Jump to content
msperl

HDR: different results for same image when using different types

Recommended Posts

While playing with an applescript workflow that:

* exports the selected images from aperture (as 16 bit tiff)

* triggers "merge HDR" in Affinity Photo

* exports/saves file again

* imports those back into aperture

 

i did a lot of tests using one of my HDR panoramas (from a long time ago with a camera that only supported JPG at that time - so the source is not ideal).

 

One of the observations was that some of the generated HDR images are "cut of at wrong levels" (I guess black/white points). When I did look at the individual sources in the HDR workflow pane I see that those "cut of at wrong levels" happen at the "source" level already.

 

The end result is heavy solarization in the resulting image.

 

Strangely this did not happen for all the "component" images of my panorama, so I did dig into the HDR process some more and have done the following

* JPG source directly - solarization (heavy)

* 16 bit TIFF exported via Aperture - solarization (better than from raw JPG)

* 16 bit tiff exported via Aperture, resaved in Affinity Photo as 32 bit HDR - no solarization

* 16 bit tiff exported via Aperture, "converted" to 16 bit tiff using ImageMagik - no solarization

* JPG source, "converted" to 16 bit tiff using ImageMagik - no solarization

* 16 bit psd exported via Aperture - not accepted as source format for unknown reasons

 

So it seems as if there is some import issue for the sources depending on the input file format.

 

Note that I am running 1.5.1

 

Any Ideas?
 

If not then I guess I will need to add an additional "transformation" step in my workflow using ImageMagik "convert <src>.tiff -depth 16 <dest>.tiff"

 

Thanks, Martin

Share this post


Link to post
Share on other sites

This seems related, but it seems as if the input format of the HDR images also plays a major factor.

 

I have attached the 3 jpg that I have used.

Import them into aperture and then export them as 16 bit tiff and you should also see the "behaviour" that I am seeing and you should be able to reproduce the situation that I was describing.

* post-49796-0-29054800-1485108058_thumb.jpg
* post-49796-0-50566900-1485108065_thumb.jpg
* post-49796-0-62713300-1485108072_thumb.jpg

Share this post


Link to post
Share on other sites

Here also for comparison the window of the image including the histogram:

  • DSC06595.JPG loaded into AP on its own:
    post-49796-0-31495500-1485108806_thumb.jpg
  • the same image selected in Sources during the HDR merging:
    post-49796-0-33488900-1485108805_thumb.jpg

The problem is immediately apparent when comparing those two images - I would expect both of them to look identical.

Share this post


Link to post
Share on other sites

Note that I get the same effects when using Canon EOS 5Dm3 raw images as sources (exported via aperture as 16 bit tiff).

 

The following observations:

* no artifacts:

  * raw CR2
  * 16-bit exported tiff from aperture exported (with AP) as 32 bit tiffs

  * raw CR2 developed and exported (with AP) as 32 bit tiffs

* shows solarization:

  * 16-bit exported tiff from aperture
  * raw CR2 developed and exported (with AP) as 16 bit tiffs

 

Any ideas?

 

Is it possible that there is an implicit "develop" step in the HDR process that may trigger this? (I wonder because .afphoto files can not get used as a "source" file format)

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×