[Astronouth7303]

What the camera is sending is the bounding box of the blob.

x1 - minimum X value
y1 - minimum Y value
x2 - maximum X value
y2 - maximum Y value
mx - median X value (average of x1 and x2, IIRC)
my - median y value (average of y1 and y2)

There is also a confidence value that describes how "full" the bounding box is, IIRC. higher confidence values mean that more of the bounding box is taken up by the blob. You'll have to look up the exact formula for it.

You can also configure the camera to send you all the pixels, but this takes up a lot more time.