In portraits this technique can be used to reduce unwanted shadows, or deepen them for a more dramatic look. In landscape and architectural photography it can create more dramatic, dynamic images. There are several different ways to Dodge and Burn and I’m planning on exploring them further, but lets start with the basics.
Dodging and Burning in Photoshop: some folks will dodge and burn right on the image. This is considered “destructive” editing because it changes the actual file. If there’s just a little bit of work to do, it can be a quick way to get it done. However, one of the problems with dodging and burning is that you’re essentially drawing black and white lines of varying widths and densities on your image, and these can make larger areas look kinda gray and muddy. A non-destructive way which is a little more subtle, is to create a separate layer:
1. Create a New Layer
2. Fill it with 50% gray: Hold down the Shift key and hit f5 and the box will appear (another way is to go to Edit, then Fill). Select 50% Gray from the drop down menu. Now you can’t see anything but gray.
3. Set your Blend More to Soft Light or Overlay. Now you can see your image just as it was before.
4. Select you Brush tool. Select a low Density or a low Flow (see below for more info)
5. Paint on the gray layer using white to lighten, and black to darken.
6. Dodge and Burn your image.
It’s important to keep your brush tool on either a low density, or a low flow. By low I mean around 10%. If you set your density to 10%, then the opacity of your painting will never go above 10%. If you set your flow to 10%, it means your brush will paint 10% with each pass. I typically set my Density to about 25%, and my Flow to about 5%. This means I’ll apply 5% with each stroke, up to a total of 25% but no greater.
How to Dodge and Burn an image is too complicated and visual to explain in words. Do a Youtube search for it. Keep in mind that you can Dodge and Burn to reduce unwanted shadows, but also to increase shadows and highlights for a more dramatic image.