Speckle Debugging

So there appears to be a significant number of particles that are receiving RGBA values of (0, 0, 0, 0).  I wrote a quick C++ code to separate the black particles out of an initial set. 

Here are the stats I got off of that test:

Initial Set                            Black                              %                          Location
13914                                    9997                               70%                     Mesh Edge
217088                                  9423                               4.3%                    Mesh Center
1737411                               105750                           6.1%                    Total Mesh

So it looks like the script fails more on the outer edges of the mesh, with an overall failure of about 6%.  I tried running these black particles back through the culling script.

Upon reculling inside of maya, ie running the script through the script editor, all were assigned a color. 

Upon reculling outside of maya, ie running the script in the terminal through mayapy, only 125 were black (1.25%).

So what's going on????  When running the script I am getting this error:

TIFFReadDirectory: Warning, /u/klitaker/Desktop/particle_generation/sourceimages/s106-3d-layer-BLUE_flat2125.tif: wrong data type 7 for "RichTIFFIPTC"; tag ignored.

It might be possible that this is causing the texture lookup operation to fail in the terminal and not to fail in the maya interface.  Which is really weird/stupid.