light to emphasise the subject of our photograph, we're effectively shining the spotlight on it. Woodland can be very effective for this, such as this picture of wood sorrel. On a stormy day gaps in the clouds can yield similar effects, such as the photo of Cinque Torri in the Dolomites in my gallery.

Spotlight On Sorrel

Colour

After light, the next most obvious tool to draw your viewer's attention to the subject of your image is colour. Bold, saturated colours are more effective than muted, pastel colours (think of text highlighters!). Many photographers will talk to you about colour theory and complementary colours. All well and good, but instinctively I think we know which colours work together and which colours contrast (if we didn't it wouldn't work).

The red-purple marsh orchid really stands out here against the lush green spring grass of this Bavarian pastorale.

Outstanding in its Field

Depth of Field

Depth of field is the relative distance in front of and behind the focal plane of the camera that is effectively in focus. A narrow depth of field means that focus is restricted to a short distance around the focal plane, a wide depth of field puts more of the image into focus.

We can control it using the camera's aperture; an open aperture generally results in a narrow depth of field, a closed aperture in a broader depth of field. The closer the subject matter and the longer the focal length of the lens used, the narrower the depth of field too.

Like with light, the eye naturally wanders to what's in focus in an image, like the dandelion centre left here. The closer and further flowers are out of focus thanks to an aperture of f5.6. A similar effect can be achieved in mist or fog.

The Eye Flies to the Focal Point

Composition

I showed you an example of poor composition earlier; the beech tree against the woods. It was literally impossible to see the wood for the trees! If, though, you can put a bit of empty space between your subject and its environment - what photographers classically call separation - it's easier for the eye to find its intended target. It's a frame within a frame, like a bulls-eye saying 'look at me'.

Parting is such sweet sorrel - using the roots to frame the greenery

Texture

Texture is particularly powerful in monochrome photography, where it can really stand out in an image. It's probably the subtle changes in light that stick out and help the viewer to focus.

I didn't have a good example of this from the field so I took this quick and dirty photo of the rug in our living room. Rug, edging and tiles all have a very similar tones, but the texture of the three elements is clearly distinct, allowing the viewer to visually separate the elements.

Texture Separation

Separation using Light	Separation using Colour
Separation using Depth of Field	Separation using - Separation

Bringing it all Together

So how do we put all this into practice? When something catches your eye in the field, before you raise your camera to your eye, stop a second and analyse what's in front of you. What is that has caught your attention? Let your eye wander over the scene for a second or two so that you can identify the subject and ask yourself whether there's enough separation between the subject and the background that the viewer can identify what the subject is. If there isn't, how can you generate separation using some of the tools I've mentioned above?

If you've found this article useful, let me know below. Also, if you're in the Munich area and would be interested in exploring putting some of this into practice together in the field, drop me a line (contact details on the right) and we'll see if we can set something up!