Documentation about palettes

[martinber]

Hello, I'm implementing false color in my project using the same palette method you are using. If possible, I would like to know how are you generating the N19-HRPT-Falsecolor.png palette (or to know from where did you get it).

Also, adding an acknowledgement to my program pointing here (Xerbo/aptdec) is ok?

[Xerbo]

The N19 false color palette was made from a single HRPT pass, its just a LUT of ch1/ch4 to ch1/ch2.

I will have a search for the quick and dirty python script I used to make it but its probably long gone.

[Xerbo]

I couldn't find the original so heres a version I threw together in 10 minutes:

import sys
from PIL import Image
import numpy as np

if len(sys.argv) != 5:
    print("Usage:", sys.argv[0], "ch1.png ch2.png ch4.png out.png")
    exit();

ch1 = Image.open(sys.argv[1])
ch2 = Image.open(sys.argv[2])
ch4 = Image.open(sys.argv[3])
ch1_px = ch1.load()
ch2_px = ch2.load()
ch4_px = ch4.load()

sum   = np.zeros((256, 256, 3))
count = np.zeros((256, 256))
for y in range(ch1.size[1]):
    for x in range(ch1.size[0]):
        ch1x = int(ch1_px[x, y] / 256)
        ch2x = int(ch2_px[x, y] / 256)
        ch4x = int(ch4_px[x, y] / 256)
        sum[ch1x, ch4x, 0] += ch2x
        sum[ch1x, ch4x, 1] += ch2x
        sum[ch1x, ch4x, 2] += ch1x
        count[ch1x, ch4x]  += 1

out = Image.new('RGB', (256, 256));
out_px = out.load()

for y in range(256):
    for x in range(256):
        if count[x, y] != 0:
            r = int(sum[x, y, 0]/count[x, y])
            g = int(sum[x, y, 1]/count[x, y])
            b = int(sum[x, y, 2]/count[x, y])
            out_px[x, y] = (r,g,b)

out.save(sys.argv[4])

If the images are from SatDump you will need to replace the xxx / 256 with xxx / 240 and if you are using LeanHRPT you will need to disable calibration (just delete calibration.ini)

[martinber]

It took me time to look at it. Thank you very much!

In the end I didn't use it because I never received HRPT and I don't have images. But I like the method, it's a good idea