isDistinct problem with B&W photos

[Havock94]

Hi, I've noticed a problem in the isDistinct method while comparing colors in a black and white picture.
I was comparing black (RGB 0, 0, 0) and white (RGB 1, 1, 1) and I noticed that the method only compares colors channels by subtracting R and G channels and then R and B channels.

So 0 - 0 < threshold and 1 - 1 < threshold, resulting in white and black colors being considered as not distinct, returning only one color while processing my image.

I then added a control for luminance as well, so if the colors are both in a grey scale AND their luminance is within the threshold, then consider them to be similar, otherwise not.

// Check for grays
if (abs(r - g < 0.03f && fabs(r - b) < 0.03f)) {
if (fabs(rc - gc) < 0.03f && (fabs(rc - bc) < 0.03f)){
//Also check for luminance!!
if(fabs([self luminance] - [color luminance]) < threshold){
return NO;
}else{
return YES;
}
...

After this, the color extraction works for B&W images as well, returning a grey scale palette.
I hope this can help you!

[timominous]

@Havock94 Feel free to submit a PR on this :)