With the advent of CAD-CAM technology and laser/digital image processing, I wonder why this hasn't taken hold in the coin industry. I can see it now. Coin graders will use a base of MS69/70 and PR70coins and they are scanned for their mint strike, color, etc. Laser technology can analyse within 20 microns and it is far better than the human eye for objectively analysing surfaces and objects. Average in a number of MS70 struck and PR70DCAM coins (without mint marks and year) for a partiular coin and this will be your "given" database for coin A. Each part of the coin (with the exception of the mint mark and year) will have ts own 20 square microns storing data for height, depth, markings, etc. Now comes a coin for grading...it too is scanned and this scan becomes "Y". In other words, the difference between X and Y would determine the grade of the coin. The larger the difference, the lower the grade. The laser (picture Dr. Evil using quotations) can scan worn surfaces against normal surfaces as well as "see" nicks and marks not seen be the human eye. Too picky?...well adjust the settings to that of a human eye using a 7X magnifier. In other words, if the eye can't see it under normal grading conditions, then it should not be important. What about color, luster and all of that eye appeal stuff? Well, working for a color measurement company...(ever been to Home Depot to match paints with a sample you bring in?)...there is technology already out there that measures color for industries that demand accurate color measurement (auto, paint, print ads in publications, etc.). Using basic algebra, they are able to formulize colors so that x equals x...and the color is perfect. Color is measured in a controled area with a calibrated light source so that everything is equal. Even rainbow toned coins could be measured as the computer can easily pick up changes in RGB levels (red, green blue). Also, the data base of the graded coin can have its own fingerprint so that it can be indentified later against the stored data for this particular coin. I saw a patent for this type of grading in the early 90's, but the project of doing it versus the idea seems to be the issue. But then again, we will probably be arguing against the machine if we aren't satisfied with the grade our submitted coins received!
This is not a new thought. Computer aided grading or photo aided grading have been attempted and some services are still trying to work the system in. The idea would work for technical grading. Eye appeal makes a huge difference and is something that all mint state and proof coins need. No computer or photo can take the place of seeing the coin. This is exactly why so few coins are bought strictly sight-unseen. Despite who guarantees what, or what the photos look like, a coin must be seen to be truely appreciated. While I hope that one day, a true technical grade will happen, I just don't see how it will work. At least with what is available today.
Even if it were possible to use technology to consistently apply the same *technical* grade to a coin -- and at some point, it could be -- it still wouldn't reflect changing market tastes. At times, maybe toned coins are "in"; other times, maybe blast-white silver pieces are the in thing. And while two coins may already reach the same grade technically, you still need to see the coin to determine if it has the right eye appeal for you. In that respect, no grading service or magical whizzbang technical grading machine will ever fully be able to replace seeing the coin, examining it under magnification, swirling it around under a light and applying your own tastes and opinions on eye appeal in terms of valuing the coin. It can serve as a useful baseline, perhaps, but that's about it.
Apparently, something like this has been tried, although there is no mention of using lasers: ---------- 10 Myths of the Modern Coin Market Scott Travers - June 1, 1995 Myth No. 3: Certification services grade coins by computer. A few years ago, with much fanfare and hoopla, the Professional Coin Grading Service introduced The Expert, a computer which, according to claims in PCGS literature, would be able to grade coins accurately on a systematic basis. Initially, it was programmed exclusively to grade Morgan dollars. But after just a year or so of service, The Expert was "temporarily" retired, and the Irvine, California, certification service now grades coins strictly by means of human experts' eyes, not by computer. ---------- reference: http://www.pcgs.com/articles/article191.chtml
It is going to work someday...sooner than we think! The use of lasers are used today in a variety of industries that deal with custom services and appliances. This technology is in fact REAL. For instance, the dental industry is now using lasers to scan teeth whereby milling machines can exactly duplicate the tooth in a new restored material...in about 15 minutes while you wait. Height, depth, size, etc., are all scanned, assigned data, calculated and now made instantly. Of course, no two people have the same exact teeth and the same can be said about coins. Matter of fact, laser digital companies have perfected the use of strands of laser lines (versus the single dot we see in pointing devices) to scan surfaces "repeatedly" whereby changes in surfaces can be detected within 10 to 20 microns. Way smaller than the human eye! Imagine lasers scanning the surface of a Walking Liberty and givining each "pixel" area a value (x,y,z). Compare the scanned data to that of the "standard" and the computer using calculations for each area will aside a technical grade based on the collected data from 1 to 70! I have seen this technology work and spoke to the engineers about our little hobby and how much of an uproar takes place between a MS64 grade and MS65 grade...by grading experts with their subjective opinions. They told me that it can happen. The key is what will be considered the "perfect" coin that all others (the variable) will be compared against. That will be a debate all by itself, I bet you this...it is going to happen. At least at the very minimum, grades will eventually be assigned from a technical standpoint for strike, relief, etc. using this type of technolgy. PCGS may have tried this...but the key is how to accurately scan coin's surfaces for all of the important items that determine grade. I am sure PCGS knows coins, but laser scanning is something only a few understand.
You are right Ed. The number of people who can accurately grade coins is small. This is not to say that some can't grade some coins, but overall the ability to grade coins is an art lost on most. I have worked with people who use lasers to do all sorts of things. They are amazing in their craft. If this ever translates into a workable area of numismatics will remain to be seen. Grading is without a doubt, the most argued point of numismatics. The major grading companies serve an important service, with regards to authenticating, and attributing. I believe that they honestly try to do their best on any given day. In order for any grading system to work properly, we must get away from market grading. The preservation of a coin does not change for the better. It can only remain the same or get worse. Despite what the value of the coin is, or which pedigree it has, the grade should be based on that particular coin.
But, like all technology, more and more are understanding it everyday, and it will be applied to more esoteric disciplines (coin collecting) in the not too distant future. In fact, if anyone is willing to bankroll me, I'll get started on it tommorow
Buyers and sellers would have to learn how to interpolate! "Let's see...MS-64.673246...that's a little more than 2/3 of the way from 64 to 65...this Greysheet shows $1,000 in 64 and $3,000 in 65...must be about...think...$2,350 or so..."
Laser technology has been in use for this purpsoe for quite some now. And yes it is very effective and very difficult to detect.
Excellent article. Scanning is the way but most people think pictures. Scanning gives 3 dimensional to look for wear.
Look what happened to judging in figure skating. Ok, a 0.03mm scratch is a manditory 0.5 point deduction.
I think for the system to work you would have to scan thousands of coins, not just a reference coin. Take the coins that have already been graded and amass a database. That way the computer not only knows the "perfect" coin, but it has a huge sampling of every grade in between. When thousands of grades are averaged out they become accurate. That's basically what they have to do for voice recognition. You can't use enter one audio file of a word for a computer, you have to enter thousands of files of thousands of people saying that word. That way the computer can have a database of comparison. To the computer, me saying the word "coin" and you saying the word "coin" are just a different bunch of binary digits. I think if this was employed as a way to give a base grade, with a maximum deviation stipulated, grading could get more accurate.
The basic patents on computerized grading belong now to PCGS. You can find them on the appropriate government public databases. They have had them quite a while, and could have made it part of their laser scanning for identification, but did not publicly label or release. This patent data on google gives a large number or original patents granted, to find current owner if any, you would have to research further. Some go back to the 90's/ http://www.google.com/patents/US5224176
Under Alice, (https://en.wikipedia.org/wiki/Alice_Corp._v._CLS_Bank_Int'l), you can't patent an abstract idea. Especially you can't just throw a computer into an existing process and claim protection. The ruling continued with these points: A mere instruction to implement an abstract idea on a computer "cannot impart patent eligibility." "[T]he mere recitation of a generic computer cannot transform a patent-ineligible abstract idea into a patent-eligible invention." "Stating an abstract idea 'while adding the words "apply it"' is not enough for patent eligibility." "Nor is limiting the use of an abstract idea to a particular technological environment." So read the claims, ignoring the use of a computer and tell me if the patent is an abstract idea or new and novel: as is always true with patents, the gap between what it seems to say and what it actually says is large. Only the claims matter... Drop the computer and isn't that just teaching somebody how to grade from photographs?
So perhaps the software patent atmosphere has evolved into a place where such a process isn't patentable and this one would evaporate in court. I get an odd musty feeling about posting in a thread which was just raised after 11 years in the grave, especially when we're already discussing the topic live in another.
Lasers are used in all kinds of sports and other applications. With computers and systems improving, it is just a matter of time when computer is one of the tools you use in the grading process.
