Method for comparing image sections to determine similarity ther

Image analysis – Pattern recognition – Template matching

Patent

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

382257, G06K 968, G06K 940

Patent

active

055398413

ABSTRACT:
A method for comparing two image sections consisting of a plurality of image signals, or pixels, where each image section represents a token (e.g., character, symbol, glyph, string of components, or similar units of semantic understanding), in order to identify when similar tokens are present within the image sections. The invention further operates without the need for individually detecting and/or identifying the components making up the tokens. In one embodiment, the method relies upon the detection of connected components within words to first isolate individual word tokens and then applies a two stage process where dilated images of the tokens are compared with model representations of the tokens to determine the relative similarity therebetween.

REFERENCES:
patent: 2905927 (1959-09-01), Reed
patent: 3133266 (1964-05-01), Frishkopf
patent: 3295105 (1966-12-01), Gray et al.
patent: 4010445 (1977-03-01), Hoshino
patent: 4155072 (1979-05-01), Kawa
patent: 4326190 (1982-04-01), Borland et al.
patent: 4400828 (1983-08-01), Pirz et al.
patent: 4477926 (1984-10-01), Linger et al.
patent: 4481664 (1984-11-01), Linger et al.
patent: 4495644 (1985-01-01), Parks et al.
patent: 4558461 (1985-12-01), Schlang
patent: 4644585 (1987-02-01), Crimmins et al.
patent: 4701960 (1987-10-01), Scott
patent: 4731857 (1988-03-01), Tappert
patent: 4747148 (1988-05-01), Watanabe et al.
patent: 4764972 (1988-08-01), Yoshida et al.
patent: 4809344 (1989-02-01), Peppers et al.
patent: 4821333 (1989-04-01), Gillies
patent: 4864628 (1989-09-01), Scott
patent: 4918740 (1990-04-01), Ross
patent: 4926490 (1990-05-01), Mano
patent: 4933977 (1990-06-01), Ohnishi et al.
patent: 4949281 (1990-08-01), Hillenbrand et al.
patent: 4949392 (1990-08-01), Barski et al.
patent: 4956869 (1990-09-01), Miyatake et al.
patent: 4977603 (1990-12-01), Irie et al.
patent: 5142589 (1992-08-01), Lougheed et al.
patent: 5179419 (1993-01-01), Palmquist et al.
patent: 5179596 (1993-01-01), Weingard
patent: 5214719 (1993-05-01), Budd et al.
patent: 5216725 (1993-06-01), McCubbrey
patent: 5272764 (1993-12-01), Bloomberg et al.
patent: 5291560 (1994-03-01), Daugman
patent: 5305389 (1994-04-01), Palmer
"F6365 Japanese Document Reader" Fujitsu Sci. Tech. J., 26, 3, pp. 224-233 (Oct. 1990).
"An Efficiently Computable Metric for Comparing Polygon Shapes," by Arkin; Chew, Huttenlocher, Kedem & Mitchell; Proceedings of First Annual ACM-SIAM Symposium on Discrete Algorithms, Jan. 1990 (pp. 129-137).
"Comparing Images Using the Hausdorff Distance" by Hutfenlocher, Klanderman and Rucklidge; Dept. of Computer Science, Cornell University, Ithaca, NY 14853; CUCS TR 91-1211 (revised) Jun. 1991.
"A Multi-Resolution Technique for Comparing Images Using the Hausdorf Distance" by Huttenlocher and Rucklidge; TR 92-1321 Dec. 1992; Dept. of Computer Science, Cornell University, Ithaca, NY.
Conference Paper; K. Yokosawa; "Human-Based Character String Image Retrieval from Textual Images", 1989.
Journal Paper; R. M. Bozinovic, S. N. Srihari; "Off-Line Cursive Script Word Recognition"; 1989.

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Method for comparing image sections to determine similarity ther 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 comparing image sections to determine similarity ther, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for comparing image sections to determine similarity ther will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-720359

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.