Sunday, February 16, 2014

Digital Workflow 3 - Black and White Conversion

This is a continuation of my Digital Workflow series. In my previous posts, I talked about camera settings and the importance of white balances. In this article, I will discuss one of the optional approaches to IR post-processing – black and white (BW) conversions. 

Initial Photograph without (left) and with (right) the LifePixel
White Balance.  Nikon D90 830 nm converstion
My initial motivation for shooting IR photographs was to use them as an entrĂ©e into black and white (BW) photography. I was enamored with the sharp contrasty photos and the ability to create surreal images.  These effects can be obtained with any IR conversion but if you are a dedicated BW photographer, the 830nm Deep Infrared conversion seems to be the ideal solution.  With this conversion, you will get BW images right out of the camera if you use the LifePixel white balance preset.  However, this filter will produce extremely red images if you do not use the preset WB.  

Cropped image without any post-processing.
Here is the initial cropped image. Right out of the camera, these images usually lack contrast and appear to be visually flat. As I mentioned in my Camera Settings post, I set my Picture Controls to +1 contrast, Vivid, and Sharpening at 6. This image was brought into Nikon Capture NX2 (CNX2), cropped, and saved as a TIFF file.  The camera settings are applied during this process.  

Autotone in Photoshop
The first step in post-processing is to use the Autotone function in Photoshop or the Auto Levels function in CNX2  

Image > Autotone (Photoshop)
Adjust > Light > Auto Levels (CNX2)

This step makes a big difference but the clouds and some of the grasses still don't have much detail.

One process I find extremely useful when processing Deep Infrared photos in Photoshop is to increase the microcontrast using the unsharp mask functions.  The Detail Extractor filter in NIK Color Efex Pro 4 the Clarity brush in Lightroom will produce similar results with various levels of post-processing control. I happen to like this approach because it works with my workflow.

For increasing microcontrast:
   PhotoShop Unsharp Mask:
      Amount — 5-20%
      Radius — 30-100 pixels (smaller radius enhances smaller scale detail)
     Threshold — 0

Unsharp Mask 20% and radius 30
My normal workflow is to duplicate the background layer and apply an unsharp mask at 20% Sharpening and a Radius of 50 pixels. If the scene starts to look too contrasty, I adjust the opacity of the layer to make the image look right. If I want to increase the local contrast in certain areas and not others, I change the duplicate layer to a layer mask and paint in the local contrast with a black brush. When processing a color IR photo, I set the blending mode to “Luminosity” to prevent local color saturation changes.

Choosing the proper radius is the key to this approach. High resolution images or those where light-dark transitions are large, require a larger radius value.  Very low resolution images may require a radius less than 30 pixels to achieve the effect. 

This process can produce noise in some images so this is a good time to do a noise reduction treatment.

Adjust Levels, noise reduction, and sharpening.
This final image is after adjusting the midtone levels to 0.81 to darken the sky and adding 10% Smart Sharpening.  

Processing of color IR photos is very straightforward because you have more color information in the image.  The Photoshop, CNX2, and NIK SilverEfex Pro 2 software will do a nice job with the conversion. Many people increase the saturation of the images before BW conversion but I have not found that necessary with a properly white-balanced image.

No comments:

Post a Comment