Pixel shift post-processing

I have recently been doing a series using pixel shift to capture as much detail and texture as possible but was disappointed with the lack of sharpness I was getting using Lightroom - no matter what adjustments I applied it just wasn't there. This is weird because it hasn't been a problem before so I just did basic settings in Lightroom, set the sharpness to zero, and then tried the oddly named Unsharp Mask in Photoshop and bingo! It works a treat. Here are some examples:

I think that Rawtherapee does a better job with pixelshift than the Adobe programs..

Davek

---

My PBase: www.pbase.com/davekitson

To be honest it's impossible to see if they are sharp because of the size.

Pixel shift is something I have always meant to try but never got round to as my preferred software (DXO Photolab) doesn't process it.

---

. My Flickr

Fair enough, here are crops from each which I hope will show how much detail is captured.







They were all taken on a tripod with a DFA 24-70mm at the longer end and aperture f8 to overcome the difference in distances between the centre and edges of the frame, in live view and a 2 sec. delay on the shutter.

Don't generally use pixel shift as I'm unsure if Lightroom can process them?

I bought my K1 used from Wex I think from memory and it didn't have the Silkypix CD

---

Gareth
Welsh Photographer

My outfit: K1 gripped - K3ii - two Z-1P - Pentax D FA 24-70mm - Sigma 70-200mm OS HSM - Pentax modified DA* 60-250mm f4 - Irix 15mm Firefly - Pentax FA 35mm - FA 50mm - Sigma EX 20mm - FA 28-70mm f4 - Tamron SP 90mm macro - Pentax AF 540 FGZ II

My Flickr
My PPG
Foundation NFT

Gareth, I find pixel shift works fine normally in Lightroom but for some reason the ones I took above wouldn't sharpen. There could be a couple of factors, I was using autofocus in live-view in rather poor light and the exposure times were between 1/4 to 1/10 sec so maybe a tiny bit of wobble although they look fine on the back of the camera. The Unsharp Mask in Photoshop has some sort of smart function which I think compensates for whatever is causing blur because there are other menu options to correct blur caused by camera movement and lens issues.

If you are shooting a static object like a building or still life it is definitely worth giving it a try, use a sharp lens, get everything right and the detail is amazing.

Thanks Rob,
I'll give it a go

---

Gareth
Welsh Photographer

My outfit: K1 gripped - K3ii - two Z-1P - Pentax D FA 24-70mm - Sigma 70-200mm OS HSM - Pentax modified DA* 60-250mm f4 - Irix 15mm Firefly - Pentax FA 35mm - FA 50mm - Sigma EX 20mm - FA 28-70mm f4 - Tamron SP 90mm macro - Pentax AF 540 FGZ II

My Flickr
My PPG
Foundation NFT

This is a straight reaction to looking at the images: they appear to have the most amazing textures - almost every grain of sand and tiniest bit of moss. I would expect pixel shift to produce this - detail to the 'n'th degree! However, is all that microscopic detail actually swamping the major detail of the entire gravestones? It's like looking closely at a pointillist painting. All you see is dots, until you step way back. Have you any non-pixelshift images to compare them with? Of course, I could be talking complete rubbish. It wouldn't be the first time!

---

Both the *istDS and the K5 are incurably addicted to old glass

My page on Photocrowd - link

Fair point David, these are part of a series exploring how time and the weather are gradually wiping away the lettering, in an analogy of the loss in human memory. The subject matter I want to focus on is the texture, lichens and other growth rather than the full details of the person interred which are often illegible anyway, and with the idea that these could be printed quite large.

Is there any in-camera facility for processing pixel-shift files, even if it is just assembly into a single initial image?

---

Regards,
Kea828

The four images are automatically combined into a single file in the camera which you process like a standard image in appropriate software. I believe where other manufacturers have a similar feature they are retained as separate images which have to be combined afterwards in special software (if you can identify them amongst the photos on your card).

The four images are automatically combined into a single file in the camera which you process like a standard image in appropriate software. I believe where other manufacturers have a similar feature they are retained as separate images which have to be combined afterwards in special software (if you can identify them amongst the photos on your card).[/quote]

RobL,
thanks for the clarification.

---

Regards,
Kea828