Part XVIII - Tools - Effective Cropping, Straightening and Leveling
Introduction: Adobe Camera RAW Do's and Don'tsSo far, we've been dealing with Lightroom's basic develop tools and we've seen how the tonal scale of an image can be flexibly tweaked relying only on these tools.
Nevertheless, it's important to understand that most of those tools are provided by Adobe Camera RAW, running within Lightroom under the hood, and you may think of them as of your digital darkroom. More often than not I see people stuck with Camera RAW controls while trying to get the picture they want but, sadly, without success. Some beginners fail to realize the purpose of Camera RAW.
This is important: read carefully on.
Adobe Camera RAW it's the engine that reads and interprets RAW files to produce a final image that can be processed by an image editing tool such as Photoshop or Lightroom. To put it in another way, Camera RAW is the bridge between you and the proprietary format of the RAW file produced by your camera sensor. That's why Adobe ships Camera RAW inside those programs: in order for photographers to seamlessly manage their RAW files without a workflow disruption. You don't have to rely to an external program (such the one that comes with your camera) to produce an image from your RAW files: the develop stage is just part of the workflow and you do it seamlessly inside Lightroom (and in Photoshop to a lesser extent).
For this reason, you should consider Camera RAW develop controls the starting point of your develop workflow and be aware of their purpose, drawbacks and limitations.
The purpose of shooting RAW and using a raw processor such as Camera RAW is simply taking advantage of the additional information gathered by your camera sensor. In fact, it's during the develop stage when you take full advantage of it: Adobe Camera RAW uses the information found on RAW files to produce the final image according to the parameters you've set. When you set the white and the black point of an image using tonal scale controls such as exposure, recovery and blacks, Camera RAW uses the data of your sensor to produce the final image. Since sensors can have very high dynamic ranges (13 stops are common nowadays), you can use those information to produce a faithful resulting image even if you're applying strong exposure corrections, even in the 1-stop range or more. Depending on your sensor dynamic range, you will be able to get more details from the shadows when you're raising the exposure, or more details in the highlights when you're lowering it. It all boils down to what we've explained more extensively in previous posts: you should always shoot RAW.
Why Tone Reproduction Is DifficultThis is an introductory section about tone reproduction issues and dynamic range of some common mediums used by photographers. You can skip this section if these concepts are already known to you. As a more detailed introduction about the concepts used hereon, you can read this post: "Tones and Dynamic Range. Why You Should Shoot RAW".
The human eye has a very high dynamic range, much higher than your camera sensor (at least nowadays): it's an estimated 30 f-stops. But the physics of the human eye is very different than the physics of a sensor: the eye cannot see very distant luminance values at the same time since it needs some "adjustment time". You're surely are aware of if: try to go out on a sunny day and fix your attention to a very dark spot. You will be able to see details in the shadow as well as you can in the bright, but not at the same time.
The problem with camera sensors is that, when you take a photo, you're recording sensor data in a given instant of time. That's the reason why you simply cannot have satisfactory sensor reading of a scene whose dynamic range is bigger than your sensor's. That's also why techniques such as HDR were invented: multiple exposures are taken to faithfully record data of very different parts of an image (luminance-wise) and then blend them together.
This problem is effectively worsened when you consider other mediums you use during your workflow whose dynamic range is nowhere near to the theoretical value you need, such as your monitor or your printer.
To solve this problem, you "fix" tones in your image to give a "proper", although subjective, luminance value to different zones of your image. In fact, many steps in a photographer's workflow may involve non-linear tone mapping functions. As an example, think about gamma correction: it's a non-linear transformation meant to adjust luminance output taking into account the physics of the human eye.
Tone CurveLightroom, as well as most image editing programs, offer a tool that let you modify such tone mapping function: it's usually called the tone curve, or simply curves.
|Lightroom Develop Module - Tone Curve Panel|
This tool is found into the Tone Curve panel of Lightroom's Develop module. In the upper part of the image you can see a graph that has the following characteristics:
- The x-axis represent the original tone value of the image from black to white (the input value).
- The y-axis represent the modified tone value of the image from black to white (the output value, what we get on the picture).
- The current image histogram in the background.
- The mapping function plotted in white above the histogram.
An identical transformation (in which case no modification is applied) is represented by the y = x equation, which is plotted a straight line from the origin of the graph to the upper right corner. The typical S-curve used by default by Lightroom and shown in the previous picture adds some contrast to your image: dark tones are mapped to darker ones (y < x), the middle tones doesn't get remapped (y = x) and light tones are mapped to lighter ones (y > x).
You can tweak the mapping function to remap a specific tone value to another one in a variety of ways:
- Applying adjustments to the Region sliders.
- Applying free-hand transformation to the curve.
- Using the adjust tone tool and dragging the mouse over the part of the image you want to change.
|Tone Curve - Darkening Shadows|
Sliders are an intuitive way to modify the tone curve and use the same naming conventions and histogram subdivisions we've seen on basic tone controls:
- Highlights, roughly corresponding to the zone affected by the Recovery tool.
- Lights, roughly corresponding to the zone affected by the Exposure tool.
- Darks, roughly corresponding to the zone affected by Fill Lights tool.
- Shadows, roughly corresponding to the zone affected by Blacks tool.
You are not limited to using the Region sliders or applying monotonic transformations, though. If you need to apply more complex tone transformation functions, you can switch to the free-hand transformation mode (using the small graph icon on the bottom right corner of the panel) to add and drag control points on the mapping function according to your needs:
|Tone Curve - Custom Curve|
Also, if you want to modify the mapping of a specific tone in your photo, you can use the following procedure:
- Grab the Adjust Tone Curve tool, clicking on the small circular icon on the left of the graph: your mouse pointer will change to a cross with a slider control.
- Click on the point of the image whose tone you want to change.
- Keep the mouse clicked and drag upwards or downwards to raise or lower the mapping for that tone.
Why This Duality?At the beginning it may seem to you that the Tone Curve tool sort of duplicates the functionality of the basic tone controls we've seen in the Basic tools.
To answer this question with a quick "No", is sufficient to note that basic tone tools won't allow for non monotonic transformation to be applied. You can tweak each region of the histogram, but the side effect will always be a monotonic transformation.
There are is a difference that is even more important to grasp: the tone mapping curve only maps the currently displayed tones. On the other hand, when "developing" a photo with Camera RAW you are able to effectively choose which tones will end up in your photo, moving the entire histogram up and down, and using additional data from your RAW file. That means that the effect of a mapping curve on the clipping points is much more limited than what you can achieve using Camera RAW tone tools.
The latter can be used, as we know, to recover blown out data in the highlights or details from the shadows. The former simply remaps the information that's available at the moment, that is, the image produced by Camera RAW.
In the following section we'll see an example of this effect to help you understand when to use one tool or the other.
An Example ImageSome days ago I entered, almost by chance, Amsterdam's Openbare Bibliotheek and I was captivated by the incredible light that was in there. The building is entirely white and there's a strong, diffused, soft white light throughout the library. I soon decided to take some shots. This is one of the final images:
|Amsterdam's Openbare Bibliotheek|
This is a classical example of the tone reproduction problem we were talking about in the first sections of this post: the human eye is very good at quickly jumping from one zone to another with very different luminance levels but recording this on a camera sensor is a tough problem. The scene in question, in fact, is a scene whose tone cannot be reproduced either without being remapping or being clipped. In fact, the resulting image is the result of a specific process whose goal was to maximize the luminance of the environmental lights while preserving the "correct" exposure of the darker zone (such as the floor): in practice, this was achieved "squeezing" the luminance information into the histogram taking care to maintain a natural contrast between darker and lighter zones.
Taking the PictureThe first step obviously was taking the picture. You have already learnt that more often than you'd like you've got to come to a compromise when adjusting your image exposure. In this case, being a strongly lit environment, I decided to give preference to highlight preservation, despite looking for a high key image.
Since I know that my camera meter was going to be fooled by that lightning condition, I manually underexposed the image by 1.5 f-stops, to be sure not to completely burn out the highlights, double checking this with the on-camera histogram. The result was this:
|Original Image - Highlights preserved|
This is a conservative approach: in my mind I already knew that I wanted a high key image and I could have accepted some clipping on the highlights. But if you clip a channel, that information is gone forever, and we don't want that: sometimes, you may envision a different use for a picture in post-production. Moreover, in this case, I haven't lost details in the shadows even if the image is strongly underexposed. And even if I had, I knew that I could recover much of them using additional data from my camera sensor: over time I learnt that my sensor allows for recovering shadow details when underexposed more than a 1-stop and, in this case, I was underexposing in that range.
Of course, every camera is different and you've got to know yours when assessing the lightning conditions and taking a decision and remember: there are times when you simply cannot record all of a scene's dynamic range and you'll have to make a choice. The camera histogram will help you check that the result is what you want. If you're unsure, bracket the exposure instead of losing a good shot.
Basic Image DevelopmentThe first step in the workflow is "developing" the RAW file to produce a suitable image, as explained so far in the previous parts of this tutorial. The first step was correcting the white balance and the tint: as you can see in the original image, it's pretty warmer than it should be. My camera wasn't calibrated for that lightning condition nor was I using a grey card. However, there's plenty of white in the picture to use Lightroom's White Balance Selector to fix that.
|White Balance Correction|
The second step was setting the initial tone settings for the image. I raise the exposure by 1.55 f-stops and checked the highlights:
This is where we are going to see a fundamental difference between basic tonal settings (from Camera RAW) and tone curve adjustment. Look at the difference between the previous image and the image we want to get: the previous image is still much darker, especially in the white zones in the roof. The floor, however, is correctly exposed. We could try and boost the exposure:
Now, the roof is as exposed as I'd like but, obviously, the bottom half of the image is really washed out. Remember that this is subjective: I want strong lights in the roof, but correct exposure in the floor. If I wanted a higher key image, I'd probably raise exposure a little bit more. If you feel it's right to push exposure further, go on: it all depends on the result you want to achieve.
How could we fix that using basic tone controls? Well, the only way of pushing information back on the dark side of the histogram is raising the blacks:
As soon as we begin raising the blacks, we notice that two things happen: the contrast and the color saturation rise as well. Why? Because rising the blacks stretches the histogram left and discards information that goes beyond the black point. You end up with a subset of the original data stretched throughout the histogram. It's a good technique to create a punchy sunset, but that's not we want in this case. Look at some dark zone even with a very moderate increase of the blacks:
The shadows in the floor are exaggeratedly deep. That's the effect of raising the black: you're effectively establishing where the black point is and information below it is discarded.
That's why, in this case, we're going to use a tone mapping curve instead. Basic tone controls are used to generate a picture form the RAW: in this case, we set the overall exposure of the image to a suitable level. and we cannot go further. From here on, we're going to rely on tone curves for additional editing instead.
Adjusting the Tone CurveNow that we've got a suitable image to edit, all we want is: raising the lights. Let's adjust the curve until we're happy with it and we end up with the final image we've seen at the beginning of this post:
The adjustment used to get the final image is the following:
|Tone Curve Adjustment - Lights: +57, Highlights: +8|
The curve is an S-shape more pronounced on the right part of the histogram. Its effect is lightening the tones from the darks zone upwards (where y > x) and the bigger x, the more pronounced its effect, since the difference (y - x) gets increasingly bigger.
A Much Higher KeyAlthough this is not a suitable image for this kind of mood, let's see how you could achieve a "stairway to heaven" effect with a custom curve. Pushing lights further and reducing highlights to preserve some details gives us the following result:
|Higher Key Image|
The corresponding tone curve is the following:
|Tone Curve Adjustment - Lights: +100, Highlights: -86|
ConclusionCurves are a powerful tool to customize the representation of tones in your image and, because of the very nature of the human eyes and of our camera sensors, they are a fundamental part of a photographer's everyday workflow. They may seem intimidating to the novice but they're pretty intuitive once you understand their principle of operation.
Also, tone curves are complementary to basic (Camera RAW) tone controls and it's important you understand and master both to achieve the results you want.
If you want to help me keep on writing this blog, buy your Adobe Photoshop licenses at the best price on Amazon using the links below.