Image analysis – Pattern recognition – Template matching
Reexamination Certificate
1999-09-08
2002-06-18
Au, Amelia M. (Department: 2623)
Image analysis
Pattern recognition
Template matching
C382S228000, C707S793000, C707S793000
Reexamination Certificate
active
06408093
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally a method for comparing a scheme for ranking the similarity of a set of objects to a standard object, to a standard ranking scheme. One specific application is taught whereby a document image ranking scheme can be compared with a given standard ranking scheme.
BACKGROUND OF THE INVENTION
There are many applications where document images are processed to determine their similarity with other document images, or models of document image types. These include classifying groups of similar document images. Document type classification is used in a variety of applications including database management and document routing through a computer network. Furthermore, by identifying the class to which a document image belongs, one can expect certain information contained in the document to appear in selected regions of the document. Thus, once a document type is identified, data may be relatively easily extracted from the document by focusing on that specific region.
Due to the demand, various schemes have been developed and are known, for ranking document images by their degree of closeness to a set standard document image. Furthermore, over time and experience with a particular scheme, those skilled in the art become familiar with what they can expect from that scheme in terms of reliability and performance. However, there is no objective method for measuring the comparative performance of one ranking scheme to another for a particular application.
Similarly, ranking schemes are used in a variety of applications, not only document images. Indeed a ranking scheme can be used whenever it is desired to order the relative similarity of multiple objects to a standard object. An object herein refers to any item that can be compared with a given standard, including, but not limited to a document, text file, or image file. However, there is no known objective method for comparing the various schemes for ranking the relative similarity of the various objects to the standard.
SUMMARY OF THE INVENTION
In accordance with the present invention an object ranking scheme is compared with a known standard, or ideal ranking scheme for a comparative assessment of the performance and reliability of the ranking scheme being tested as compared with the ideal ranking scheme. The same set of objects are processed and ranked by both ranking schemes. A higher ranking indicates a relatively closer similarity to a standard object than a lower ranked object.
The ranking scheme being tested is examined for all objects of interest. Referring to each of those objects as subject objects, the test ranking scheme is charged for each object, referred to herein as a swapped object, which it ranked higher than a subject object, which the ideal ranking scheme ranked lower than the subject object. In one specific embodiment of the present invention the objects are document images.
REFERENCES:
patent: 5991755 (1999-11-01), Noguchi et al.
patent: 6012056 (2000-01-01), Menlove
patent: 6104835 (2000-08-01), Han
patent: 2001/0013035 (2001-08-01), Cohen
Tim Kam Ho, “Decision Combination in Mutiple Classifier Systems”, Jan. 1994, IEEE ISBN: 0162-8828, vol. 16, No. 1, pp. 66-75.*
Kim et al, “A Systematic Approach to Classifier Selection on Combining Multiple Classifiers for Handwritten Digit Recognition”, 1997, IEEE ISBN: 0-8186-7898-4, pp. 459-462.*
Tubbs et al, “Measures of Confidence Associated with Combining Classification results”, IEEE ISBN: 0018-9472, May/Jun. 1991, vol. 21, No. 3, pp. 690-692.*
Henry S. Baird, “Background Structure In Document Images”, International Journal of Pattern Recognition and Artificial Intelligenc vol. 8, No. 5 (1994) 1013-1030.
Wei Zhu, “Image Organization and Retrieval using a Flexible Shape Model”, pp. 31-39.
Andreas Dengel and Gerhard Barth, “High Level Document Analysis Guided by Geometric Aspects”, International Journal of Pattern Recognition and Artificial Intelligenece vol. 2 No. 4 (1988) 641-655 (1988).
Hanno Walischewski, “Automatic Knowledge Acquisition for Spatial Document Interpretation”, pp. 243-247 1997 IEEE.
David Doermann, Huiping Li and Omid Kia, “The Detection of Duplicates in Document Image Databases”, pp. 314-318 1997 IEEE.
Jonathan J. Hull and John F. Cullen, “Document Image Similarity and Equivalence Detection”, pp. 308-312 1977 IEEE.
R.S. Kashi, J. Hu, W.L. Nelson, “On-line Handwritten Signature Verification using Hidden Markov Model Features”.
Douglas E. Critchlow, “Metric Methods for Analyzing Partially Ranked Data,” in 34 Lecture Notes in Statistics (D. Brillinger, et al eds. 1985).
Jianying Hu, Michael K. Brown, and William Turin, “HMM Based On-Line Handwriting Recognition”, IEEE Transactions on Patern Analysis and Machine Intelligence, vol. 18, No. 10, Oct. 1996.
John D. Ferguson, “Variable Duration Models for Speech” pp. 143-147.
Hiroaki Sakoe and Seibi Chiba, “Dynamic Programming Algorithm Optimization for Spoken Word Recognition”, IEEE Transactions on Acoustics, Speech and Signal Processing vol. ASSP-26, No. 1 pp. 43-49 (1978).
S.E. Levinson, “Continuously variable duration hidden Markov models for automatic speech recognition”, Computer Speech and Language pp. 29-45 (1986).
Hu Jianying
Kashi Ramanujan S.
Wilfong Gordon Thomas
Au Amelia M.
Dastouri Mehrdad
Lucent Technologies - Inc.
LandOfFree
Method for comparing object ranking schemes 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 object ranking schemes, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for comparing object ranking schemes will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2910675