|
- 07.29.10 -
menu
- types
- history
|
|
Hand print OCR
Computer systems for recognizing printed text have enjoyed a lot of success in
the recent years. Among these are the input device for personal digital
assistants such as those running Palm OS. The algorithms take advantage of
the order, speed, and direction of each individual line's segments at input
being known. Also, the user can retrain the computer to recognize any particular
character. This method cannot be used in software that scans paper documents, so
accurate recognition of hand-printed documents is still largely an open problem.
Accuracy rates of 80% to 90% on neat, clean hand-printed characters can be
achieved, but that accuracy rate still translates to dozens of errors per page,
making the technology useful only in very limited contexts. This variety of OCR
is now commonly known in the industry as "ICR" or intelligent character
recognition for short.
Cursive OCR
Recognition of cursive text is an active area of research, with recognition
rates even lower than that of hand-printed text. Higher rates of recognition of
general cursive script will likely not be possible without the use of contextual
or grammatical information. For example, recognizing entire words from a
dictionary is easier than trying to parse individual characters from script.
Reading the Amount line of a cheque (which is always a written out number) is an
example where using a smaller dictionary can increase recognition rates greatly.
Knowledge of the grammar of the language being scanned can also help determine
if a word is likely to be a verb or a noun, for example, allowing greater
accuracy. The shapes of individual cursive characters themselves simply do not
contain enough information to accurately (greater than 98%) recognize all
handwritten cursive script.
Music OCR
Early research into recognition of printed sheet music was performed in the mid
1970s at MIT and other institutions. Successive efforts were made to localize
and remove musical staff lines leaving symbols to be recognized and parsed. The
first proprietary music-scanning program, MIDISCAN, was released in 1991. Three
proprietary products are now available but music OCR software does not recognize
handwritten scores.
MICR
One area where accuracy and speed of computer input of character information
exceeds that of humans is in the area of magnetic ink character recognition,
where the error rates range around one read error for every 20,000 to 30,000
checks.
Other
A particularly difficult problem for computers and humans is that of old church
baptismal and marriage records containing mostly names. The pages may be damaged
by age, water or fire and the names may be obsolete or contain rare spellings.
Another research area is cooperative approaches, where computers assist humans
and vice-versa. Computer image processing techniques can assist humans in
reading extremely difficult texts such as the Archimedes Palimpsest or the Dead
Sea Scrolls.
Generally, for more complex recognition problems neural networks are commonly
used as they generally can be made indifferent to both affine and non-linear
transformations.
A related area is raster to vector conversion, converting bitmap images (for
example, maps including drawings, text, and map symbols) into vector graphics
that are easier to work with.
More Resources:
|
Related:
omni page ocr, mac osx ocr free, ocr b font, use of ocr software by the usaf |
