Assign scalar value to 'RGB' of few dominant colors of an image using python

Question

I have an image which is made up of some discrete colors such as green, blue, red, yellow, navy. I would like to read this image and make an array which has following two attributes:

one numerical value for each discrete color,
size of each axis of the array to be proportionate (approximately) to the image dimensions.

P.S.: All the questions that I have come across with respect to reading images deal with rgb/pixels. This is just one question that would have been useful if the objective of the question was reversed, in which case it may have been similar to the question I am asking.

So, will each row have the same values all across its width or will it change across the width - if so, how? What is supposed to happen to the brown layers at the top and bottom? Are there other pictures and do they have these same five colours? — Mark Setchell, Nov 19 '15 at 20:27
@MarkSetchell: Thanks for your comment. Here are some clarifications in response to your question: `1. `I would like if the rows contain values corresponding to the small undulations as shown in the image, which means few indices of a row may contain a separate value corresponding to a separate color. For example, let's look at green layer: the rows corresponding to that color would contain same value except at some indices where navy color is intermixed in between. `2. `Let's assume the brown layers as a discrete brown color...so, there are now total of 6 colors. `3. `Yes, same colors. — , Nov 19 '15 at 20:58

Mark Setchell · Answer 1 · 2015-11-19T22:18:06.040

Not sure if this is close to what you want, but if I use ImageMagick which is installed on most Linux distros and available for OSX and Windows, I can run this command in the Terminal:

convert image.png -alpha off -colors 6 -sample 50x20! result.png

And, if I ask for the output in text, rather than as an image, I get this:

convert image.png -alpha off -colors 6 -sample 50x20! txt:

# ImageMagick pixel enumeration: 50,20,255,srgb
0,0: (47517,15011,17755)  #B93A45  srgb(185,58,69)
1,0: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
2,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
3,0: (52533,46293,16329)  #CCB440  srgb(204,180,64)
4,0: (52533,46293,16329)  #CCB440  srgb(204,180,64)
5,0: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
6,0: (47517,15011,17755)  #B93A45  srgb(185,58,69)
7,0: (52533,46293,16329)  #CCB440  srgb(204,180,64)
8,0: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
9,0: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
10,0: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
11,0: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
12,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
13,0: (52533,46293,16329)  #CCB440  srgb(204,180,64)
14,0: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
15,0: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
16,0: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
17,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
18,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
19,0: (52533,46293,16329)  #CCB440  srgb(204,180,64)
20,0: (47517,15011,17755)  #B93A45  srgb(185,58,69)
21,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
22,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
23,0: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
24,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
25,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
26,0: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
27,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
28,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
29,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
30,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
31,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
32,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
33,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
34,0: (47517,15011,17755)  #B93A45  srgb(185,58,69)
35,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
36,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
37,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
38,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
39,0: (47517,15011,17755)  #B93A45  srgb(185,58,69)
40,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
41,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
42,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
43,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
44,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
45,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
46,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
47,0: (17144,20406,18979)  #434F4A  srgb(67,79,74)
48,0: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
49,0: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
0,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
1,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
2,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
3,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
4,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
5,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
6,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
7,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
8,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
9,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
10,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
11,1: (17144,20406,18979)  #434F4A  srgb(67,79,74)
12,1: (17144,20406,18979)  #434F4A  srgb(67,79,74)
13,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
14,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
15,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
16,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
17,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
18,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
19,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
20,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
21,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
22,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
23,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
24,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
25,1: (17144,20406,18979)  #434F4A  srgb(67,79,74)
26,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
27,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
28,1: (17144,20406,18979)  #434F4A  srgb(67,79,74)
29,1: (17144,20406,18979)  #434F4A  srgb(67,79,74)
30,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
31,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
32,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
33,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
34,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
35,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
36,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
37,1: (17144,20406,18979)  #434F4A  srgb(67,79,74)
38,1: (17144,20406,18979)  #434F4A  srgb(67,79,74)
39,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
40,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
41,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
42,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
43,1: (17144,20406,18979)  #434F4A  srgb(67,79,74)
44,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
45,1: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
46,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
47,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
48,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
49,1: (52533,46293,16329)  #CCB440  srgb(204,180,64)
0,2: (52533,46293,16329)  #CCB440  srgb(204,180,64)
1,2: (52533,46293,16329)  #CCB440  srgb(204,180,64)
2,2: (52533,46293,16329)  #CCB440  srgb(204,180,64)
3,2: (52533,46293,16329)  #CCB440  srgb(204,180,64)
4,2: (52533,46293,16329)  #CCB440  srgb(204,180,64)
5,2: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
6,2: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
7,2: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
8,2: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
9,2: (47517,15011,17755)  #B93A45  srgb(185,58,69)
10,2: (17144,20406,18979)  #434F4A  srgb(67,79,74)
11,2: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
12,2: (17144,20406,18979)  #434F4A  srgb(67,79,74)
13,2: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
14,2: (17144,20406,18979)  #434F4A  srgb(67,79,74)
15,2: (17144,20406,18979)  #434F4A  srgb(67,79,74)
16,2: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
17,2: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
18,2: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
19,2: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
20,2: (17144,20406,18979)  #434F4A  srgb(67,79,74)
21,2: (17144,20406,18979)  #434F4A  srgb(67,79,74)
22,2: (17144,20406,18979)  #434F4A  srgb(67,79,74)
23,2: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
24,2: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
25,2: (17144,20406,18979)  #434F4A  srgb(67,79,74)
26,2: (17144,20406,18979)  #434F4A  srgb(67,79,74)
27,2: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
28,2: (17144,20406,18979)  #434F4A  srgb(67,79,74)
29,2: (17144,20406,18979)  #434F4A  srgb(67,79,74)
30,2: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
31,2: (52533,46293,16329)  #CCB440  srgb(204,180,64)
32,2: (52533,46293,16329)  #CCB440  srgb(204,180,64)
33,2: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
34,2: (17144,20406,18979)  #434F4A  srgb(67,79,74)
35,2: (17144,20406,18979)  #434F4A  srgb(67,79,74)
36,2: (17144,20406,18979)  #434F4A  srgb(67,79,74)
37,2: (17144,20406,18979)  #434F4A  srgb(67,79,74)
38,2: (52533,46293,16329)  #CCB440  srgb(204,180,64)
39,2: (52533,46293,16329)  #CCB440  srgb(204,180,64)
40,2: (52533,46293,16329)  #CCB440  srgb(204,180,64)
41,2: (52533,46293,16329)  #CCB440  srgb(204,180,64)
42,2: (52533,46293,16329)  #CCB440  srgb(204,180,64)
43,2: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
44,2: (52533,46293,16329)  #CCB440  srgb(204,180,64)
45,2: (17144,20406,18979)  #434F4A  srgb(67,79,74)
46,2: (17144,20406,18979)  #434F4A  srgb(67,79,74)
47,2: (52533,46293,16329)  #CCB440  srgb(204,180,64)
48,2: (52533,46293,16329)  #CCB440  srgb(204,180,64)
49,2: (52533,46293,16329)  #CCB440  srgb(204,180,64)
0,3: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
1,3: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
...
...
46,18: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
47,18: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
48,18: (17144,20406,18979)  #434F4A  srgb(67,79,74)
49,18: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
0,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
1,19: (52533,46293,16329)  #CCB440  srgb(204,180,64)
2,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
3,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
4,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
5,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
6,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
7,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
8,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
9,19: (52533,46293,16329)  #CCB440  srgb(204,180,64)
10,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
11,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
12,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
13,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
14,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
15,19: (47517,15011,17755)  #B93A45  srgb(185,58,69)
16,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
17,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
18,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
19,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
20,19: (47517,15011,17755)  #B93A45  srgb(185,58,69)
21,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
22,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
23,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
24,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
25,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
26,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
27,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
28,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
29,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
30,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
31,19: (47517,15011,17755)  #B93A45  srgb(185,58,69)
32,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
33,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
34,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
35,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
36,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
37,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
38,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
39,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
40,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
41,19: (52533,46293,16329)  #CCB440  srgb(204,180,64)
42,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
43,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
44,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
45,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
46,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
47,19: (17144,20406,18979)  #434F4A  srgb(67,79,74)
48,19: (50918,44309,39507)  #C6AC9A  srgb(198,172,154)
49,19: (52533,46293,16329)  #CCB440  srgb(204,180,64)

Maybe this will get you started...

Adding -median 7 after -colors 6 decreases the small "noisy" patches somewhat.

convert image.png -alpha off -colors 6 -median 7 -sample 50x20! result.png

Note that there are Python bindings for ImageMagick too.

Another option may be to make your own little colour swatch of the 6 colours you want identified like this:

convert xc:"rgb(255,230,11)" xc:"rgb(16,128,90)" xc:"rgb(200,160,120)" xc:"rgb(200,0,40)" xc:"rgb(18,140,190)" xc:"rgb(14,17,116)"  +append swatch.png

and then to tell ImageMagick to map all the colours in the image to the swatch:

convert image.png -quantize sRGB -alpha off -remap swatch.png  -sample 50x20! result.png

If you want the palette of colours used, add -verbose info: at the end instead of the output image name, then you will get this - look at the last 6 lines - they show the indices of the colours:

convert image.png -alpha off -colors 6 -sample 50x20! -type palette -verbose info:

Image: image.png
  Format: PNG (Portable Network Graphics)
  Mime type: image/png
  Class: PseudoClass
  Geometry: 50x20+0+0
  Resolution: 59.05x59.05
  Print size: 0.84674x0.338696
  Units: PixelsPerCentimeter
  Type: Palette
  Base type: Palette
  Endianess: Undefined
  Colorspace: sRGB
  Depth: 8/16-bit
  Channel depth:
    red: 16-bit
    green: 16-bit
    blue: 16-bit
  Channel statistics:
    Pixels: 1000
    Red:
      min: 2 (0.00935378)
      max: 204 (0.801602)
      mean: 114.15 (0.447646)
      standard deviation: 85.674 (0.335977)
      kurtosis: -1.77949
      skewness: -0.174985
      entropy: 0.994611
    Green:
      min: 58 (0.229053)
      max: 180 (0.706386)
      mean: 127.02 (0.498117)
      standard deviation: 45.2218 (0.17734)
      kurtosis: -1.4666
      skewness: -0.2851
      entropy: 0.994611
    Blue:
      min: 64 (0.249165)
      max: 175 (0.687266)
      mean: 107.26 (0.420626)
      standard deviation: 43.1734 (0.169307)
      kurtosis: -1.48194
      skewness: 0.477601
      entropy: 0.994611
  Image statistics:
    Overall:
      min: 2 (0.00935378)
      max: 204 (0.801602)
      mean: 116.143 (0.455463)
      standard deviation: 61.2345 (0.240135)
      kurtosis: -1.14278
      skewness: -0.180352
      entropy: 0.994611
  Colors: 6
  Histogram:
       146: (  2,144,105) #029069 srgb(2,144,105)
       150: ( 10,125,175) #0A7DAF srgb(10,125,175)
       199: ( 67, 79, 74) #434F4A srgb(67,79,74)
       148: (185, 58, 69) #B93A45 srgb(185,58,69)
       200: (198,172,154) #C6AC9A srgb(198,172,154)
       157: (204,180, 64) #CCB440 srgb(204,180,64)
  Colormap entries: 6
  Colormap:
         0: ( 67, 79, 74) #434F4A srgb(67,79,74)      <--- index 0 in palette
         1: (185, 58, 69) #B93A45 srgb(185,58,69)     <--- index 1 in palette
         2: (  2,144,105) #029069 srgb(2,144,105)
         3: (204,180, 64) #CCB440 srgb(204,180,64)
         4: (198,172,154) #C6AC9A srgb(198,172,154)
         5: ( 10,125,175) #0A7DAF srgb(10,125,175)    <--- last palette entry

I have two comments: `1) `I don't see the array containing the numeric values (between 1-6) that correspond to those six different colors. `2) `The pattern of color does not correspond well to the original image in a lot of places. I think that's because we don't have any control over the algorithm which you suggested (though not sure what that algorithm is). However, I would prefer to keep the consistency of original image to the obtained image closer than shown here. In other words, the matching of colors between the two image is quite off. — , Nov 19 '15 at 21:53
The edited image looks much better. Which means, disregard my 2nd comment out of the two from earlier. — , Nov 19 '15 at 21:56
Is there some output matrix containing a single numeric value corresponding to each color for the final figure? — , Nov 19 '15 at 22:12
I have added some more information about matching the indices to the colours. — Mark Setchell, Nov 19 '15 at 22:19
`-verbose info:` gives the color pallete...okay. Where would those color pallete be saved? I looked for its python library and there are three suggested packages suggested [here](http://www.imagemagick.org/script/api.php#python)...which one among those three is the appropriate one? Also, what would be the corresponding function in python or how could I identify this function in any of those bindings? — , Nov 19 '15 at 23:12

Mark Setchell · Answer 2 · 2015-11-20T12:20:32.830

My other answer is getting a bit too long and I have had some more thoughts I wanted to share... look at this answer as a pre-processing step before the other one maybe.

If you convert your image in the HSL colorspace, like this:

convert image.png -colorspace hsl -separate hsl.png

You will get the Hue in file hsl-0.png like this:

and the Saturation in file hsl-1.png like this:

You can see that the Saturation image is lighter where your saturated red, green, blue and yellow colours are and darker where the brown is that you are not interested in. You can also see that the colours are fairly well differentiated in the Hue image. That leads me to the idea of using the Saturation channel as a mask and the Hue channel to split into colours...

convert image.png -colorspace hsl -channel s -threshold 80% -channel h -colors 12 -colorspace rgb result.png

Or, starting with your original image, you could go directly with this:

convert image.png -fx "u.saturation>0.8?u:0" -median 3 result.png

You could increase the -median 3 to 5 to remove more speckles.

So, I would say my current best shot at this is as follows:

# Make a swatch of the colours want in the output image
convert xc:black xc:"rgb(255,230,11)" xc:"rgb(16,128,90)" xc:"rgb(200,0,40)" xc:"rgb(18,140,190)" xc:"rgb(14,17,116)"  +append swatch.png

# Clean up image using Saturation to select coloured areas, median to reduce speckles, then remap to swatch created above and resample down to 50x20
convert image.png -fx "u.saturation>0.8?u:0" -median 3 -remap swatch.png -sample 50x20! result.png

Now print the palette indices - which is what you actually asked for - I know how to do this easily in PHP:

#!/usr/local/bin/php
<?php
// Read in a palettised image
$im = imagecreatefrompng("result.png");

for ($r = 0; $r < imagesy($im); $r++) {
    printf("Row %d: ",$r);
    for ($c = 0; $c < imagesx($im); $c++) {
        $pixel = imagecolorat($im, $c, $r);
        $index = $pixel & 0xFF;
        printf("%d ",$index);
    }
    printf("\n");
}
?>

Output

Row 0: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Row 1: 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Row 2: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 2 0 0 0
Row 3: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 1 1 1 1 1 1
Row 4: 0 1 1 1 1 2 1 3 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 1 0 0 1 1 1 1 0 2 2 0 0 2 0 1 0 0 0 4 0
Row 5: 0 2 2 2 2 2 2 2 2 2 0 0 4 5 2 2 5 2 5 2 2 2 2 2 2 2 2 2 5 2 2 2 2 2 2 2 2 2 2 0 0 0 0 0 0 0 0 0 2 2
Row 6: 0 2 2 2 2 2 2 2 2 2 5 2 2 2 2 2 2 2 2 2 2 2 5 2 2 5 2 2 5 5 2 4 4 2 5 2 5 2 5 2 2 2 2 2 2 2 2 2 2 2
Row 7: 0 0 0 2 0 0 2 2 2 2 2 2 2 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 2 2 0 0 2 0 2 2 5 2 2 2 2 5 5 2 5 5
Row 8: 0 2 2 2 2 2 2 2 2 2 2 2 2 2 5 5 2 5 5 5 2 5 2 2 2 2 5 2 5 5 5 2 5 5 2 2 2 2 2 2 2 2 2 2 2 2 5 5 5 4
Row 9: 0 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5
Row 10: 0 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 0 0 4 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5
Row 11: 0 4 0 0 1 1 1 1 1 1 1 1 1 1 1 3 3 3 3 3 3 3 3 1 3 3 3 3 0 0 0 0 0 4 0 2 5 5 5 5 5 5 5 5 0 0 5 2 5 5
Row 12: 0 1 1 1 1 1 0 3 3 1 3 1 3 3 3 3 1 3 3 3 3 3 5 3 3 3 5 3 3 3 3 0 0 4 0 0 0 3 3 5 3 1 3 5 3 3 3 3 3 3
Row 13: 0 3 3 1 1 1 3 3 3 3 3 3 3 3 3 0 2 0 0 0 0 0 5 0 5 0 3 3 3 3 3 3 3 3 3 3 3 3 1 3 3 3 3 3 3 3 3 3 3 5
Row 14: 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 3 3 0 0 0 3 3 3 3 3 3 3 3 0 0 0
Row 15: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 4 4 4 4 4 4 4 0 0 0 0 0 0 0 0 0 4 0 0 0
Row 16: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Row 17: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Row 18: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4
Row 19: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Mark, thanks for this. I am more inclined to use python and I think I have sort of figured how to do it using that. If your answer were to use python then it could have been really helpful...thanks. — , Nov 20 '15 at 14:03
Yes, I understand that but I don't speak Python and my answer should give you a concept and a technique that will get you started. Maybe Eric (@emcconville) will feel like and be able to translate the concept into Python for you... — Mark Setchell, Nov 20 '15 at 17:58

Assign scalar value to 'RGB' of few dominant colors of an image using python

2 Answers2