Choosing Colors for Grayscale and Colored Maps
Started by
Textularia
, Jul 07 2006 09:06 AM
12 replies to this topic
#1
Posted 07 July 2006 - 09:06 AM
Dear List Members,
With papers being published at work and in an outside journal, I have a problem with choosing the colors for the polygons of geologic maps, which in my figures. In each case there is a printed and PDF version of the publication. At this time, I am only allowed to submit one version of each figure for publication. The problem is that the PDF version is in color, while the printed version is in grayscale. Thus, the colors, which I choose for the colored PDF version must also be distinguishable and look good when they are converted to grayscale print version.
Does anyone have any advice about how to choose colors, patterns, or combinations of both, which look nice in color but also contrast with each other as to be readily distinguishable in grayscale?
Are there any web pages, papers, or books that would help me with this problem?
Yours,
Paul Heinrich
Baton Rouge, LA 70803
With papers being published at work and in an outside journal, I have a problem with choosing the colors for the polygons of geologic maps, which in my figures. In each case there is a printed and PDF version of the publication. At this time, I am only allowed to submit one version of each figure for publication. The problem is that the PDF version is in color, while the printed version is in grayscale. Thus, the colors, which I choose for the colored PDF version must also be distinguishable and look good when they are converted to grayscale print version.
Does anyone have any advice about how to choose colors, patterns, or combinations of both, which look nice in color but also contrast with each other as to be readily distinguishable in grayscale?
Are there any web pages, papers, or books that would help me with this problem?
Yours,
Paul Heinrich
Baton Rouge, LA 70803
#2
Posted 07 July 2006 - 09:18 AM
My cartography teacher always told us we should use no more than 4 shades of gray in a map. So you get black/white, and then 20%, 40%, 60%, 80%. I think some people would even say only use 3 greys + b/w, that being 25%, 50%, and 75%.
But that was advice for a map prior to its creation. Since you already have the maps created, I'm not sure if this works for you.
But that was advice for a map prior to its creation. Since you already have the maps created, I'm not sure if this works for you.
#3
Posted 07 July 2006 - 10:11 AM
I'd have to agree with Ben here. When you're limited to grayscale, there's not that much you can do. 3 or 4 shades of gray and that's it. In fact, I'm not too sure on the 80% grey, it'd probabely be too much to black.
Of course this may not be an issue for your map.
If you want to show more different types of areas (or don't want to use the really dark shades of gray), you'll have to resort to a combination of grays and pattern fills.
Of course this may not be an issue for your map.
If you want to show more different types of areas (or don't want to use the really dark shades of gray), you'll have to resort to a combination of grays and pattern fills.
Hans van der Maarel - Cartotalk Editor
Red Geographics
Email: hans@redgeographics.com / Twitter: @redgeographics
Red Geographics
Email: hans@redgeographics.com / Twitter: @redgeographics
#4
Posted 07 July 2006 - 11:46 AM
I'd have to agree with Ben here. When you're limited to grayscale, there's not that much you can do. 3 or 4 shades of gray and that's it. In fact, I'm not too sure on the 80% grey, it'd probabely be too much to black.
Of course this may not be an issue for your map.
If you want to show more different types of areas (or don't want to use the really dark shades of gray), you'll have to resort to a combination of grays and pattern fills.
I was just looking through Cynthia Brewer's "Designing Better Maps" book yesterday, looking at some color techniques. I noticed in there that she does show a gray scale that incorporates 9 different shades.
Personally, I wouldn't recommend more than 5 if possible. 6 if necessary.
If you follow Ms. Brewer's techniques, here are her recommendations:
5: 3, 20, 41, 61, 85
6: 3, 15, 26, 41, 61, 85
David Toney, GISP
GIS Manager
United States Marine Corps
West Coast Installations
GIS Manager
United States Marine Corps
West Coast Installations
#5
Posted 07 July 2006 - 11:52 AM
5: 3, 20, 41, 61, 85
6: 3, 15, 26, 41, 61, 85
In this case are you excluding black (100) and white (0) as classes?
#6
Posted 07 July 2006 - 12:01 PM
Here is another handy quote to remember... from Cynthia Brewer's " Designing Better Maps" "Equal percentage steps do not equal visual steps."
I would stick with 4-5 classes and start at 2-3% black as well.
I would stick with 4-5 classes and start at 2-3% black as well.
#7
Posted 07 July 2006 - 12:12 PM
5: 3, 20, 41, 61, 85
6: 3, 15, 26, 41, 61, 85
In this case are you excluding black (100) and white (0) as classes?
3% and 85% are effectively white and black when printed.
Nick Springer
Owner: Springer Cartographics LLC
Director of Design and Web Applications: ALK Technologies Inc.
Chief Creative Officer: Dashflo.com
Owner: Springer Cartographics LLC
Director of Design and Web Applications: ALK Technologies Inc.
Chief Creative Officer: Dashflo.com
#8
Posted 07 July 2006 - 12:48 PM
It all depends on the actual map. If you print them right next to eachother, so that the areas actually touch, 85% grey and full black can be distinguished (assuming it's a high-quality print of course). If you print them some distance apart, it's going to be much harder, if not impossible.
Hans van der Maarel - Cartotalk Editor
Red Geographics
Email: hans@redgeographics.com / Twitter: @redgeographics
Red Geographics
Email: hans@redgeographics.com / Twitter: @redgeographics
#9
Posted 07 July 2006 - 02:37 PM
My two cents...
Many years ago in a cartography class the professor talked at length about the lighter shades of gray being more distinguishable than the darker shades of gray. This has sort of stuck with me so I usually do something like this when making grayscale maps:
5: 0, 10, 30, 50, 85
6: 0, 10, 25, 40, 60, 85
I don't find shades of gray over 80% to be distinguishable at all especially in print.
Rich
Many years ago in a cartography class the professor talked at length about the lighter shades of gray being more distinguishable than the darker shades of gray. This has sort of stuck with me so I usually do something like this when making grayscale maps:
5: 0, 10, 30, 50, 85
6: 0, 10, 25, 40, 60, 85
I don't find shades of gray over 80% to be distinguishable at all especially in print.
Rich
#10
Posted 10 July 2006 - 10:04 AM
As far as colors that work well as grayscale, I haven't found any helpful resources. I run into a similar problem with my maps, which are constantly being photocopied and faxed. I've just done a lot of trial and error by Xeroxing the printed version to ensure everything still shows up. So far I've learned that subtlty does not work and contrast is king.
You may want to do the same experimenting by converting your maps into grayscale through Photoshop.
Mike
Mike Pesses
Grad Student | CSUN
GIS Specialist | County of Kern
You may want to do the same experimenting by converting your maps into grayscale through Photoshop.
Mike
Mike Pesses
Grad Student | CSUN
GIS Specialist | County of Kern
#11
Posted 10 July 2006 - 09:21 PM
Also bear in mind the printing method: if you are printing offseet with an average printer, any light tints (like the 3%) Cindy Brewer suggests, will effectively pring much darker because of dot gain... probably actually more like 8-10% of the actual page will be inked.
Likewise, laser printers, especially if you are using a fine dot screen (133 lines per inch or more), will not produce 100 different screens. For every "cell" in qhich a dot is placed, there may not be 100 pixels of toner available to toggle on or off. For a test, produce a smooth gradation across a page, say from 0 to 50%, and count how many bands there are. This will tell you how many percentages you actually have available to you for the dot-screen density at which you are printing.
Nat Case
Hedberg Maps
Minneapolis
Likewise, laser printers, especially if you are using a fine dot screen (133 lines per inch or more), will not produce 100 different screens. For every "cell" in qhich a dot is placed, there may not be 100 pixels of toner available to toggle on or off. For a test, produce a smooth gradation across a page, say from 0 to 50%, and count how many bands there are. This will tell you how many percentages you actually have available to you for the dot-screen density at which you are printing.
Nat Case
Hedberg Maps
Minneapolis
Nat Case
Head of Production, Hedberg Maps, Minneapolis, MN USA
maphead.blogspot.com
"Life's too short for bad maps"
Head of Production, Hedberg Maps, Minneapolis, MN USA
maphead.blogspot.com
"Life's too short for bad maps"
#12
Posted 12 July 2006 - 10:05 PM
Hi,
you may be interested in some research done over at Northwestern university:
http://www.cs.northw...820/color2gray/
They developed some code to minimize the information loss in the color to grayscale transformation.
cheers
ned
you may be interested in some research done over at Northwestern university:
http://www.cs.northw...820/color2gray/
They developed some code to minimize the information loss in the color to grayscale transformation.
cheers
ned
#13
Posted 13 July 2006 - 02:17 AM
Thanks, Nedmohr, that's very interesting !
0 user(s) are reading this topic
0 members, 0 guests, 0 anonymous users


Sign In
Create Account

United States
Back to top
Netherlands
France








