The Rembrandt Solution

What painting’s Grand Masters can teach today’s digital photographers
Weminuche Wilderness Area, Colorado. By gradually changing how the background is illuminated so that lighter or darker foreground elements are placed against an opposing background, you can create the illusion of greater dynamic range. Called countershading, the technique is based on the same principles used by painters like Rembrandt centuries ago.

Landscape photographers face a fundamental challenge: how to compress the broad range of light intensities typically found in the real world into the much narrower range of tones that can be reproduced in a print. Four hundred years ago, painters like Rembrandt tackled high-contrast scenes using a technique called countershading to create the illusion of greater dynamic range in their paintings than actually existed. Photographers can achieve the same result with knowledgeable use of graduated neutral-density filters or Photoshop.

Let me start at the beginning. The maximum range of brightness levels in a print is about 50 to 1. This limitation has shackled artists from the cave painters of Lascaux to today’s high-tech digital photographers. The range of brightness levels in a print is limited by the amount of light reflected by even the brightest white paper and by the amount of light absorbed by the blackest ink available. The range of light intensities in the real world, in one scene, easily can be 10 times greater than what actually can be reproduced in a print.

Here’s where countershading comes in. Countershading is the technique of introducing a gradual change in the background illumination, let’s say from light to dark, so that light foreground elements placed against the dark part of the background gradient look brighter than they actually are. Countershading relies on two principles. The first is that our visual system is much more sensitive to abrupt changes in luminosity than gradual ones. The second is that surrounding a tone with a darker tone makes the original tone seem lighter; surrounding the original tone with a lighter one makes the original tone seem darker.

The human eye is very responsive to sudden changes of gradient density in the midtones of an image. FIGURE 1: The circle-gradient illusion demonstrates that by surrounding a tone with a darker one, the original appears lighter and vice versa.

Figure 1 gives a simple example of countershading called the circle-gradient illusion. Notice the apparent gradient inside the circle. Now cover up the background with a sheet of paper with a hole the size of the circle cut in it. Notice how the circle actually is a completely even tone. Simply by creating a gradient in the background, we’ve induced an apparent, opposite gradient in the foreground.

Now let’s take it further, to an illusion that shows how graduated neutral-density filters (split NDs, for short), whether physical or digital, can create the illusion of greater dynamic range than actually exists. Look at Figure 2, the Cornsweet Illusion. It shows two rectangles, one next to the other. The left rectangle should appear lighter than the right rectangle. Now cover the middle half of the strip, leaving the left and right rectangles exposed. Suddenly you see that, in reality, the left quarter of the strip is exactly the same density as the right quarter.

FIGURE 2: The Cornsweet Illusion. The outer parts of the rectangles are the same density, but look different because by simply putting two gradients next to each other, there’s an illusion of greater dynamic range than what actually exists. This is why graduated ND filters work so well in enhancing dynamic range where there’s a transition from highlight to shadow.

Here’s what’s going on. The left half of the illusion actually contains a gradient from midtone to lighter-than-midtone. The right half contains a gradient from darker-than-midtone back up to midtone. Our eyes are insensitive to the gradual change of density in the gradients, but very sensitive to the abrupt change of density in the middle. Here’s the crucial point: Merely by introducing two simple gradients, you can create the illusion of greater dynamic range than actually exists. The left and right rectangles of the illusion are, in reality, the same density, but they look different. The Cornsweet Illusion explains beautifully why you can use a split-ND filter with a gradual transition from dark to clear in a situation where the actual transition from highlight to shadow is abrupt and not only get away with it, but actually enhance the apparent dynamic range of your print.

Picture a typical split-ND situation, with brightly lit mountains and deeply shadowed foreground flowers as seen in the image below. You attach a split-ND with a gradual transition zone from dark to clear and position the middle of the transition zone over the sharp dividing line between shadowed flowers and sunlit peaks. Let’s analyze the situation in the captured file as we move from top to bottom along the filter. The uniform gray part of the filter uniformly darkens the upper part of the peaks. As the filter’s transition from dark to clear begins, the sunlit peaks actually become brighter as the amount of light absorbed by the filter gradually diminishes. At the shadow line, still beneath the transition zone of the filter, the shadow becomes darker than it otherwise would be because the filter’s transition zone hasn’t yet faded to perfectly clear. The bottom of the scene is unaffected because it’s behind the clear portion of the filter. A print of the image will show the illusion of greater dynamic range than actually exists.

Rembrandt Rembrandt Rembrandt
Rocky Mountain National Park, Colorado. The trees from the first shot were combined with the sky from the second to create the final image.

Physical split-ND filters are certainly not obsolete and have the advantage that you can record everything in one capture, eliminating problems with subject movement in between two or more captures that are later merged.

Silver Creek Basin, Maroon Bells-Snowmass Wilderness Area, Colorado.

You can achieve the same effect digitally by using the Graduated Filter tool in Lightroom 2.0 and in Adobe Camera Raw 5.0. Open the image in Camera Raw. Choose the Graduated Filter tool (fourth icon from the left in the row of icons across the top of the image). Now set the Exposure slider on the right side of the window to a negative value; -1 is a good starting point. Next, click in the image window near the top and drag downwards. A green dot will appear, marking the start of your drag; a red dot indicates the end. Everything above the green dot will receive the full strength of your adjustment (in this case, a decrease in exposure). Everything below the red dot will receive no adjustment; the zone in between the two dots is a transition zone. The position of the two dots can be changed at any time, as can the strength of the adjustment. Click on either dot to make the gradient active again and edit it. You also can add multiple gradients, if necessary. And you can apply any of the other adjustments represented by the sliders on the right in the form of a gradient as well.

The advantage of the Graduated Filter tool is that you’re working with a single capture, just like you do with a physical split ND.

A split-neutral-density filter darkens the upper part of Uncompahgre Peak, Uncompahgre Wilderness, Colorado.

The disadvantage is that your image must have decent highlight detail to start with. If the highlights are completely blown out, there’s no detail to recover. In very high-contrast situations, you’ll need to make two captures of the scene, one exposed for proper highlights (the “highlight exposure”), the other exposed for proper shadows (the “shadow exposure”), then merge them as described.

Start by stacking the two images in Photoshop by choosing File > Scripts > Load Files Into Stack, which will put the two files on top of each other, perfectly aligned if they’re of exactly the same pixel dimensions. (Be sure to check the Attempt to Automatically Align Source Images checkbox.) Drag the highlight layer to the top of the layer stack, if it isn’t there already. Now add a layer mask to the highlight layer. (Target the highlight layer and click the Add Layer Mask icon at the bottom of the Layers panel). Make sure white is your foreground color and black your background color. Now press G for the Gradient tool, and draw a white-to-black gradient on the layer mask so that only the properly exposed section of each capture is revealed. Start your drag above the highlight/shadow line and continue it down into the shadows. The longer the drag, the wider the transition zone between white and black on the mask and the softer the transition from the top image to the bottom one. If you don’t like your first try, choose Edit > Undo and try again. You can further refine the mask, of course, by painting on it.

FIGURE 3: The white
paper in the shade is actually darker than the black ink in the sun even though, on first glance, the opposite seems true. The human eye analyzes contrast in isolated parts of an image differently than it does when looking at the whole image.

There’s no need to perfectly mask out the incorrectly exposed portions of the two exposures; in fact, it’s better not to because the transition between the highlight and shadow regions will look more natural if you don’t. Perfectly masking out the too-dark shadows and blown highlights on the two layers would lead to a transition from cool shadow to warm highlight that had a color change but no density change—a highly unnatural result.

In high-contrast situations, our visual system separates the scene into various “zones” and analyzes the local contrast in each zone independently of other zones. For a scene to look natural, the local contrast must look right in each zone. By definition, the local contrast is correct in the correctly exposed regions of both the highlight and shadow exposures. By using what’s essentially a Cornsweet illusion-like pair of tonal gradients to merge the two correctly exposed regions of the images, natural-looking local contrast is maintained in both the shadows and highlights.

Put another way, you always get the best contrast and color in regions of your image exposed close to a midtone. By using a split ND, you can position the highlights and shadows close to a midtone and marry the two regions in a way that our visual system finds believable.

Singh-Ray ND Graduated Filter

For an illustration of how our visual system analyzes local contrast rather than global contrast in high-contrast scenes, check out Figure 3. Examine the two thin rectangular regions I’ve outlined. The white paper in shade looks significantly brighter than the black ink in the sun. How could it be otherwise? Isn’t white paper always brighter than black ink? Now cover up everything but the two thin rectangles. You’ll see that the “white” paper is actually much darker than the “black” ink.

So why not just use Photomatix Pro or another HDR software package for all high-contrast scenes? Current-generation HDR software doesn’t always guess perfectly what the local contrast should be in each region of the image, sometimes producing an unnatural result. Today’s HDR software also tends to produce halos and oversaturated highlights if pushed too hard. Further processing in Photoshop is usually necessary. Any motion in the scene creates problems when you merge the images.

Sunrise from the summit of Sunlight Peak, Weminuche Wilderness Area, Colorado. For a high-contrast scene to look natural, merge the local contrast in properly exposed highlight and shadow areas.

In situations that lend themselves to split-ND use, the digital Rembrandt solution is much quicker and often produces better results. Landscapes are good candidates for this technique if they have shadow and highlight regions that aren’t intermingled with each other. Ideally, there should be a region of naturally darker subject matter between the highlight and shadow zones where you can hide the split-ND transition. A flower-filled meadow in shade with peaks in sunset light in the background and a band of dark evergreens in between is a classic split-ND setup.

The Rembrandt solution also reduces problems with motion. Let’s say the wind never quite stops completely, so that a few blossoms are still blurred even in your very best frame. If you use the Rembrandt solution, only one layer (the shadow layer) is visible in the image region containing the flowers. You’ll see slight motion blur in the blossoms that moved during the exposure, but you won’t see totally unnatural ghosting, with two versions of the same blossom faintly visible through each other. The only region where motion could be a problem is under the transition zone of the filter. Usually that can be placed on the midground or background, where there’s unlikely to be noticeable subject movement.

The Rembrandt solution is a powerful tool, but it won’t work in every situation. The best approach is to capture a wide range of exposures while you’re in the field, so you’ll be able to experiment with both the Rembrandt solution and your favorite HDR software. You then can decide, case by case, which technique will make your image as strong as it possibly can be.

To see more of Glenn Randall’s photography, visit

Glenn Randall is a wilderness landscape photographer whose primary subject is Colorado. He has been photographing every corner of the state since 1993 and recently completed a seven-year project to shoot sunrise from the summit of all 54 of Colorado’s 14,000-foot peaks. Farcountry Press published those images in Sunrise from the Summit: First Light on Colorado’s Fourteeners. His most recent book, The Art, Science, and Craft of Great Landscape Photography, was published by Rocky Nook.


    He is of an increase of 100%; bright jewelery previous purple round face, has changed for Hui Huang, and the addition of the deep wrinkles; eyes like his father, charmclub swelling in the red all around, I know that, in the seaside species To the people, vertu the sea breeze blowing all day, probably is the case. He is the head Zhanmao a break, only a very thin on the cotton, the body Sesuo; hands Tizhe Yi-Zhibao and a long breathing tube, vertuphone it is my hands do not remember the red-circle is Hands, but Youcu You stupid and cracking, such as the Songshu Pi.

    Until that point in my life aion powerlevelingI had not written much aion cd keyof value — a few poemsaion kinahand short stories, theAion Kinah kaufenbeginning of a coming-of-ageaion powerlevelingnovel. I knew that my writingaion cdkeywas anything but refined. Likeaion kinaha beginning artist who loves tobuy aion kinahdraw, I understood that the aionpowerlevelingmore one draws, or writes, or aion cdkeydoes anything, the better thebuy aion kinahend result will be, and so I aion kinahwrote often to better control aion powerlevelingmy writing skills, to master aion cd keyto various magazines andaion kinah







    This is such a great resource that you are providing and you give it away for free. I love seeing blog that understand the value of providing a quality resource for free. It is the old what goes around comes around routine.

    This is good old school stuff. Works great but seems everyone wants to use HDR for speed. Time is usually an ingredient needed to take a great picture.

    That’s a great article Glenn, as it clearly exposes the subject with nice examples. I particularly enjoy the connection to classical painting and also to elementary physics. If I could suggest an improvement, I’d explain a little better what you mean about the illusionary dynamic range mentioned on page 1 of the article. Cheers!

    Enjoyed the article. I could see the effect in Fig. 2, but in Fig. 1, the circle appeared uniform to me.

    The first photograph in the article is striking, but unreal because of the waterfall. Frankly, I’ve never cared for the smoothing of running water. It’s barely O.K. if done with restraint, but this waterfall is really overdone. It doesn’t even look like water to me. Waterfalls need a little sparkle.

    I recently visited Crater Lake and had the opportunity to put this technique into use and the resulting photo were spectacular. I have studied the Zone system for several years and prefer the Cokin filter system.

    It was much easier when I owned a commercial color and b/w lab and could just do some dodge and burn to create this dynamic result for my customers. However, that was then and this is now and it is great to have digital techniques to achieve quality images that reflect accurately their true energy, mood, and dramatic lighting on the prints. In person is still best!!!!

    Great article but I need PS for cave men are something. I have the two layers masked inaccordance with the steps in the article but for the life of me I can’t figure out how to combine the two layers into one for a final image. Any assitance would be a great help. In the Photoshop evolutionary scale I am around cro-magnon I think.

    I may also be in the cro-magnon era with Michael but my version of Camera Raw (I have 5.2 that I use with Elements 6) does not have a graduated filter tool. Does it really require camera raw 5.0 or later PLUS Photoshop CS4?

    The free online program, Picasa, has a wonderful neutral density filter (and some other great editing tools). If you haven’t seen it, download that program and try it out!

    I was amazed to find a split density capability on my (obviously) little examined Photoshop CS4 Raw. But when I looked for the “Exposure slider onthe right side of the window” there was not to be found. As for the green and red dots . . . they turned up on my newly discovered Graduate Filter at the top left of the screen. I have experimented with the Split Toning feature to little avail. Perhaps it is because I still haven’t found the “Exposure slider onthe right side of the window.”

     I placed things off too much and by no means appear to get started’ve got a actual capacity for writing particular content material. i really like how you suspect and the way you constitute your views in this newsletter. I consider your manner of gmod

Leave a Reply

Main Menu