My Post Processing Work Flow
On the left is the out of the camera RAW without any processing, on the right my final file.
Since I get asked quite frequently, how I process my images and what my “secrets“ are, I thought I share my workflow in a detailed tutorial.
I chose and image of the critical endangered Regent Honeyeater. Firstly, because it is a stunning bird and secondly, because it offers a few challenges.
The conditions when I took the picture were quite challenging and I did not use a fill flash, in order to not risk spooking the bird. I ended up with a slightly too dark picture. It also provides challenges, because the background is quite bright and the bird has strong contrasts as well.
For browsing through my images I use FastStone Image Viewer. I came across this program on one of my workshops, where my friend Maxis introduced it to me. I was instantly hooked and find it to be the best program by far to browse through your images. It is amazingly fast and you can check your images for sharpness @ 100% simply by holding down the mouse button. Using FastStone has reduced the time I need to go through my images and delete the bad ones dramatically. The best thing about it…..it’s FREE! You can download it here: http://www.faststone.org/FSViewerDetail.htm
After selecting the file I want to process, I have to simply press “E” and Photoshop opens, where I can use ACR to convert my RAW file.
As you can see in the images above, the picture is a bit dark, but because the bird has such a bright back, I could not expose any lighter, because I would have lost the detail on the back otherwise.
Once we are in Photoshop (I use CS5), I just adjust a few things. As picture style mode I chose “landscape” for this image. This is a style I only rarely use, but for this image, it offered the best colors. I mostly use “neutral” or “camera standard”, however, there’s no right or wrong here and it mostly depends on the image you are processing, what you want to use. If the image contains a lot of red, the choice will be harder, because most of the styles will burn the reds. You can always go with neutral and adjust the image in Photoshop afterwards, but I prefer to use whatever I think gives me the best base to start off with.
You can see my adjustments in the images below. I like to use “vibrance”, but I absolutely hate “clarity”. In my opinion clarity makes the images dark, dull and grey and I stay away from it.
Once I am satisfied, I click “open”. Make sure that when you adjust the image, the histogram on top is okay and everything stays within the borders. You don’t want to leave a bit of space to either edge.
The first step in Photoshop is cropping. I use the crop with a preset 2×3 ratio.
This images is actually a bit of a bigger crop, but the files of the Mark IV easily hold up and you can print a file like this still quite large. Once I am happy with my crop, I adjust the histogram.
I use the short cut “CTRL + L” to open the box. By holding down the “Alt” key when I move the slides, I can see that nothing gets burned out or clocked up.
The next thing I do is make a selection of the background and saved it. I do this by selecting the background with the lasso tool, which is very easy in this case here. Once I have selected it, I create a new empty layer, by clicking on the “create a new layer” icon on the bottom right of the layers panel. Then I go to “Edit” > “Fill” > “Black” and click ok. Now I make the layer invisible by clicking on the “eye” icon next to the layer. Then I deselect it. Whenever you need the selection you your background now, you simply hold down “Ctrl” and click on that back layer with your mouse and the background will be selected again.
This selection will proof helpful when we process the image later on.
The next step I did is increase the saturation. Since this image is quite dull and I like my images to pop, I gave it quite a decent boost of +18. Be aware and check your histogram afterwards, sometimes such a big boost will be too much. If that’s the case, adjust accordingly and check again. We don’t want any color leaving the edge of the histogram.
What I do afterwards is getting rid of unwanted colors. Canon has a tendency to put a magenta cast and magenta “bits” on images, especially when shot under overcast conditions. I counter that by selecting magenta in the color panel and reduce it almost completely. On this picture I did the same with blue and cyan, since both negatively affected the blacks and the perch.
The next thing I didn’t like about this image was the color of the perch. It has such a beautiful grey color, which I wanted to enhance. I did so b de-saturating it completely. To get a quick and easy selection of the perch, I held down “Ctrl” and clicked onto our saved background selection layer. Then I clicked on the quick mask button in the panel on the left and used the “brush” tool to deselect the bird. Once only the perch is red, I clicked on the quick mask button again, to leave the quick mask mode and ended up with a nice selection of the perch. All I needed to do then is to reduce saturation by about 80 or more, until I liked how the perch looked.
The next step was one I don’t do very often, but here I saw a few spots on the bird that I didn’t like. To get rid of some of the white feathers on the head and the dirt on the beak, I used the “healing brush” and the “clone” tool. See the before/after below to see what I did. It was not much, but improves the image in my opinion.
While I did that I also noticed that the image would be greatly improved, if I lightened the eye and the warty area around it. To lighten it, I just used the “doge tool” @ 8%.
In the two images you can see the instant impact this little adjustment has on the overall image.
Next I wanted to make the image “pop” a bit more. To do so, I ran several curve adjustments. Firstly I ran a little “s-curve” on the whole image.
Afterwards I was still not happy, but because I did not want to change the background anymore, I selected the background selection layer again, but selected the inverse this time, so I was working only on the bird. To get it right. I ran a slight highlights adjustment. This darkened the bright areas on the bird’s back.
After I adjusted that, I ran another curves adjustment, but only on the bird and the perch.
To give the image the final touches, I used the “burn” tool to darken the bright areas on the back and to darken the back on the head a bit.
The image is not very noisy at all, but I decided to de-noise the background nevertheless. To do so, I selected the background selection on last time, but click onto it, while holding down “Ctrl”. To de-noise my images, I use Topaz DeNoise 5. In the Topaz add-in, I just select the RAW- medium setting and let Topaz do its magic.
Now we are basically done. All I do to post the image on the web, is let my sharpening action run and save it J
Some people might be shocked, that I only use little to no layers, but I only use them when I have to. I know that there are many advantages of using layers, but this workflow as described above is what I do most of the time.
And here’s the final image
The Perfect Web Image
I often receive comments that my web images look so perfectly sharp and well processed and I thought this blog is a good way to share my web image editing process. For all of you who think I sit in front of my images for hours and process them heavily, I have to say that’s not the case. When I prepare an image for my website it usually takes 2-3 minutes. I start in Lightroom2 , where I correct the exposure and WB if needed and then I move the image to Photoshop. This post does not intend to change your post processing or has any intention to become a guideline. It’s just a description of how I process my images.
In Photoshop I adjust the levels, add saturation and add contrast. Then I select the background and use some gaussion blurr on it. Sometimes I also use the internal PS Noise reduction. That was basically it. So far I think I do nothing special and most of you do it likes this.
The important and different part comes now . Instead of resizing my image in one step and sharpen it at the end. I resize my images in 500 pixel steps and use an USM at every of those steps.
I describe my technique for a Mark IIn Image. It has a width of 3500 pixel on the long side.
My initial step is to sharpen it heavly and strength 120 and radius 1. (120/1) Then I resize it to 3000 pixel on the long side. In the next step I sharpen it a little less. 80 strength and radius 1. Then I proceed as the following:
resize to 2500 – sharpen 90/0,6
resize to 2000 – sharpen 60/0,4
resize to 1500 – sharpen 80/0,4
If you use a different camera or file size you might to change these values slightly or add more/less steps. Just play around a bit and you will find your perfect fit.
I then resize it to my designated web size, which is usually 999 or 750 pixel. If I resize them down to 600 pixel I sometimes add one more sharpening step or use the sharpening brush in Photoshop, if I find some parts too soft. To make things easier I created a PS action, which also adds my signature and a small frame. The whole process may take 5 seconds.
Sometimes you get some dark or bright halos around the bird. I get rid of them by using the softening tool(what’s the right name of it?) with a strength of 85% and a pixel size of 2. And go over them and they disappear.
I add two samples one with my method and one sized down and sharpened in one step.
hmmm it’s hard to see a difference with these small images, but with larger files you see a clear difference. However I think also here you see more details in the first one…
What do you guys think? How do you prepare your images?