Image analysis – Applications – Mail processing
Reexamination Certificate
2007-10-23
2007-10-23
Bella, Matthew C. (Department: 2624)
Image analysis
Applications
Mail processing
C382S161000, C382S310000
Reexamination Certificate
active
11087504
ABSTRACT:
In the present invention information is first defined as a complete description of the meaning of the content, in addition to the required layout of the characters, words, delivery addresses or images of the entire delivery surface, which contains the delivery addresses, as sample elements. The images of the sample elements belonging to the defined reference information are then generated with the aid of type libraries for the defined character fonts and languages and said images are then stored, after having been assigned to the reference information, in the form of a pixel representation for the sample. The images of the sample elements that have been generated in the previous step are subsequently varied by means of known image processing algorithms to create at least one statistical variation corresponding to the previously collected, grouped deliveries, according to the defined sample scope.
REFERENCES:
patent: 5774588 (1998-06-01), Li
patent: 5796410 (1998-08-01), Baird
patent: 6269171 (2001-07-01), Gozzo et al.
patent: 0 461 793 (1991-12-01), None
patent: WO 00/70549 (2000-11-01), None
Govindan V K et al: “Artificial Database for Character Recognition Research” Pattern Recognition Letters, North-Holland Publ. Amsterdam, NL. vol. 12, No. 10, Oct. 1, 1991, pp. 645-648, XP000230128, ISSN: 0167-8655, Section 2.
“Optical Character Recognition Test”: IBM Technical Disclosure Bulletin, IBM Corp. New York, US, vol. 36, No. 3, Mar. 1, 1993; pp. 201-202, XP000354749; ISSN: 0018-8689, the whole document.
Miyojim M et al: “Synthesized images for pattern recognition” Pattern Recognition, Pergamon Press Inc. Elmsford, N.Y. US, vol. 28, No. 4, Apr. 1, 1995, pp. 595-610, XP004013174, ISSN: 0031-3203, abstract; figures 11,12.
Gierszewski Gerd
Lindwurm Rainer
Miletzki Udo
Bella Matthew C.
Siemens AG
Tucker Wes
LandOfFree
Method for generating learning and/or sample probes does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for generating learning and/or sample probes, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for generating learning and/or sample probes will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3887384