Hue Clock "Time Guide"

Share your quick and clever Photoshop, Elements, or Curvemeister techniques here!
ggroess
Posts: 5342
Joined: Wed May 24, 2006 2:15 am
Contact:

Postby ggroess » Wed May 15, 2013 1:33 pm

I have been asked by a student in the 101 class if there is a "guide" for using the hue clocks based on time; like we do for Skin tones.  Where you have a general area of the hue clock you use to denote certain colors.

For instance, in the lessons we talk about making sure that grass and foliage are in the "more yellow than green" area but, we do not denote a time like 12:30 to 1:30 on the hue clock like we do for skin tones.

The student is asking if you all have suggestions for areas like that and what they are. 

Post your reply here so we can begin a "guide" for everyone to use...
Greg

leeharper_admin
Posts: 263
Joined: Tue Aug 20, 2013 2:39 pm

Postby leeharper_admin » Sat May 18, 2013 2:26 pm

Hi everyone,

This is a really interesting and important topic. There are certain objects where our expectation about their color requires that our corrections put them into a specific region of the hue clock. Obviously skin tones are a clear example, but the list of objects we need to be careful about is a little longer...

This topic has been much studied - under the name 'Memory Colors'; a good introduction is available here: http://vanhurkman.com/wordpress/?p=58

Alexis Van Hurkman mentions two lists of objects that need special consideration:




  • Skin Tones

  • Green Grass

  • Blue Sky



And the second:



  • Red Brick

  • Green Grass

  • Dry Grass

  • Blue Sky

  • Flesh

  • Tan Flesh

  • Green Foliage

  • Evergreens

  • Inland Soil

  • Beach Sand




The question is: which of these should we consider useful?

As a start, I think that defining time values for the following would be most helpful:



  • Skin Tones

  • Green Foliage

  • Green Grass

  • Evergreens

  • Blue Sky




Whilst beach sand is a useful memory color (and something that requires attention), it does vary a lot - so whether we can come to some values that are generally useful is up for debate; the same applies for bricks...

Most of those items listed above are covered by CM's color pin files, so putting together some guide numbers shouldn't be too difficult. The issue - mentioned in the Van Hurkman article is that memory color preferences are (sometimes) geographically specific; therefore, any suggestions given are areas of the clock that a reading ought to be "in the neighborhood of", and not values to match exactly.

I'll try to put some numbers together, and will post them here.

Looking forward to hearing other opinions : )

Lee.

imported_ganna
Posts: 460
Joined: Tue Nov 07, 2006 11:41 am

Postby imported_ganna » Sat May 18, 2013 3:15 pm

Thanks a lot for your reply Lee.  For what its worth, I battle a bit with blue sky because air pollution definately plays a role here. We live in a relatively dry country, therefor dust in the air and also a lot of veld fires this time of the year (dry grass ).
I find that on many of my images, the blue (cloudless) sky just does'nt look right, although the rest of the colours seem to be OK; even when I try to pin it.

ggroess
Posts: 5342
Joined: Wed May 24, 2006 2:15 am
Contact:

Postby ggroess » Sat May 18, 2013 9:12 pm

ganna,
Is it too green??

check the CM sky pins for the RGB values for those sky pins and see if you have too much green.  It is one of the most common problems in Sky color.

Greg

imported_ganna
Posts: 460
Joined: Tue Nov 07, 2006 11:41 am

Postby imported_ganna » Sun May 19, 2013 10:22 am

Thanks, your'e right, always too much green, tend to be to close to the 6 o clock (cyan) side of blue. I must admit, I was unaware of pins being spesific for RGB or Lab. Sometimes easy to miss a simple thing along the way and then the top structure keeps on cracking. Thanks for this, shows, it is worth going back to the basics from time to time :) ("when all else fail, read the instruction manual")

mikemeister_admin
Posts: 4927
Joined: Fri Sep 20, 2013 8:29 pm

Postby mikemeister_admin » Sun May 19, 2013 1:45 pm

It sounds like a very sensible suggestion, but...

after I completed the CM course, I would have loved to have had this information to help me adjust images.  But since that time I have read more and more about how we perceive colour compared to a camera.

You will notice from the references Lee posted, that pure grey can sometimes be perfect for skin colour.  I have been finding examples where we are easily fooled - for instance a yellow-red tint (hue 41) is perceived as a lovely pink shade (hue 353)!  When you check old painting, it seems that most tree leaves are yellow hue rather than green!  I think this is the painter knowing how to fool our eyes and create an effect that is not there in (digital) nature, but which we find appealing.

The contrast, and colour, of an area surrounding an object in an image can result in our brains doing strange things.  I'm sure you all know the chess squares (bottom of http://www.broadhurst-family.co.uk/lefteye/mainpages/oh_no.htm) image (that has been discussed at length on the ColorTheory forum) where we think the squares are not the same.

As well as this surrounding contrast effect, we also have the tone of the colour.  Taking a leaf out of the Grand Masters, they change the chroma of a colour as it gets darker to create their fantastic paintings.  There is an accepted (?) relationship suggested by Frank Reilly (I've done a bit of analysis at http://www.broadhurst-family.co.uk/lefteye/mainpages/reillymunsell.htm) that defines this change.

I use to think that when I corrected an image, I was to try to get everything 'correct' by numbers - some chap called Dan thought this was a good idea!  But he was wrong and I believe his current thinking is rapidly moving away from exact numbers to human perception and making a statement/impact with an image.

I now think that my digital photography is an art form and as such one should post-process an image to try and allow the viewer to see some of the emotion and feeling we experienced when we took the photo (I'm not referring to snapshots!!).  You are telling a story, or trying to show why you thought the photo was worth recording in the first place.  You saw the scene through your eyes, not a camera sensor - so I think the final image should be our view of it, not the cameras.  That means post-processing for Human Perception not necessarily for Actual/Factual colours and tones.  Therefore this idea of recording Hue values is a good idea, providing we do not religiously follow it or believe it to be correct for our particular scene.

Lastly there is the awful problem of how we are going to view the image - print, profile aware program or web and even size, plus Calibrated or not monitors.  There are just so many variables it becomes a nightmare to get 'right'.

Humph - better in than out or the ramblings of a mad man?
Zog

imported_ganna
Posts: 460
Joined: Tue Nov 07, 2006 11:41 am

Postby imported_ganna » Sun May 19, 2013 2:13 pm

Ag- No Chris, just as I beginning to think colour correction is easy, you pulled the dam from under the goose  :D

leeharper_admin
Posts: 263
Joined: Tue Aug 20, 2013 2:39 pm

Postby leeharper_admin » Sun May 19, 2013 7:25 pm

Thank you for the link to your article Chris, it was very interesting. Greg and I have worked on this idea of limiting saturation boosts to specific areas of luminance (also inspired by Munsell's work) using a special luminance mask; you may have seen Greg use it in some of his videos. I think, however, that saturation enhancement (inc. reduction of saturation in shadows) deserves the attention of another thread in the forum...

You rightly mention that the colours we might prefer for particular objects might not match their actual values as recorded by a camera, and that what is perceptually correct may not be scientifically accurate. You are absolutely correct. However, 'memory colours' are the values that people associate with objects; as I recall, the research papers mentioned on Alexis Van Hurkman's blog recommend hue values that differ from the actual coloration of the objects in question.

Guidelines for where items ought to sit on CM's Hue Clock are still valid - as is working "By the Numbers" - so long as the numbers we are working towards denote human preference, rather than the recordings of a spectrophotometer. For example, a recommendation for 'Blue Sky' should recommend a value that people tend to like.

This is of course where things get complicated. If we are trying to plot items according to human preference, the inevitable question raised is "who's preference do these values refer to?" As is also mentioned on Van Hurkman's site, is that colour preferences are geographically distinct. As mentioned in one of Dan Margulis' new videos, Italian's seem to prefer paler skin tones than Americans.

For this reason, I don't think that specific values can be suggested, but we can arrive at some ballpark suggestions. For example, the Hue Clock skin tone recommendation is 'between 12:30 and 1:30', rather than '12:52'; I think that something similar will be possible for other common objects...

Thank you again for your link - and Ganna, don't give up hope just yet  ;)

Cheers,
Lee.

ggroess
Posts: 5342
Joined: Wed May 24, 2006 2:15 am
Contact:

Postby ggroess » Tue May 21, 2013 6:40 pm


You will notice from the references Lee posted, that pure grey can sometimes be perfect for skin colour.  I have been finding examples where we are easily fooled - for instance a yellow-red tint (hue 41) is perceived as a lovely pink shade (hue 353)!  When you check old painting, it seems that most tree leaves are yellow hue rather than green!  I think this is the painter knowing how to fool our eyes and create an effect that is not there in (digital) nature, but which we find appealing.


A couple of things here Chris...
I always try to look at the underlying colors and systems that we use to perceive them.  For instance, Green foliage to me is easy to see as yellow.  When Green plants die they typically turn yellow or brown.  to me the under lying color is always present in the living plant just waiting for the chlorophyll to go away and reveal the true colors.

I have often been fascinated by the colors things display under extremely low lighting conditions.  I absolutely love the time of the day when there is enough light to see but not enough to distinguish colors.  I find it compelling to try figure out the true color of objects under these conditions.  Even when I know the true colors I am still amazed at the complexity of the visual sense.


The contrast, and colour, of an area surrounding an object in an image can result in our brains doing strange things.  I'm sure you all know the chess squares (bottom of http://www.broadhurst-family.co.uk/lefteye/mainpages/oh_no.htm) image (that has been discussed at length on the ColorTheory forum) where we think the squares are not the same.

As well as this surrounding contrast effect, we also have the tone of the colour.  Taking a leaf out of the Grand Masters, they change the chroma of a colour as it gets darker to create their fantastic paintings.  There is an accepted (?) relationship suggested by Frank Reilly (I've done a bit of analysis at http://www.broadhurst-family.co.uk/lefteye/mainpages/reillymunsell.htm) that defines this change.


Color contrast perception is more about the understanding of the process for me than it is about the actual application of the theory.  If I understand the process, I can choose when and how much to affect the colors and tones in my images.  Understanding that the perception of my viewer is part of the process is key to being able to apply it.  I think the masters knew this or studied it to add the dimensional aspect of color perception to the frames.  I personally think that if I try too hard to apply the theory I make bigger mistakes with my images.


I use to think that when I corrected an image, I was to try to get everything 'correct' by numbers - some chap called Dan thought this was a good idea!  But he was wrong and I believe his current thinking is rapidly moving away from exact numbers to human perception and making a statement/impact with an image.


I'd agree that the "perfect" by the numbers correction is a bad example to have set and we are teaching it a bit differently now.  When we have a natural set of points that allow the full by the numbers correction we do it but if you only have some of the parts say a solid shadow or a real highlight that should be neutral we can accept that and make sure the rest of the image makes sense.


I now think that my digital photography is an art form and as such one should post-process an image to try and allow the viewer to see some of the emotion and feeling we experienced when we took the photo (I'm not referring to snapshots!!).  You are telling a story, or trying to show why you thought the photo was worth recording in the first place.  You saw the scene through your eyes, not a camera sensor - so I think the final image should be our view of it, not the cameras.  That means post-processing for Human Perception not necessarily for Actual/Factual colours and tones.  Therefore this idea of recording Hue values is a good idea, providing we do not religiously follow it or believe it to be correct for our particular scene.


I think I have a much better understanding of "your actual mileage may vary" as a part of the process than I had say 3 years ago.  I know I teach it differently and will be changing even more as time goes on.


Lastly there is the awful problem of how we are going to view the image - print, profile aware program or web and even size, plus Calibrated or not monitors.  There are just so many variables it becomes a nightmare to get 'right'.

Humph - better in than out or the ramblings of a mad man?
Zog


I would agree that the destination of the images is extremely important in regards to the corrective actions we are making.  I handle images differently if I am making a print or just making a web based image.

I'd venture to guess that madness may not be the cause, but rather passion for good images is at the root of your ramblings...

Respectfully
Greg 

ggroess
Posts: 5342
Joined: Wed May 24, 2006 2:15 am
Contact:

Postby ggroess » Wed May 22, 2013 8:36 pm

Saw this and could not resist....

Greg
Attachments
l1ssvzm-jpg
l1ssvzm-jpg (166.91 KiB) Viewed 4433 times


Return to “Tips and Tricks”

Who is online

Users browsing this forum: No registered users and 1 guest