Suppose I have an RGB24 image in which the color of each pixel evenly encodes a (raw) data byte. Scheme: Channel red encodes Bit 7:6 (R: 0-63 = 0b00, 64-127 = 0b01, 128-191 = 0b10, 192-255 = 0b11) Channel green encodes Bit 5:3 (G: 0-31 = 0b000, 32-63 = 0b001, 64-95 = 0b010, 96-127 = 0b011, ...) Channel blue encodes Bit 2:0 (B: 0-31 = 0b000, 32-63 = 0b001, 64-95 = 0b010, 96-127 = 0b011, ...) Example: RGB = [150, 40, 94] represents (raw) data byte 0b10001010. Question: I would like to use the Python Imaging Library (PIL) to decode my image. If I got it right, then I would need to define a palette and apply it to my image using the quantize() method (as described here). However, I don't understand how the mapping of the colors does work. How does the palette translate the original 24 Bits of color information into 8 Bits? Finally: How would I need to set up my palette so that the decoding works according to my scheme? Continue reading...
