Post Processing

General workflow to do Post Processing

Article written by: Siegfried Seierlein
Last Updated: 16th May 2008

Post processing is a dynamic process and changes as one learn new techniques. I can remember with the older MyCanonG7 site I also wrote an article on post processing that looked very different to what we discuss today. There was nothing wrong with the technique, describing it was just much more intense. Today I use better and more accurate techniques. There are so many different editing variations and that makes it really difficult to say the one is better than the other. A good understanding of one software package like Photoshop will help you to master picture editing.

As said, this is a dynamic subject and I will therefor keep on updating this page as I learn more techniques. This page is also focussed on the JPEG version of the picture and not the RAW version.

While reading and practicing these techniques I will always try to point out when a technique can damage the file. For example the inherent nature of a JPEG file is to systematically destroy the file. Many photographers stop using a JPEG file if it was saved a 3rd time.

Then before we start, keep in mind that this article will only discuss post processing techniques and not techniques on how to change pictures. By change I mean combining pictures or HDR techniques or any other changes to the picture other than optimizing the taken picture.

Important to know

The Photoshop editing engine has a “destructive” effect on pictures.
Increasing saturation can result in loss of detail
Increasing contrast can result in loss of detail.
The most important post processing parameter to “correct” is white balance (WB) and therefor color.
Do not neglect the significance of good sharpening techniques.
The most important Photoshop technique to master is how to work with Layers.
Partial editing can be of great help you maximize parts of the picture and dynamic range.
The two most powerful editing tools you should get to know well are levels and curves.
Do not under estimate the power of the quick editor in Elements it applies levels and curves in a clever way.




		Original picture - not edited

The above picture is a typical example of a picture taken in JPEG format. The camera was set on auto white balance and I typically use aperture priority. Are the colors correct and is the picture as sharp as it could be? Keep in mind that the higher the sharpness setting in the camera the more damage one can have to the picture. For this reason picture sharpness is reduced in most high end cameras. Many photographers prefer to sharpen their pictures during post processing.

In this article on Post Processing I will discuss:

How to do levels (Levels are used to correct white balance, exposure, saturation and contrast)
How to remove any further color casts.
How to fine tune contrast and saturation.
How to sharpen the picture.

The Histogram

Brief description of the technique I will discuss...

You most probably read all about the histogram and how important it is. In Picture 1 you will see I draw a big RED arrow pointing at the start of the histogram graph. Just above the histogram graph you can see the drop down menu and written in there is RGB. The typical technique people will use to correct contrast and color is to select each of the RGB colors from the drop down menu and to then move the sliders at the bottom of the histogram graph up to the point where the graph begins. When done, typically the color (WB) and contrast are OK. In extreme cases it does not work as well though..

The aim of the described procedure is therefore to correct the picture’s white balance (color) and in the process to correct as much as 85% of the picture’s contrast. Using the white, gray and black pens in the histogram adjustment window is highly accurate.

Typically RAW converters does not have all three pens resulting in a lets guess where the gray point is effort. That’s why the described method is often more accurate than RAW editing.

Step 1. Duplicate the file

The first step I normally do is to duplicate the file and to close the original file. This helps to prevent any damage to the original jpeg file. When the picture is open in the editor, go to the file menu and select duplicate, give the new file a name.

I only edit pictures I need. The rest I leave alone and will only edit or use them if they needed. That way I preserve my pictures plus I reduce the risk of loosing data.

Step 2. Adjust the color or White Balance

There are various ways of doing this like for example moving the sliders to where the histogram graph begins as indicated at point H. Typically one would do this for each of the RGB colors.
Recently I learned a much more accurate way of correcting the white balance using the three pens right below the “auto” button on the histogram window. Using the pens correctly also improves the saturation and contrast significantly.
Please keep in mind that the histogram is just another form of the “Curves” function.




		Picture 1

The question is, how does one know where to point those pens in the picture, or is it really a matter of experience?

The good news is, there are a very accurate way of determining where to point those pens to get the right results. To do that you need to go to the Layers Pallet. The black and white circle is the adjustment layers button.

Layers is the best way to edit pictures without destroying any picture data. As you can see you can do levels, contrast, saturation and all the other editing functions using adjustment layers.

Keep in mind the left pen will be pointed to the true black point in the picture, the most right pen to the true white point and the middle pen will be pointed to the true gray point in the picture.

How to determine the correct black point in the picture?

Select the adjustment layer pallet as in the above picture.
The move your curser down and select the Threshold adjustment layer.
You will see something like the picture on the right.
By moving the slider at the bottom of the histogram completely to the left and then slowly back to the middle, the first black areas that will appear in the picture will be the true black areas we looking for.
Move the slider back and forth until you find a clear spot in the picture you will be able to identify later.
When you find that point, press Cancel

How to determine the correct white point in the picture?

Repeat the above procedure, the only difference is, to find the white point move the slider to the right and instead of turning white the picture will go black.
Again move the slider left and right until you find a point you can clearly identify as the “white” points.

To determine the correct gray point in the picture see the article I wrote here.

Knowing where the correct white, black and gray points are in the picture, you can now select the adjust levels function and with each of the pens you can point to the areas you identified in the picture.

I also used the middle slider on the histogram window in figure 1, to reduce the exposure a little by sliding it to the right.

Practice will help you to perfect this procedure,

When the black, white and gray point is identified:

The above description was to determine exactly where to find the white, gray and black areas in the picture. With that knowledge we can do the following step:

Delete any the layers you created to help you find the white, gray and black points. (To delete, click on the layer and drag it to the waist bin)
Click on the adjustment layers button and select levels.
First click on the most left (black) pen in the histogram and then click on the black point in the picture. Repeat for the gray and white pens.
The above step will correct the picture’s white balance.
Next step is to move the middle slider below the histogram, red arrow in above picture, left or right to correct the picture’s brightness or exposure.


	Final picture, color adjusted or WB corrected

Final tip on the histogram.

It often happens that there still is a color cast after levels was done. One especially see it on skin tones. To correct that select from the blending drop down menu Luminosity. You can then adjust the Opacity to increase or decrease the effect. This technique normally works great.

Saturation and Contrast

Saturation and contrast are important adjustments and often they are also incorrectly used by photographers. I think important to remember is that a high level or general contrast and saturation has already happen with the above histogram adjustment. What is left to do is a fine tune setting or it could also happen that only part of the picture needs to be adjusted at this point. See the article on partial editing I wrote.

There are basically three methods of doing this final adjustment:

Do the adjustments on the picture with the Enhance, Adjust Lightning or Adjust Color drop down menus from the to menu. (Not the preferred method)
Go to the Adjustment Layer button and first create a Adjust Saturation layer and then a Adjust Contrast layer. (Preferred method)
One can also use the Curves function. Elements 5 and 6 has a basic curves function in the Enhance menu or one can use the Master Curves plug in.

Saturation

At this point I only adjust the saturation of a picture and not the Hue or the Lightness. The reason is that the Hue will typically adjust a color cast and that has been done with the histogram.

Lightness has been adjusted using the middle slider in the histogram.

Keep in mind that at any time one can go back and one can adjust one of the adjustment layers. All that’s required is to double click on the graph symbol on the adjustment layer and the adjustment window will appear.

How much saturation is a personal taste. Personally I will adjust between 7 and 17, depending on the condition of the picture.

Keep in mind that to much is not good.

Contrast:

Again click on the adjustment layer button and select the Adjust Lightness/Contrast option. Do not adjust the lightness, again its been done in a better way using the histogram.

Contrast adjustment is important and the following should be kept in mind. When sharpening is applied to a picture then all the software does is to increase the contrast at the edge areas in the picture.

Therefore if one apply a general contrast setting then it will have a sharpening effect on the picture. Its critical not to apply to much contrast. Typically I will use anything between 3 and 6 on the contrast slider.

The final step is to Sharpen the picture:

Sharpening is one of the most important steps during post processing. Most SLR’s will deliver “soft” pictures and often high end compacts too. The reason is that to much sharpening can permanently damage a picture. The photographers that therefor use these type cameras prefer to adjust sharpness themselves.

A next aspect to keep in mind is that there are a number of different techniques that photographers use to sharpen a picture with. For example with portrait photography one often sharpen the persons eyes separately with using the sharpening pen. Sharpening is such a important part of editing that one can write a complete article on that only.

In this article I will only discuss the Unsharp Mask technique.

The Unsharp Mask Window or Command:

Go to the Enhance Menu and right at the bottom select the Unsharp Mask option.

The following general points are important before we start:

The picture must be at 100% - this is critical for getting good sharpening results.
Different cameras will use different settings.
Different type pictures will have different settings.
Amount: Amount will determine how much sharpening
Radius: Radius determine how many pixels away from the edge will be effected by the sharpening command.
Threshold: Threshold determines how different pixels should be to be regarded as an edge - interesting is that the lower the number the more intense the sharpening and the higher the number number the lesser the effect.

I used “The PSE 5 Book” from Scott Kelby for info.

The table below has typical settings one will use for different type pictures. Keep in mind that different cameras will also require different sharpening settings. For example my G7 requires less sharpening than my E-330. Does the mean the E-330 is not a good camera? You will agree that the person who say that does not know better at this stage....The following table as I said is a guideline, you can fine tune it for your camera.

Type of Picture	Amount	Radius	Threshold
Soft Subjects: Like Max, the dog or the cat or the pillow, or hair	150%	1	10
Portraits: People, faces, close-ups	75%	2	3
Landscapes: More moderate type sharpening	225%	0.5	0
Buildings - Cities: More aggressive sharpening settings	65%	4	3
Web Pictures: Web pictures requires different settings.	400%	0.3	0


	Please keep in mind that the above settings are only a guideline and its good to develop your own set of settings based on your camera, in-camera settings and the lens you use.

Conclusion or final steps.

This basically concludes post processing. You can now save your picture. In the final step one can also decide to add a watermark to the picture or one can add a frame to the picture.

A post processed picture can also be used for further editing or for inclusion into other projects.

Siegfried