Visualize values in numpy array as colors in a bitmap
I have been given a segmentation mask: a HxW numpy array in which each location represents a pixel. The value at each location is not RGB, however; it is an arbitrary integer in the range 0-60, where the integer is simply a code representing a type of image content.
E.g. array locations whose value is 12 indicate pixels categorized as 'table'; locations whose value is 34 indicate pixels categorized as 'carpet'.
I want to load this segmentation mask into a canvas using tkinter, so I need to make an image out of it, and I need each of the values 0-60 to correspond to colors whose hues are as distinct as can be.
How do I choose these colors? How do I convert my HxW array into an RGB image?
python numpy hue
asked Nov 27 '18 at 21:27
JellicleCatJellicleCat
14.1k1680121
add a comment |
I have been given a segmentation mask: a HxW numpy array in which each location represents a pixel. The value at each location is not RGB, however; it is an arbitrary integer in the range 0-60, where the integer is simply a code representing a type of image content.
E.g. array locations whose value is 12 indicate pixels categorized as 'table'; locations whose value is 34 indicate pixels categorized as 'carpet'.
I want to load this segmentation mask into a canvas using tkinter, so I need to make an image out of it, and I need each of the values 0-60 to correspond to colors whose hues are as distinct as can be.
How do I choose these colors? How do I convert my HxW array into an RGB image?
python numpy hue
asked Nov 27 '18 at 21:27
JellicleCatJellicleCat
14.1k1680121
you shouldn't need to convert to rgb. simply visualise your array via pcolor using a custom colormap
– Tasos Papastylianou
Nov 27 '18 at 22:01
add a comment |
I have been given a segmentation mask: a HxW numpy array in which each location represents a pixel. The value at each location is not RGB, however; it is an arbitrary integer in the range 0-60, where the integer is simply a code representing a type of image content.
E.g. array locations whose value is 12 indicate pixels categorized as 'table'; locations whose value is 34 indicate pixels categorized as 'carpet'.
I want to load this segmentation mask into a canvas using tkinter, so I need to make an image out of it, and I need each of the values 0-60 to correspond to colors whose hues are as distinct as can be.
How do I choose these colors? How do I convert my HxW array into an RGB image?
python numpy hue
asked Nov 27 '18 at 21:27
JellicleCatJellicleCat
14.1k1680121
I have been given a segmentation mask: a HxW numpy array in which each location represents a pixel. The value at each location is not RGB, however; it is an arbitrary integer in the range 0-60, where the integer is simply a code representing a type of image content.
E.g. array locations whose value is 12 indicate pixels categorized as 'table'; locations whose value is 34 indicate pixels categorized as 'carpet'.
I want to load this segmentation mask into a canvas using tkinter, so I need to make an image out of it, and I need each of the values 0-60 to correspond to colors whose hues are as distinct as can be.
How do I choose these colors? How do I convert my HxW array into an RGB image?
python numpy hue
python numpy hue
asked Nov 27 '18 at 21:27
JellicleCatJellicleCat
14.1k1680121
asked Nov 27 '18 at 21:27
JellicleCatJellicleCat
14.1k1680121
asked Nov 27 '18 at 21:27
JellicleCatJellicleCat
14.1k1680121
asked Nov 27 '18 at 21:27
JellicleCatJellicleCat
14.1k1680121
asked Nov 27 '18 at 21:27
JellicleCatJellicleCat
14.1k1680121
14.1k1680121
you shouldn't need to convert to rgb. simply visualise your array via pcolor using a custom colormap
– Tasos Papastylianou
Nov 27 '18 at 22:01
add a comment |
you shouldn't need to convert to rgb. simply visualise your array via pcolor using a custom colormap
– Tasos Papastylianou
Nov 27 '18 at 22:01
you shouldn't need to convert to rgb. simply visualise your array via pcolor using a custom colormap
– Tasos Papastylianou
Nov 27 '18 at 22:01
you shouldn't need to convert to rgb. simply visualise your array via pcolor using a custom colormap
– Tasos Papastylianou
Nov 27 '18 at 22:01
add a comment |
2 Answers
2
active
oldest
votes
You don't need to convert to RGB. Just use pcolor with an appropriate colormap.
It could be something as simple as this:
import matplotlib.pyplot as plt
import numpy
M = numpy.array([[1,2,3,4,5,6,7,8,9,10], [11,12,13,14,15,16,17,18,19,20]])
plt.pcolor( M , cmap = 'hsv' )
plt.show()
Added by Mark Setchell:
answered Nov 27 '18 at 22:06
Tasos PapastylianouTasos Papastylianou
11.1k11133
2
It would be helpful to show how that looks and also to explain how to make that into something OP can load into aTkIntercanvas. I hope you don't mind my adding the plot - just delete it if you do.
– Mark Setchell
Nov 27 '18 at 22:13
1
@MarkSetchell ah apologies, I missed the tkinter bit, thanks for pointing it out. I haven't done it in a while, but effectively, one can use a matplotlib object in a tkinter gui, e.g. as in this post: stackoverflow.com/q/4073660/4183191 Again, there would be no need to convert to RGB, just use the resulting figure in the tkinter gui.
– Tasos Papastylianou
Nov 27 '18 at 22:28
1
I had to usepcolormeshinstead ofpcolorbecause the former was too slow
– JellicleCat
Nov 28 '18 at 0:04
add a comment |
IMHO, the obvious thing to do would be to make a palettised image where you have a palette of 60 colours and an index into the palette at each location. That is very simple with PIL/Pillow.
Now you come to choosing those 60 colours. I would think of using HSL (Hue, Saturation and Lightness) colourspace. Now you need 60 different values. So you could use full Saturation and Lightness for all pixels and spread the available 360 degrees of Hue into 6 degree chunks for 60 colours. Or you could use 100% Lightness and 50% Lightness plus 100% Saturation and 50% Saturation to give you 4x as many Lightness/Saturation combinations meaning you could rotate 24 degrees of Hue for greater separation.
See here for example of how to make an image from a Numpy array, add a palette to an image and make into Pillow image and save.
I am basically suggesting a 60-entry palette, with the following values:
Hue, Saturation, Lightness
0 50% 50%
0 50% 100%
0 100% 50%
0 100% 100%
23 50% 50%
23 50% 100%
23 100% 50%
23 100% 100%
47 50% 50%
47 50% 100%
47 100% 50%
47 100% 100%
61 ...
answered Nov 27 '18 at 21:55
Mark SetchellMark Setchell
91.2k779183
I like this method, but I'm having scant luck. I'm trying the toy exampleImage.fromarray([[0,1,2,3,4]], mode='P')and am getting a single, solid color which matches the first color in my palette.
– JellicleCat
Nov 27 '18 at 23:21
add a comment |
Your Answer
StackExchange.ifUsing("editor", function () {
StackExchange.using("externalEditor", function () {
StackExchange.using("snippets", function () {
StackExchange.snippets.init();
});
});
}, "code-snippets");
StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "1"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});
function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});
}
});
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53508413%2fvisualize-values-in-numpy-array-as-colors-in-a-bitmap%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
2 Answers
2
active
oldest
votes
2 Answers
2
active
oldest
votes
active
oldest
votes
active
oldest
votes
You don't need to convert to RGB. Just use pcolor with an appropriate colormap.
It could be something as simple as this:
import matplotlib.pyplot as plt
import numpy
M = numpy.array([[1,2,3,4,5,6,7,8,9,10], [11,12,13,14,15,16,17,18,19,20]])
plt.pcolor( M , cmap = 'hsv' )
plt.show()
Added by Mark Setchell:
answered Nov 27 '18 at 22:06
Tasos PapastylianouTasos Papastylianou
11.1k11133
2
It would be helpful to show how that looks and also to explain how to make that into something OP can load into aTkIntercanvas. I hope you don't mind my adding the plot - just delete it if you do.
– Mark Setchell
Nov 27 '18 at 22:13
1
@MarkSetchell ah apologies, I missed the tkinter bit, thanks for pointing it out. I haven't done it in a while, but effectively, one can use a matplotlib object in a tkinter gui, e.g. as in this post: stackoverflow.com/q/4073660/4183191 Again, there would be no need to convert to RGB, just use the resulting figure in the tkinter gui.
– Tasos Papastylianou
Nov 27 '18 at 22:28
1
I had to usepcolormeshinstead ofpcolorbecause the former was too slow
– JellicleCat
Nov 28 '18 at 0:04
add a comment |
You don't need to convert to RGB. Just use pcolor with an appropriate colormap.
It could be something as simple as this:
import matplotlib.pyplot as plt
import numpy
M = numpy.array([[1,2,3,4,5,6,7,8,9,10], [11,12,13,14,15,16,17,18,19,20]])
plt.pcolor( M , cmap = 'hsv' )
plt.show()
Added by Mark Setchell:
answered Nov 27 '18 at 22:06
Tasos PapastylianouTasos Papastylianou
11.1k11133
2
It would be helpful to show how that looks and also to explain how to make that into something OP can load into aTkIntercanvas. I hope you don't mind my adding the plot - just delete it if you do.
– Mark Setchell
Nov 27 '18 at 22:13
1
@MarkSetchell ah apologies, I missed the tkinter bit, thanks for pointing it out. I haven't done it in a while, but effectively, one can use a matplotlib object in a tkinter gui, e.g. as in this post: stackoverflow.com/q/4073660/4183191 Again, there would be no need to convert to RGB, just use the resulting figure in the tkinter gui.
– Tasos Papastylianou
Nov 27 '18 at 22:28
1
I had to usepcolormeshinstead ofpcolorbecause the former was too slow
– JellicleCat
Nov 28 '18 at 0:04
add a comment |
You don't need to convert to RGB. Just use pcolor with an appropriate colormap.
It could be something as simple as this:
import matplotlib.pyplot as plt
import numpy
M = numpy.array([[1,2,3,4,5,6,7,8,9,10], [11,12,13,14,15,16,17,18,19,20]])
plt.pcolor( M , cmap = 'hsv' )
plt.show()
Added by Mark Setchell:
answered Nov 27 '18 at 22:06
Tasos PapastylianouTasos Papastylianou
11.1k11133
You don't need to convert to RGB. Just use pcolor with an appropriate colormap.
It could be something as simple as this:
import matplotlib.pyplot as plt
import numpy
M = numpy.array([[1,2,3,4,5,6,7,8,9,10], [11,12,13,14,15,16,17,18,19,20]])
plt.pcolor( M , cmap = 'hsv' )
plt.show()
Added by Mark Setchell:
answered Nov 27 '18 at 22:06
Tasos PapastylianouTasos Papastylianou
11.1k11133
edited Nov 27 '18 at 22:22
answered Nov 27 '18 at 22:06
Tasos PapastylianouTasos Papastylianou
11.1k11133
answered Nov 27 '18 at 22:06
Tasos PapastylianouTasos Papastylianou
11.1k11133
answered Nov 27 '18 at 22:06
Tasos PapastylianouTasos Papastylianou
11.1k11133
11.1k11133
2
It would be helpful to show how that looks and also to explain how to make that into something OP can load into aTkIntercanvas. I hope you don't mind my adding the plot - just delete it if you do.
– Mark Setchell
Nov 27 '18 at 22:13
1
@MarkSetchell ah apologies, I missed the tkinter bit, thanks for pointing it out. I haven't done it in a while, but effectively, one can use a matplotlib object in a tkinter gui, e.g. as in this post: stackoverflow.com/q/4073660/4183191 Again, there would be no need to convert to RGB, just use the resulting figure in the tkinter gui.
– Tasos Papastylianou
Nov 27 '18 at 22:28
1
I had to usepcolormeshinstead ofpcolorbecause the former was too slow
– JellicleCat
Nov 28 '18 at 0:04
add a comment |
2
It would be helpful to show how that looks and also to explain how to make that into something OP can load into aTkIntercanvas. I hope you don't mind my adding the plot - just delete it if you do.
– Mark Setchell
Nov 27 '18 at 22:13
1
@MarkSetchell ah apologies, I missed the tkinter bit, thanks for pointing it out. I haven't done it in a while, but effectively, one can use a matplotlib object in a tkinter gui, e.g. as in this post: stackoverflow.com/q/4073660/4183191 Again, there would be no need to convert to RGB, just use the resulting figure in the tkinter gui.
– Tasos Papastylianou
Nov 27 '18 at 22:28
1
I had to usepcolormeshinstead ofpcolorbecause the former was too slow
– JellicleCat
Nov 28 '18 at 0:04
2
2
It would be helpful to show how that looks and also to explain how to make that into something OP can load into a
TkInter canvas. I hope you don't mind my adding the plot - just delete it if you do.– Mark Setchell
Nov 27 '18 at 22:13
It would be helpful to show how that looks and also to explain how to make that into something OP can load into a
TkInter canvas. I hope you don't mind my adding the plot - just delete it if you do.– Mark Setchell
Nov 27 '18 at 22:13
1
1
@MarkSetchell ah apologies, I missed the tkinter bit, thanks for pointing it out. I haven't done it in a while, but effectively, one can use a matplotlib object in a tkinter gui, e.g. as in this post: stackoverflow.com/q/4073660/4183191 Again, there would be no need to convert to RGB, just use the resulting figure in the tkinter gui.
– Tasos Papastylianou
Nov 27 '18 at 22:28
@MarkSetchell ah apologies, I missed the tkinter bit, thanks for pointing it out. I haven't done it in a while, but effectively, one can use a matplotlib object in a tkinter gui, e.g. as in this post: stackoverflow.com/q/4073660/4183191 Again, there would be no need to convert to RGB, just use the resulting figure in the tkinter gui.
– Tasos Papastylianou
Nov 27 '18 at 22:28
1
1
I had to use
pcolormesh instead of pcolor because the former was too slow– JellicleCat
Nov 28 '18 at 0:04
I had to use
pcolormesh instead of pcolor because the former was too slow– JellicleCat
Nov 28 '18 at 0:04
add a comment |
IMHO, the obvious thing to do would be to make a palettised image where you have a palette of 60 colours and an index into the palette at each location. That is very simple with PIL/Pillow.
Now you come to choosing those 60 colours. I would think of using HSL (Hue, Saturation and Lightness) colourspace. Now you need 60 different values. So you could use full Saturation and Lightness for all pixels and spread the available 360 degrees of Hue into 6 degree chunks for 60 colours. Or you could use 100% Lightness and 50% Lightness plus 100% Saturation and 50% Saturation to give you 4x as many Lightness/Saturation combinations meaning you could rotate 24 degrees of Hue for greater separation.
See here for example of how to make an image from a Numpy array, add a palette to an image and make into Pillow image and save.
I am basically suggesting a 60-entry palette, with the following values:
Hue, Saturation, Lightness
0 50% 50%
0 50% 100%
0 100% 50%
0 100% 100%
23 50% 50%
23 50% 100%
23 100% 50%
23 100% 100%
47 50% 50%
47 50% 100%
47 100% 50%
47 100% 100%
61 ...
answered Nov 27 '18 at 21:55
Mark SetchellMark Setchell
91.2k779183
I like this method, but I'm having scant luck. I'm trying the toy exampleImage.fromarray([[0,1,2,3,4]], mode='P')and am getting a single, solid color which matches the first color in my palette.
– JellicleCat
Nov 27 '18 at 23:21
add a comment |
IMHO, the obvious thing to do would be to make a palettised image where you have a palette of 60 colours and an index into the palette at each location. That is very simple with PIL/Pillow.
Now you come to choosing those 60 colours. I would think of using HSL (Hue, Saturation and Lightness) colourspace. Now you need 60 different values. So you could use full Saturation and Lightness for all pixels and spread the available 360 degrees of Hue into 6 degree chunks for 60 colours. Or you could use 100% Lightness and 50% Lightness plus 100% Saturation and 50% Saturation to give you 4x as many Lightness/Saturation combinations meaning you could rotate 24 degrees of Hue for greater separation.
See here for example of how to make an image from a Numpy array, add a palette to an image and make into Pillow image and save.
I am basically suggesting a 60-entry palette, with the following values:
Hue, Saturation, Lightness
0 50% 50%
0 50% 100%
0 100% 50%
0 100% 100%
23 50% 50%
23 50% 100%
23 100% 50%
23 100% 100%
47 50% 50%
47 50% 100%
47 100% 50%
47 100% 100%
61 ...
answered Nov 27 '18 at 21:55
Mark SetchellMark Setchell
91.2k779183
I like this method, but I'm having scant luck. I'm trying the toy exampleImage.fromarray([[0,1,2,3,4]], mode='P')and am getting a single, solid color which matches the first color in my palette.
– JellicleCat
Nov 27 '18 at 23:21
add a comment |
IMHO, the obvious thing to do would be to make a palettised image where you have a palette of 60 colours and an index into the palette at each location. That is very simple with PIL/Pillow.
Now you come to choosing those 60 colours. I would think of using HSL (Hue, Saturation and Lightness) colourspace. Now you need 60 different values. So you could use full Saturation and Lightness for all pixels and spread the available 360 degrees of Hue into 6 degree chunks for 60 colours. Or you could use 100% Lightness and 50% Lightness plus 100% Saturation and 50% Saturation to give you 4x as many Lightness/Saturation combinations meaning you could rotate 24 degrees of Hue for greater separation.
See here for example of how to make an image from a Numpy array, add a palette to an image and make into Pillow image and save.
I am basically suggesting a 60-entry palette, with the following values:
Hue, Saturation, Lightness
0 50% 50%
0 50% 100%
0 100% 50%
0 100% 100%
23 50% 50%
23 50% 100%
23 100% 50%
23 100% 100%
47 50% 50%
47 50% 100%
47 100% 50%
47 100% 100%
61 ...
answered Nov 27 '18 at 21:55
Mark SetchellMark Setchell
91.2k779183
IMHO, the obvious thing to do would be to make a palettised image where you have a palette of 60 colours and an index into the palette at each location. That is very simple with PIL/Pillow.
Now you come to choosing those 60 colours. I would think of using HSL (Hue, Saturation and Lightness) colourspace. Now you need 60 different values. So you could use full Saturation and Lightness for all pixels and spread the available 360 degrees of Hue into 6 degree chunks for 60 colours. Or you could use 100% Lightness and 50% Lightness plus 100% Saturation and 50% Saturation to give you 4x as many Lightness/Saturation combinations meaning you could rotate 24 degrees of Hue for greater separation.
See here for example of how to make an image from a Numpy array, add a palette to an image and make into Pillow image and save.
I am basically suggesting a 60-entry palette, with the following values:
Hue, Saturation, Lightness
0 50% 50%
0 50% 100%
0 100% 50%
0 100% 100%
23 50% 50%
23 50% 100%
23 100% 50%
23 100% 100%
47 50% 50%
47 50% 100%
47 100% 50%
47 100% 100%
61 ...
answered Nov 27 '18 at 21:55
Mark SetchellMark Setchell
91.2k779183
edited Nov 27 '18 at 22:26
answered Nov 27 '18 at 21:55
Mark SetchellMark Setchell
91.2k779183
answered Nov 27 '18 at 21:55
Mark SetchellMark Setchell
91.2k779183
answered Nov 27 '18 at 21:55
Mark SetchellMark Setchell
91.2k779183
91.2k779183
I like this method, but I'm having scant luck. I'm trying the toy exampleImage.fromarray([[0,1,2,3,4]], mode='P')and am getting a single, solid color which matches the first color in my palette.
– JellicleCat
Nov 27 '18 at 23:21
add a comment |
I like this method, but I'm having scant luck. I'm trying the toy exampleImage.fromarray([[0,1,2,3,4]], mode='P')and am getting a single, solid color which matches the first color in my palette.
– JellicleCat
Nov 27 '18 at 23:21
I like this method, but I'm having scant luck. I'm trying the toy example
Image.fromarray([[0,1,2,3,4]], mode='P') and am getting a single, solid color which matches the first color in my palette.– JellicleCat
Nov 27 '18 at 23:21
I like this method, but I'm having scant luck. I'm trying the toy example
Image.fromarray([[0,1,2,3,4]], mode='P') and am getting a single, solid color which matches the first color in my palette.– JellicleCat
Nov 27 '18 at 23:21
add a comment |
Thanks for contributing an answer to Stack Overflow!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
To learn more, see our tips on writing great answers.
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f53508413%2fvisualize-values-in-numpy-array-as-colors-in-a-bitmap%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
you shouldn't need to convert to rgb. simply visualise your array via pcolor using a custom colormap
– Tasos Papastylianou
Nov 27 '18 at 22:01