Jump to content

noise QC failure


Recommended Posts

Just had a rejection on grounds of noise from a 100 ISO image, taken on a Canon EOS 5M, lightly processed. Looks fine to me at 100% - anyone else noticed a shift ( inexplicable or otherwise) in noise rejection policy?

Link to comment
Share on other sites

No, but its unusual at ISO 100. You could post the image here at 100% size for others to have a look. Equally you can contact alamy directly and ask them to have a look. QC are generally pretty good, but everyone can make a mistake.

Link to comment
Share on other sites

Post a full size JPEG image for an opinion if you like. If the image was underexposed or you opened up deep shadows then an ISO 100 image could show noise although given that you say lightly processed it seems unlikely. Dropbox is good for posting example pics as it also allows posting of raw images. 

Link to comment
Share on other sites

I can't see any significant noise at all. You have used moderate dehaze and a little clarity which does increase noise in general (together with the older and mild default LR sharpening) but I would regard the noise here as insignificant as far as QC is inconcerned. The only thing I can see that could possibly make a QC failure is the foreground is a bit soft and the bush is very blurred (out of focus and wind blow?) so if anything it could be failed for SoLD but not for noise. The main subject is in focus though so SoLD would seem harsh here. 

Link to comment
Share on other sites

1 minute ago, spacecadet said:

THere's a lot of heat haze in the distance. I've been failed for that before. In England.

 

I would be surprised if that was the case here as the main building is fine and the haze is not causing any significant softening of the distant buildings. Whatever the case, it is certainly not digital noise.

Link to comment
Share on other sites

5 minutes ago, MDM said:

 

I would be surprised if that was the case here as the main building is fine and the haze is not causing any significant softening of the distant buildings. Whatever the case, it is certainly not digital noise.

I was surprised too, but fail it did.

I have to say they look quite soft to me, enough to fail for SoLD IMO- I wouldn't submit an image that soft. There are even jaggies on the latticework on the right. But as you say they're not the main subject.

As Matt says sometimes the wrong failure reason is picked.

I'd ditch it and move on and accept that there can be a small (very small) amount of subjectivity in QC, as evidenced by the differing opinions here. Hence not sending anything close to the borderline is a good policy

Edited by spacecadet
Link to comment
Share on other sites

I would agree with Spacecadet. There is a little noise in the darkest part of the shadows of the main building, but hardly enough to justify a fail, I would have thought. Overall though I think the image is a little too soft on the main subject to be a reliable pass and I would probably have (reluctantly) deleted it from my submission list. 

Link to comment
Share on other sites

Thanks for the (supportive!) comments. Some of you commented on softness: the main part of the building to the left is pin sharp (see the vertical edge), but perhaps lack of depth of field renders some on the right a tad less than that.

Link to comment
Share on other sites

All's well that ends well - Alamy QC have just got back to me:

"I’ve had another look over the images and I have seen some areas that have colour noise which I’ve screenshotted below."

 

It’s mostly seen in the dark patches of the image but can also be seen within the sky.

 

We think maybe it was a bit of a harsh decision as there isn’t a large amount that can actually be seen.

On second look we think they’re passable images and shall correct it now."

Link to comment
Share on other sites

Your color noise is a bit high. And PIE says you have set color noise reduction to 0. Try 20 or 25. Set the image to 200% to fully appreciate the effect.

You also have a shallow depth of field and a lot of heat noise or blur. That makes some of the out of focus lines on the building look jaggy, probably more so because of the local contrast increase (=sharpening) of the dehaze and clarity settings. Try applying those in the zone that's in focus only. (Difficult because of the color shift of the dehaze.) The same for the saturation and vibrance settings.

 

wim

 

edit: good to see QC agrees 😉

Edited by wiskerke
  • Like 1
Link to comment
Share on other sites

Yes. I was looking for luminance noise.  ooking at it again the colour noise is actually quite bad and quite visible in the building, especially at 200% as wim says. I leave color noise amount set permanently at 25-50-50 as a default in LR for all images for all cameras, all images regardless of ISO and adjust if I need to.

Link to comment
Share on other sites

9 minutes ago, MDM said:

 I leave color noise amount set permanently at 25-50-50 as a default in LR for all images for all cameras, all images regardless of ISO and adjust if I need to.

Same here, I'm not aware of setting them myself so assume it came out of the box like that.

I seem to remember that too much NR desaturates the colour a bit at high ISO, but that's usually the least of your problems.

Edited by spacecadet
Link to comment
Share on other sites

26 minutes ago, spacecadet said:

Same here, I'm not aware of setting them myself so assume it came out of the box like that.

I seem to remember that too much NR desaturates the colour a bit at high ISO, but that's usually the least of your problems.

 

It is the LR default setting. From the time I got my first DSLR I have always set colour noise reduction at that level as a starting point and I rarely change it except at very high ISO (maybe 1600 and above depending on camera and other factors). Colour at high ISO can be very strange anyway. Luminance NR I tend to leave at zero which is also the default but adjust from 400 up as a rule of thumb.

Edited by MDM
Link to comment
Share on other sites

With smaller sensors (RX100) color noise can be a problem. However the cure: applying color noise reduction can be a problem too. I have seen details disappear completely at 50% which was my default for the earlier RX100s and my older Canons. The Canons particularly suffered from color noise. And while they have improved on that, they still cannot touch the Sony sensors in that department.

 

wim

Link to comment
Share on other sites

4 minutes ago, wiskerke said:

With smaller sensors (RX100) color noise can be a problem. However the cure: applying color noise reduction can be a problem too. I have seen details disappear completely at 50% which was my default for the earlier RX100s and my older Canons. The Canons particularly suffered from color noise. And while they have improved on that, they still cannot touch the Sony sensors in that department.

 

wim

 

Yes my first DSLR was a Canon 20D and the colour noise was horrendous if not corrected. Nikons are way better (also Sony sensors although there is some debate it seems about the origin of the latest backlit ones). I don't really care who makes them though - the performance is incredible. Colour noise correction is still required though. 

Link to comment
Share on other sites

Mmm.... I wouldn't have expected a fail for noise, but some of the areas of foliage (ignoring the foreground bush) look a bit mushy (SoLD). Looks much better when downsized to 2000 x 3000 IMHO.

 

Also, what's the black dot in the sky? Probably a bird, but would be better if removed.

 

Mark

Edited by M.Chapman
Link to comment
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
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.