With the October Contest underway where the winner will walk away with a fully licensed copy of Noise Ninja, courtesy of the folks at PictureCode, today is a great day to give everyone an idea of what makes this program one of the industry leaders. Since I have talked about the presence of noise in your images before (read that post here), re-hashing the principles of what introduces noise and the artistic benefits and image quality downsides of it in your images isn’t really needed.
Instead, here I will simply share some of the interface, and a few before and after shots here so you can get an idea of how well this software handles noise in your images. So, enjoy!
First up (after entering your license code), you’ll have a screen that looks like this:
You can see the yellow squares in the image on the right that shows the sections scanned to determine the noise profile for the image selected. On the right, a closer view of what the image will look like (or at least a close approximation) after the noise reduction is run. You see how you can make certain adjustments on one of several tabs. On this main tab, you can see the options for Image Profile, Profile Chart and radio buttons for luminance and color values, along with the reset option. I actually leave these alone most of the time.
On the second tab is where I vary my noise reduction adjustments (and I would venture to guess most people make their valuation adjustments based on the image used). Here’s the options available to you:
You can see here that many more detailed adjustments can be made to the strength, smoothness, contrast, USM amount, radius, color noise, and then checking whether you want the processing done quickly (turbo) and applying the settings to coarse noise situations. I made my adjustments to 12 for the strength, smoothness and contrast (and typically use these ones for most images with noise).
In the next tab, as shown below, you can see the noise brush, where you can apply the noise on a paint mask or an erase mask, specify the channel (luminance, chroma, or all) ad the quality of the brush trail by not only the type of edge, but also by radius and strength. (This is very handy if you want the noise reduction faded across the image so as to leave some noise behind for artistic effect! Normally I don’t make adjustments here either – but it is fun to play!) Take a look at the interface:
The Actions tab, as shown next, defines how Noise Ninja operates when it’s loaded. Specifically, here is where you can tell NN wither the profile is by default configured based on EXIF data or to create a profile from the image, or to use a specific profile. This can be very handy as well if you are batch processing a lot of photos that were shot under the same or similar circumstances so that a profile doesn’t have to be loaded each and every time. The fallback, and filter policies can also be defined here as well. Here typically I leave the default settings alone as I prefer not to batch profess since my type of work is more geared toward fine art and individual edits rather than massive quantities. To each their own I guess – take a look at the interface:
The next tab is the annotations tab. I almost never use this, as this is a place where you can add additional EXIF and metadata. Since I prefer to take the values from camera rather than adding my own specific notations, this tab doesn’t get any use from me – but it can be handy if you want to vary settings based on things like manufacturer, Model, FIlm type Megapixel count, ISO level, etc. A second tab lets you add even more…see the interface captures below for reference. Noise Ninja to me is isn’t about editing exif data though – it’s about noise reduction, so I skip these tabs. Take your pick:
I do occasionally reference the Image Info tab though, because some useful information can be obtained there as well. Things like the manufacturer, model, and more specifically, the ISO setting at which the image was taken. ISO settings can be an indication on the amount of noise that was introduced due to camera settings and thus, how much can be reasonably applied.
You can see that my first image, shot at ISO 3200, was subject to some pretty intense noise on the 40D. So, I ran the process, and take a look at the before and after results:
Some pretty heavy noise, as expected, so let’s just look at the image after processing:
See the difference in grain? Yeah, I was impressed too. But is it always the case? I ran another couple images through NN and got the same or similar results on other objects in my print box. (Which, by the way, is a great resource for test shots when looking into lens and gear performance on a wide variety of things like noise, white balance, macro, depth of field, and much more.) Here’s another sample crop just to show you for comparison sake:
Again, some impressive results. When push comes to shove, results is where it counts in my book and PictureCode delivers at a superb value.
Now granted, there are other noise handling programs out there, and if you ask a dozen people what they use, 5 will say one, 5 will say another, and 2 will offer up another alternative. When push comes to shove, it really is up to personal preference on which one “works best”. For me, it doesn’t really matter as long as you are aware of what noise is, what it can do to an image, and how to fix it. The tool you use is up to you. I think Noise Ninja is probably recognized well enough for a reason, and the results here seem to suggest why.
The rest I’ll leave up to personal preferences, and if you would like to contribute your own thoughts, by all means, share them here in the comments or with me via email. If you want to try it – you can download a free copy from the folks at PictureCode here – or simply participate in the October Contest over at Flickr and win a copy from them for free!
That’s it for today – happy shooting all, and we’ll see you back here again tomorrow!