The "Q-5" Process- It's Just Math!

My technique to establish "face ID" which I call the Q-5 process, short for Five Point Quintangulation, is a simple mathematical technique, but totally dependent on computer technology, namely Photoshop. It is also dependent on having two photographs of faces taken from basically the same angle. I pick out at least five points on a face, mark them with a color, then lift them and apply them to a different face, to establish the degree of similarity. It starts with the eyes. A person's eyes don't change much during his or her lifetime. It is critical to mark just the iris of the eyes, capturing their size and the distance between them; the nostrils; the upper lip, which doesn't change much over a lifetime, (until the upper teeth fall out) and the chin. All are marked as precisely as possible. It helps, if possible, to mark the top and bottom of the ears, understanding that ears do grow and sag over a lifetime. The chin is somewhat stable over the life of a person... again, until they start to lose teeth. I mark the forehead, at the part or the highest points of the hairline. I might mark outstanding cheek bones, or eyebrows. But anything made out of hair won't be too reliable! Facial hair is always a problem, and requires some guessing.

When the face-marks from the original face are lifted, they are carefully placed to align with the new face perfectly. Eyes must be laid precisely on eyes. Suddenly the likeness is affirmed or negated. That is because every person has a pretty unique mathematical pattern on his face; the distance between his or her eyes, then the distance between the eyes and the nostrils, then distance next down to the upper lip... and the chin and so on. If any one of those distances does not match, it is merely a look-alike. A doppelganger. The variability of the size of the eyes and distance between them is the crux of the whole technique, and it does not suffer doppelgangers. Except for identical twins, very few "look-alikes" can pass the test. But some do.

Then the photograph of the wanna-be look-alike has to pass other screens, such as sex, race, period clothing, dark eyes or light eyes, width of nose and ear construction, which usually eliminates the close calls. Many a doppelganger goes down in flames when the ear construction is compared. Many of my would-be purchases have been stopped in their tracks because the ears were off. Here (above) the chin was the big failure, so Billy Crystal can now rest easier knowing that he is not Ronald Kaufman, a dangerous criminal... but when Kaufman cut his hair... the similarity might have caused Crystal to be picked up by the police!

Children are very difficult to measure because they are constantly changing during their whole existence.

Computer facial recognition does a fair job, but is only as good as the technician using it. People who consult free face recognition services usually do not have enough grasp of human anatomy or aging, historical context, changes in photography technology over the last century, or the constant evolutions of fashion in order to discern a doppelganger from an actual historical discovery. Hence all of the nonsense on Ebay, where sellers have obtained the "authentication" from "Face ID" and now offer rare prizes of antique photography hoping to be worth millions.

My grading is totally subjective. Since I can sometimes lay a pattern from an original person on top of another different photo of the same person, and not get 100% match, I say that nothing ever gets better than a 95% match. On a close call, I work backwards from 95%, deducting a point or two for every inconsistency. I consider a 90% match a probable but shaky identification. I will eventually have a page where I share some of my better doppelgangers which scored less than 90%. They stood tall in my collection before I began to use the Q-5 technique.

The old saying was that "photographs don't lie." Of course, they sometimes did, especially when doctored. But I feel I can claim that mine do not lie, because Geometry doesn't! Mathematical proportions are almost fool-proof. Knowledge of history, human anatomy, and early photography processes helps to make my evaluations very reliable.