08-12-2012, 08:14 PM
Quote: Since we know the photo has been faked .....
(h1) If Doorman is Oswald, what is the probability that they would share the same right ear, the same left eye, and the large number of other features of their clothing and build, which, given the number, turns out to be extremely high approaching one.
(h2) If Doorman is Lovelady, what is the probability that he would have Oswald's right ear, Oswald's left eye, and such, and that Lovelady would repeatedly deny that he was wearing Doorman's shirt to the FBI and others, which approaches zero.
Jim... please get some help already... you cant even state the QUESTION correctly....
"IF" is a conditional statement.... If my grandmother had balls, she'd be my grandfather... doesn't make it so since the CONDITION is not possible.
The way you state your analysis is completely wrong... there is no SHOULD in probability... there are results that lead to a conclusion....
To correctly state your question:
(h1) What is the probability that, if two images CAN BE SHOWN TO "share the same right ear, the same left eye, and the large number of other features of their clothing and build", they are of the same person?
You cannot assume the conclusion in the question. Furthermore, as I've tried to explain repeatedly to you, you have to mitigate all the NON MATCHING ITEMS to 0% probability... as well as assign a % to each of your 50 items....
So, to address this probability problem we need a few more bits of data. Since we are looking at two IMAGES... what is the proof that YOU are 100% correct about these features? Have you used photogrammetry* to determine they are EXACTLY the same? or is this eyeballing?
"the same right ear, the same left eye" - so first off you feel the image quality is good enough to compare and measure - or is this eyeballing?
You are of the opinion that you are 100% correct about those TWO ITEMS.... yet you offer nothing to support such a conclusion beyond "eyeballing"
How can you be 100% sure about a MATCH Jim other than producing the math that shows the dimensions of such items IN THE PHOTO are indeed "exactly the same" If I can point out ANY differences... 100% surety drops... and continues to drop until you reach a RELIABLE % for you conclusion. Just saying so does not make it so....
Read this again Jim... carefully: Even with 10,000, 100% documented MATCHES, a single 100% MISMATCH (ie blue v red shirt) completely negates the 10,000 items and proves they are NOT the same item or person. In addition, any reduction in any of the 10,000 100% conclusions, reduces the probability of them matching... As I showed you, even a 50/50 of your 50 items creates astronomical odds AGAINST them being the same
So Jim.... btw, this is not personal Jim... this is black and white mathematics... argue the MATH, not the messenger.
The ONLY WAY to prove that the matching approaches a 100% probability is to PROVE each item's 100% conclusion. to Prove YOUR OBJECTIVE CONCLUSIONS using subjective methods - ie photogrammetry*. until then you're pissing in the wind and just stomping around and waving your hands WANTING us to believe your EYES are better than math as well YOUR CONCLUSIONS before the experiment even begins....
"Since we know the photo has been faked...." is a conclusion, not a priori knowledge... you have to PROVE that Jim.... and you haven't. Have you computed the PROBABILITY the photo was altered? do you even know HOW to attempt that?
From what I've seen, you repeatedly massacre the realities of probability with an arrogance that is legendary. And then promote such lies as indicative of your conclusions...
but you can't fool all the people all the time.... only the sheep you hope to convince
no matter how often you lean in and speak louder hoping for comprehension of your completely wrong, childish gibberish.
Talk to ANY of your math friends Jim and find a clue already... your fighting to maintain that "the world is FLAT" makes you out to be a fool...
Was that your desire?
*Photogrammetry is the practice of determining the geometric properties of objects from photographic images. Photogrammetry is as old as modern photography and can be dated to the mid-nineteenth century.
In the simplest example, the distance between two points that lie on a plane parallel to the photographic image plane can be determined by measuring their distance on the image, if the scale (s) of the image is known. This is done by multiplying the measured distance by 1/s.
Algorithms for photogrammetry typically express the problem as that of minimizing the sum of the squares of a set of errors. This minimization is known as bundle adjustment and is often performed using the LevenbergMarquardt algorithm.