BBM
Unfortunately, in my time here I've seen at least a dozen perfect photographic matches that turned out not to be matches at all, and a number of matches where it's being generous to say the recon and the photo were approximations of each other, even when the recon was based on a post-mortem photo as this one is.
See, this is interesting because it all comes down to how our individual, and very different brains, perceive what we see as well as to how we classify things. And of course all that is at the very essence of what we do here. I want to ask, what did you mean in terms of "perfect photographic matches"? Did you mean that the two people looked exactly alike? Or was there a quantitative comparative process performed on the faces? And by that I mean something like what I did, or some kind of other measurable comparison?
I'm with you on the reconstructions. Just looking at a case where numerous reconstructions are formed, it's always amazing how greatly they can differ. I think it was the Lady in the Dunes case that had about ten to twelve recons by a few different artists, with the intent of portraying age regression. Between all of the recons it looked like there were three or four distinctly different people. It can be discouraging at times when you don't know if the recon comes even close to looking like the victim did.
I know that when they have a recognizable body they often photograph the body and then photoshop it to make it look less disturbing, and more like a drawing or a sketch. In those cases, the visual information is there and so a competent reconstructionist should be able to present it without affecting it in any way.
I am on the fence about that technique, however. In some ways I would rather see the unaltered morgue photo because, with the understanding that damage to the body could have altered certain aspects of it, I know that it is as close as we can get to an accurate comparison. Or at least, in most cases. And I think the chances, with all else being equal, are greater that we will be able to identify the person. However, some of the post-mortem photos are so painful to look at...and I especially wouldn't want that person's loved one to have to live the rest of their lives with that image in their head.
I've come across many of those too, where the match appears solid but isn't. Case in point, a missing 35 year old man named Edwin and a reconstruction of a victim wearing a Superman shirt. I can't remember all the details. But both were from the same state, missing and found dates perfectly plausible, the reconstruction appears IDENTICAL to Edwin. However, DNA ruled them out. On the other hand, though, one victim was recently matched (was it Paulette Jaster, Houston Mom's match?) by a specific pattern of three freckles common to both cases). So, an entire face match (quantitatively matched, not just by sight) is very strong evidence of identity.