• site home
  • blog home
  • galleries
  • contact
  • underwater
  • the bleeding edge

the last word

Photography meets digital computer technology. Photography wins -- most of the time.

You are here: Home / The Last Word / NEX-7– purple corners

NEX-7– purple corners

January 2, 2012 JimK 3 Comments

Near the start of my NEX-7 postings, a reader asked me if I saw purple corners when the camera is used with wide angle lenses. I replied that I had noticed the effect with the 24 mm Elmar. I’ve had a chance to do some more reading and some more experimentation, and here’s what I’ve found.

The effect is real. It is related to the angle of the focused light to the sensor. In the middle of the image, no matter what the focal length of the lens, that angle is 90°. As you move toward the edges of the image, and especially the corners, the angle becomes more acute. It’s well known in film cameras that this changing angle produces images that get darker towards the corners. Some manufacturers of short focal length view camera lenses have produced filters to place in front of the lens to cancel out this effect.

In digital cameras, not only does the image get darker toward the corners, there are color shifts, usually towards magenta. Leica, in their M9, uses both angled micro-lenses on the sensor and lens specific processing in the camera to correct this effect.

The purple corner effect gets worse as the nodal point of the lens gets closer to the sensor plane. This means the wide angle lenses are more affected than longer focal lengths, and that lenses made for rangefinder cameras are more affected than those made for SLR’s, which have designs that move the nodal point further away from the sensor plane to keep the rear elements from interfering with the swinging mirror. To a small extent (unless we’re talking about macro lenses), the effect is worse when the lenses focused on infinity, and appears to be very slightly worse when the lens is stopped down.

Here’s a picture taken with the NEX-7 and the Leica f/3.8 24 mm Elmar M ASPH. Both the lens falloff and the purple-corner effect are quite visible. For reasons that are unclear to me, the right side of the image is more affected than the left.

Here’s a similar picture taken with the NEX-7 and the 16mm Sony lens.

All else being equal, you would expect more light falloff toward the corners and the corners to be more tinted purple than with the 24 mm lens. This doesn’t happen. There is slight light falloff, and essentially no purple-corner effect. Sony must be doing some processing in the camera that affects even supposedly raw files. That’s nice for people who are happy with the quality of the Sony 16mm lens, but it doesn’t help those of us who are looking for lens quality commensurate with the NEX-7 sensor quality.

I investigated what can be done about fixing images like the above in post-processing. I picked a day when the sky was flat overcast and took some pictures of the sky with the NEX-7 and the Leica 24 mm Elmar.

Here’s an example of what happens when you place the sky on Zone V:

I made exposures at various apertures and various Zone placements, analyzing the results in Photoshop. The amount of light falloff toward the corners as measured in CIELab varied with Zone placement, but the color shift as measured by the a* and b*values was remarkably consistent. From the center to the upper right corner, where the worst color shifts occur, a* went from 0 to 10, and b* went from zero to -6.

That led me to a way to make corrections to real images: create a correction image in the CIELab color space to be subtracted from a CIELab version of the image to be corrected. While mathematically fairly pure, I doubted that I’d actually use such a correction image, since I don’t normally work in CIELab. Instead, I tried to come up with an approximation which would allow me to stay in ProPhoto RGB.

I took the Zone V image of the sky and applied a Gaussian blur with 100 pixel radius to get rid of any fine structure that might exist in the image. I then inverted it, and, by moving the black point slider in the Levels control, brought the black point up to the darkest value in the image. I saved that image for further use as the corrector image. The corrector image looks like this:

I tested the corrector image by pasting it is a layer over the original sky image, setting the blending mode to “Linear Dodge”, and adjusting the opacity until the corrected image was consistent across the whole screen.

I then tried the correction process with a real image and got this:

Too flat. I like some light falloff towards the edges of wide angle pictures. I added some:

The good news is the purple corners can be fixed in post processing. The bad news is it takes about a minute an image to apply the fix (this could be improved greatly by creating a Photoshop Action), and you have to leave Lightroom to do it. There may be other Photoshop blending modes that work better.

I couldn’t figure out how to make the corrections with Lightroom’s Camera Calibration tool, but there may be a way to do that. It’s also worth looking at DxO. I could have spent more effort on getting a picture of something that’s really flat white to serve as a basis for the correction process; that’s not as easy as it sounds. I’m suspicious of the accuracy of my quick and dirty sky picture, especially since it’s darker towards the top. It worked well enough, though.

 

The Last Word

← Sony NEX-7 noise NEX-7 — test images for corner color correction →

Comments

  1. M. Jean Prussen says

    January 3, 2012 at 12:43 am

    Hello, i had a look at the Elmar 3,8/24 and noticed that it is probably a non-retrofocus lens. I would therefore assume that by using for instance a Summilux 1,4/24 the problem with the magenta corners would be much less.

    Greetings from Vienna

    Jean Prussen

    Reply
  2. Jim says

    January 3, 2012 at 8:18 am

    Jean,

    Thanks for the comment. I think you’re probably right. I don’t have the Summilux f/1.4 24mm lens, and I don’t think I want to spend $6500 to find out if it fixes the purple corners. One thing that occurs to me about the problem. Since it’s moderately easy to fix it in post-processing, once it’s bad enough to need fixing, it doesn’t matter how bad it is. So, if the Summilux only makes the problem a bit better, it doesn’t do me much good.

    On the other hand, the Summilux could be a better lens on the NEX-7 than on the M9, where it’s got to be almost impossible to get it critically sharp at full aperture. I know I have trouble with my Elmarit 24mm f/2.8 on my M9.

    If I can figure out a way to get uniform test image illumination, I’ll do some tests on other lenses.

    I’ve only been in Vienna to change trains on a trip from Berlin to Budapest. I regret that I didn’t get off an stay a few days…

    Jim

    Reply
  3. Rich says

    January 3, 2012 at 8:38 am

    Google Cornerfix. It fixes the color shift and vignetting in one step. It works on dng files. Easier than processing in PS or LR!

    Reply

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

May 2025
S M T W T F S
 123
45678910
11121314151617
18192021222324
25262728293031
« Apr    

Articles

  • About
    • Patents and papers about color
    • Who am I?
  • How to…
    • Backing up photographic images
    • How to change email providers
    • How to shoot slanted edge images for me
  • Lens screening testing
    • Equipment and Software
    • Examples
      • Bad and OK 200-600 at 600
      • Excellent 180-400 zoom
      • Fair 14-30mm zoom
      • Good 100-200 mm MF zoom
      • Good 100-400 zoom
      • Good 100mm lens on P1 P45+
      • Good 120mm MF lens
      • Good 18mm FF lens
      • Good 24-105 mm FF lens
      • Good 24-70 FF zoom
      • Good 35 mm FF lens
      • Good 35-70 MF lens
      • Good 60 mm lens on IQ3-100
      • Good 63 mm MF lens
      • Good 65 mm FF lens
      • Good 85 mm FF lens
      • Good and bad 25mm FF lenses
      • Good zoom at 24 mm
      • Marginal 18mm lens
      • Marginal 35mm FF lens
      • Mildly problematic 55 mm FF lens
      • OK 16-35mm zoom
      • OK 60mm lens on P1 P45+
      • OK Sony 600mm f/4
      • Pretty good 16-35 FF zoom
      • Pretty good 90mm FF lens
      • Problematic 400 mm FF lens
      • Tilted 20 mm f/1.8 FF lens
      • Tilted 30 mm MF lens
      • Tilted 50 mm FF lens
      • Two 15mm FF lenses
    • Found a problem – now what?
    • Goals for this test
    • Minimum target distances
      • MFT
      • APS-C
      • Full frame
      • Small medium format
    • Printable Siemens Star targets
    • Target size on sensor
      • MFT
      • APS-C
      • Full frame
      • Small medium format
    • Test instructions — postproduction
    • Test instructions — reading the images
    • Test instructions – capture
    • Theory of the test
    • What’s wrong with conventional lens screening?
  • Previsualization heresy
  • Privacy Policy
  • Recommended photographic web sites
  • Using in-camera histograms for ETTR
    • Acknowledgments
    • Why ETTR?
    • Normal in-camera histograms
    • Image processing for in-camera histograms
    • Making the in-camera histogram closely represent the raw histogram
    • Shortcuts to UniWB
    • Preparing for monitor-based UniWB
    • A one-step UniWB procedure
    • The math behind the one-step method
    • Iteration using Newton’s Method

Category List

Recent Comments

  • JimK on Goldilocks and the three flashes
  • DC Wedding Photographer on Goldilocks and the three flashes
  • Wedding Photographer in DC on The 16-Bit Fallacy: Why More Isn’t Always Better in Medium Format Cameras
  • JimK on Fujifilm GFX 100S II precision
  • Renjie Zhu on Fujifilm GFX 100S II precision
  • JimK on Fuji 20-35/4 landscape field curvature at 23mm vs 23/4 GF
  • Ivo de Man on Fuji 20-35/4 landscape field curvature at 23mm vs 23/4 GF
  • JimK on Fuji 20-35/4 landscape field curvature at 23mm vs 23/4 GF
  • JimK on Fuji 20-35/4 landscape field curvature at 23mm vs 23/4 GF
  • Ivo de Man on Fuji 20-35/4 landscape field curvature at 23mm vs 23/4 GF

Archives

Copyright © 2025 · Daily Dish Pro On Genesis Framework · WordPress · Log in

Unless otherwise noted, all images copyright Jim Kasson.