A way to filter out very light or dark colors

[EmilStenstrom]

I'm running colorgram on images of board games that I have. Many of them have a black or white box, which means the dominant color in many of the cases is just black. I would like to filter out those colors.

I see that you already calculate luma here: https://github.com/obskyr/colorgram.py/blob/master/colorgram/colorgram.py#L65 -- so what I'm asking for is way to filter out those with high or low luma.

Is this something that is a good fit for this library?

[DonaldTsang]

Maybe remove background color (dominant color) first?

[EmilStenstrom]

@DonaldTsang That would mean implementing something like this library myself... I would like help from this library for doing this, that is what this issue is about.

[sjmelchor]

@EmilStenstrom, I'm guessing you figured this out by now, but I found your comment while trying to solve the same issue. Here's what I ended up doing:

#running the colorgram tool as expected, I just set 10 as an arbitrary number of colors to extract

colors = colorgram.extract("your-favorite-jpeg", 10)

# on a scale from 0 to 255, anything with the lightness above 200 wouldn't work for me--but you could do the same in reverse for too dark.

def def too_light(trial_color):
  
      if trial_color.hsl[2] > 200:
          return True

#(colors.hsl[2], because lightness is the 3rd value in the hsl tuple)

#then later in my code, I made a list of the rgb tuples from the colors colorgram extracted, but only the colors that the function returned as False (ie, not too bright) made it into the list I used. 

I hope this helps someone! I'm new to GitHub and Python, so massive apologies if I didn't do it right. :/

[EmilStenstrom]

@sjmelchor I did, but thanks for adding more context, I love finding these nuggets when I look for solutions elsewhere. Here's my solution which depends on the color's "luma" which is a calculated value which accounts for how light different colors feel:

image = Image.open(io.BytesIO(image_data)).convert('RGBA')
try_colors = 10
colors = colorgram.extract(image, try_colors)
for i in range(min(try_colors, len(colors))):
    color_r, color_g, color_b = colors[i].rgb.r, colors[i].rgb.g, colors[i].rgb.b

    # Don't return very light or dark colors
    luma = (
        0.2126 * color_r / 255.0 +
        0.7152 * color_g / 255.0 +
        0.0722 * color_b / 255.0
    )
    if (
        luma > 0.2 and  # Not too dark
        luma < 0.8     # Not too light
    ):
        break
   else:
        # As a fallback, use the first color
        color_r, color_g, color_b = colors[0].rgb.r, colors[0].rgb.g, colors[0].rgb.b

    return f"{color_r}, {color_g}, {color_b}"