I am in the early stage process of

[justafarmer]

developing a computer program and/or I-phone app.

Basic premise

I am creating a database of images of routinely counterfeited coins and die varieties. These images will have 2 standard (landmark) anchor points embedded in them. The standard anchor points are easily located points based on each coin series obverse and/or reverse designs. It will allow in-take of a new image (target image) from a camera, scanner, etc and plot the standardized anchor points on it. Then with a click of a button it will import selected images from the database and automatically scale and overlay these images over the target image based on the anchor points.

Suppose you are at a coin show and are considering a 1909-s VDB from a dealer. You can take a picture of the coin with your I-Phone, plot the anchors and automatically overlay images of the 4 different 09-s vdb dies stored in the database.

What do you think? Worth the effort?

[Dick]

I am not familiar with the I-phone, or I-POD, but with your trend of thought, it seems very good for two people: You, the buyer, and also the Dealer, in that both will know if the coin in question is genuine, ot not. Sounds practical to me.
Kurt, you are always checking the shows, what do you think?
Dick

[justafarmer]

Don't think I would rely upon it with 100% confidence that a coin is authentic but could rely upon it with absolute confidence in determining a coin is fake.

[Bob P]

The app could have its benefits, but I think because of the limited applicability, that it might not be worth a lot of effort at this time. You would have to consider how many collectors there are who could benefit from this app. Normally, most collectors looking for something specific enough to be counterfeited are aware of the counterfeit possibilities, and either buy TPG slabs or have other ways to check authenticity. Those collectors also must have an I-phone or I-pod, and must be willing to use the app that might or might not give the correct interpretation of the coin examined. In my humble opinion, I see it as a very limited application that most collectors wouldn't or couldn't use. I would trust the eyes of an experienced grader/authenticator/attributer over a machine run app any time

[Dick]

Now If I were to try to put together, a complete set of "counterfiet coins", then that would save me a LOT of money! Or would it? The way the counterfieters are getting more expertice in their "trade", I have to wonder..... They would probably charge the same price, and then I would have to pay a TPG to show me the genuine ones. Sheesh!

[coppercoins]

You're going to have a HUGE problem with user limitations. Many people just don't get all the intricacies of detail needed to perform graphic overlays. Angle, zoom, perspective, lighting...a LOT of things that matter.

You'd be best off developing the app. so that it could serve as a mere identification guide and leaving it at that. I would think if you centered the "fat" of the app on good quality zoom-able images of the counterfeits and geniune coins, you'd come a lot farther.

Either way, it soulds like a good step in the right direction!

[eagames]

I've always thought this could be done using similar software to those little fingerprint readers.
They learn a print then recognize it again.

Only problems are that coins have different toning and wear but I'm sure the software could be tweaked to ignore everything but the points of interest that need to ID the coin or even ID a variety.

Sort of like the software that IDs people based on images at airports and casinos.

It might not work today but this junk gets better and cheaper each year so I bet the day for it will come sooner than we think.

[Dick]

Ed, I don't know anything about that software, but I did have a fingerprint reader on a computer that I had, when I first started looking at them, and trying to learn how they operated. It was a Zenith, I believe. I was partial to the Zenith, because one compuuter I had had a monitor that could switch from computer monitor, to a TV receiver.
As far as using it to recopgnize a particular coin, or coin variety, it would have to have a lot of latitude in the definition of any particular coin, variety, or in the csae of several coins the same size, and composition, but still different countries, would take some special software, capable of a certain amount of metal composition. For example the /canadian cent, and the US cent, arethe same composition, but are not the same coin. They will have their work cut out for them, before they find a solution, that is infallible.