Comment by Steve Zarelli on July 5, 2011 at 1:16pm

The ballpoint is pretty easy... much easier than I thought. (At least I think it is) :-)

 

3 - x-first

4 - o first

8 - x first

9 - x first

 

The marker is much tougher than anticipated. But, I think it would be easier on a glossy photo because it  shows streaks and the ink sits on the surface as opposed to "sinking in."

Comment by sling on July 5, 2011 at 2:47pm

A note regarding comments on scans!  The reader should be aware that the digital eye is hundreds of time in magnitude better than ther human eye in comparing handwriting--however proper software is needed.  A good quality home scanner has sufficient digital copying resolution to analyze handwriting if the proper software is used.

The human eye cannot see proper resolution beyond 300x magnification.  The digital eye can, with the right software, see a million times magnification.  And the software also allows a highly sophisticated, multiple type handwriting comparison to valid exemplars.

Digital scans are often considered not that useful, but in fact can be much better than the human eye used properly. Word.

Comment by sling on July 5, 2011 at 3:05pm

One note re digital eye comment- software allows the HUMAN eye to see above its natural limit of 300x to in excess of a million times magnification.

Also, if there is an optical way of telling which lines of the circles and x's examples, the digital eye will spot it faster and better than human eye.

Comment by Steve Cyrkin, Admin on July 5, 2011 at 3:14pm

Pet, what software do you recommend?

Comment by sling on July 5, 2011 at 5:07pm

Historically, this software has not come cheap- but affordable by labs and gov,t agencies.  However, now there are several universities working in this area and I think are offering open source software in this area.  Just do a search of such Univ programs and you should find some quickly.

Comment by sling on July 5, 2011 at 5:21pm

One note regardingh digital eye.  The software applications may be specific to handwritting analysis subject, or the softfware may be derived out of shape analysis for things such as bacteria shapes.  In both cases, the resulting computer generated results should always be read by a human and finalized by a human. Certainly this data should wiegh heavy in the human's judgement, but the human may also know of other info that may supercede all other data.  That other data being absent, then the computer results are generally the best you can have given you have authentic exemplars in the digital database-

 

just scan!

Comment by sling on July 6, 2011 at 6:22am

Stereoscopic  scanners have primarily been used to create the ILLUSION of 3d in 2d objects.  they aren't great for coparison, such as signatures, for various reasons including the fact that they only compare to ONE other object or signature in this case.

 

If you are comaring to 20 OTHER  exemplars, then your statistical reliablility is much greater than one exemplar (1/square root of N where N is the sample size is one way to calculate margins of error)

Comment by bobby greenlee on July 7, 2011 at 10:21am

wow u sound great look me uo

Comment by bobby greenlee on July 7, 2011 at 10:21am

up

Comment

• ‹ Previous
• 1
• …
• 5
• 6
• 7
• Next ›
• Page