 |
| This is a super resolution image, creating by combining four photos in Photoshop. |
We recently tested out the Pentax K-1 II’s new hand-held Pixel Shift mode which combines four images to create a ‘super resolution’ file in-camera with better detail, dynamic range and lower noise. Sadly, it also results in some unwanted processing artifacts.
But you can also create a super resolution photo without using Dynamic Pixel Shift by shooting a series of handheld images and combining them in Photoshop. Super resolution works essentially by sampling a scene multiple times with slight shifts in framing, which allows details to be localized with sub-pixel precision (since shifts are unlikely to be perfect multiples of one pixel). The result is a file with improved resolution, less noise, more dynamic range but no artifacts. The best part is you can do this with the camera of your choosing. For the sake of this example, we did it with the Pentax K-1 II.
$ (document).ready(function() { ImageComparisonWidget({“containerId”:”reviewImageComparisonWidget-10980295″,”widgetId”:603,”initialStateId”:null}) })
Following a simple step-by-step Photoshop recipe (listed below), we created a super resolution file stacking four images – the same number used by Dynamic Pixel Shift – and one stacking 20 images, just for fun. We down-sampled the files to the original resolution (36MP). Right off the bat, the difference between our four image stack and a standalone Raw file is like night and day.
Similarly, the difference$ (document).ready(function() { $ (“#icl-4014–1567822084”).click(function() { ImageComparisonWidgetLink(4014); }); }) between our 4 image stack and Dynamic Pixel Shift mode is also substantial. Areas where Dynamic Pixel Shift displays artifacts$ (document).ready(function() { $ (“#icl-4010-939092746”).click(function() { ImageComparisonWidgetLink(4010); }); }) look clean in our 4-stack. This is observable throughout our sample scene$ (document).ready(function() { $ (“#icl-4011–1901282620”).click(function() { ImageComparisonWidgetLink(4011); }); }). Interestingly, the difference between the 4 image and 20 image super resolution examples is less noticeable$ (document).ready(function() { $ (“#icl-4012–1917693503”).click(function() { ImageComparisonWidgetLink(4012); }); }). While there is some advantage$ (document).ready(function() { $ (“#icl-4013–396364473”).click(function() { ImageComparisonWidgetLink(4013); }); }) to stacking more images, returns are diminishing.
You’ll notice we’ve included two versions of our 4-stack and 20-stack: Median and Average, which refer to the stacking method used in Photoshop (described in detail below). Overall, the median method handles ghosting from moving objects$ (document).ready(function() { $ (“#icl-4015-104250703”).click(function() { ImageComparisonWidgetLink(4015); }); }) better than the averaging method.
Methodology
There are numerous tutorials providing instructions for creating a super resolution image in Photoshop – this one by Ian Norman on PetaPixel is among our favorites. Distilled down to its simplest terms, there are four easy steps:
- Bring all images into Photoshop as a stack of layers
- Resize the image to 200% width and 200% height using ‘Nearest Neighbor’
- Auto align all the layers
- Average the layers by setting each layer’s opacity to 1/layer number (the 1st layer will be 1/1 so 100% opacity, the 2nd layer will be 1/2 so 50% opacity, and the 4th layer will be 1/4 or 25% opacity, and so on).
- Sharpen the image using a Radius setting of 2, and a suitable Amount setting (we used 200% for the 4 image stack and 300% for the 20 image stack – the more images you stack the more amenable the composite will be to aggressive sharpening)
Alternatively, for the fourth step you can convert all layers to a ‘Smart Object’ and change the stacking mode to ‘Median’. This can help deal with ghosting from movement in your final image, but can also take longer to process.
Finally, you can resize the final output by 50% width and height (we prefer Bicubic resampling for this step) to get the shot back to its original resolution, but with far more detail and cleaner output. Or, you can opt to save the high-resolution file if you print big, but just keep in mind that for a 36MP camera, that’s a 144MP file. You can always re-upscale a super resolution file you’ve shrunk, and if you use the ‘Preserve Details 2.0’ resampling method in Photoshop to do so, the results are often impressive and hard to distinguish from the higher resolution super resolution file.
Takeaway
You don’t need any particularly special camera to generate images that look like they were taken with a higher resolution, larger sensor camera. Just use the technique outlined here or in Ian’s article.
And if you’re shooting landscapes and cityscapes, you likely already have multiple photos of the same composition captured with changing light. Chances are that due to the wind, natural vibrations, etc., the shots have at least some sub-pixel movement between them (you can always gently nudge your camera between exposures to ensure there’s at least some shift). So why not go back through your library and take advantage of super resolution?
Articles: Digital Photography Review (dpreview.com)
