the problem is, on each new pixel index you're allocating a new array, and saving the current brightnessvalue to each index.
but on the very first iteration the array has the length 0, because loc is 0 too. this causes the error message.


heres a shorter version, to save all brightnessvalues

1. void setup() {
2.   PImage img = loadImage("http://processing.org/img/cover/onefive.png");
3. 
   
4.   img.loadPixels();
5. 
   
6.   float[] brightness_values = new float[img.pixels.length];
7. 
   
8.   for (int i = 0; i < img.pixels.length; i++) {
9.     brightness_values[i] = brightness(img.pixels[i]);
10.   }
11.  
12.   float min_bright = min(brightness_values);
13.   float max_bright = max(brightness_values);
14.   float range = max_bright - min_bright;
15.  
16.   println("min_bright = " + min_bright);
17.   println("max_bright = " + max_bright);
18. }

and if you really just need the minimum/maximum brightness values, the following will be faster than saving all the values to an array and then getting min/max.

1. void setup() {
2.   PImage img = loadImage("http://processing.org/img/cover/onefive.png");
3.   img.loadPixels();
4. 
   
5.   float min_bright = Float.MAX_VALUE ;
6.   float max_bright = Float.MIN_VALUE ;
7. 
   
8.   for (int i = 0; i < img.pixels.length; i++) {
9.     float bright = brightness(img.pixels[i]);
10.     if( min_bright > bright ) min_bright = bright;
11.     if( max_bright < bright ) max_bright = bright;
12.   }
13.  
14.   float range = max_bright - min_bright;
15.   println("min_bright = " + min_bright);
16.   println("max_bright = " + max_bright);
17. }

That's great!