Image analysis – Pattern recognition – Classification
Patent
1996-10-16
1998-05-19
Couso, Yon
Image analysis
Pattern recognition
Classification
382160, G06K 962, G06K 974
Patent
active
057546959
ABSTRACT:
The present invention provides a method for recognizing connected and degraded text embedded in a gray-scale image. In accordance with the invention, pseudo two-dimensional hidden Markov models (PHMMs) are used to represent characters. Observation vectors for the gray-scale image are produced from pixel maps obtained by gray-scale optical scanning. Three components are employed to characterize a pixel: a convoluted, quantized gray-level component, a pixel relative position component, and a pixel major stroke direction component. These components are organized as an observation vector, which is continuous in nature, invariant in different font sizes, and flexible for use in various quantization processes. In this matter, information loss or distortion due to binarization processes is eliminated; moreover, in cases where documents are binary in nature (e.g., faxed documents), the bi-level image may be compressed by subsampling into multi(gray)-level without losing information, thereby enabling recognition of the compressed images in a much shorter time. Furthermore, documents in gray-level may be scanned and processed with much lower resolution than in binary without sacrificing the performance. This can also significantly increase the processing speed.
REFERENCES:
patent: 4177448 (1979-12-01), Brayton
patent: 4905287 (1990-02-01), Segawa
patent: 5075896 (1991-12-01), Wilcox et al.
patent: 5261009 (1993-11-01), Bokser
patent: 5289562 (1994-02-01), Mizuta et al.
patent: 5321773 (1994-06-01), Kopec et al.
patent: 5438630 (1995-08-01), Chen et al.
patent: 5459809 (1995-10-01), Kim et al.
patent: 5559897 (1996-09-01), Brown et al.
Tullken, Application of the Grey-Box Approach to Parameter Estimation in Physicochemical Models (IEEE, 1991, Proceedings of the 3oth Conference on Decision Control, Brighton, England, Dec. 1991).
"A Tree-Trellis Based Fast Search for Finding the N-Best Sentence Hypothesis in Continuous Speech Recognition" by F. K. Soong et al., Proc. DARPA Speech and Natural Language Workshop, pp. 12-19.
"Automatic Recognition of Keywords in Unconstrained Speech UsingHidden Markov Models", J. Wilpon et al., IEEE Trans. Acoust. Speech Signal Processing, vol. 38, pp. 1870-1878, Nov. 1990.
"Connected and Degraded Text Recognition Using Hidden Markov Model" by C. Bose et al., Proc. of the 11th Int. Conf. on Pattern Recognition, 1992.
"Global-to-Local Layout Analysis" by H. Baird, Proceedings of the IAPR Workshop on Syntactic and Structural pattern Recognition, France, Sep. 1988.
"Document Image Analysis" by S. Srihari et al., Proceedings of the 8th International Conference on Pattern Recognition, Paris, Oct. 1986.
"A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition" by L. Rabiner, Proceedings of the IEEE, vol. 77, pp. 257-286, Feb. 1989.
"A Segmental k-Means Training Procedure for Connected Word Recognition Based on Whole Word Reference Patterns", L. Rabiner, et al., AT&T Technical Journal, vol. 65, pp. 21-36, May 1986.
"A tree-trellis based fast search for finding the n best sentence hypothesis in continuous speech recognition" F. K. Soong et al., Proc. DARPA Speech and Natural Language Workshop, pp. 12-19, Jun. 1990. Comment: At the present time, a copy of this reference is unavailable to us. As soon as we are in receipt of it we will make the same available to the United States Patent & Trademark Office.
"Pseudo two-dimensional hidden markov models for document recognition" by O. E. Agazzi et al., AT&T Technical Journal, vol. 72, pp. 60-72, Sep./Oct. 1993.
"Fundamentals of Speech Recognition", L. Rabiner and B. Juang, PTR, Prentice Hall, 1993. Comment: Because of the voluminous nature of this book, applicants have not included it with this information disclosure statement but have elected to cite the reference. If the Examiner requires the book during prosecution of this application, Applicants will provide a copy of it.
"Hidden Markov Model Based Optical Character Recognition in the Presence of Deterministic Transformations", by O. E. Agazzi et al., Pattern Recognition, vol. 26, No. 12, 1993.
"Keyword Spotting in Poorly Printed Documents Using Pseudo 2d Hidden Markov Models", by S. Kuo et al., IEEE Trans. on PAMI, vol. 16, pp. 842-848, Aug. 1994.
"Connected and Degraded Text Recognition Using Planar Hidden Markov Models", by O. E. Agazzi et al., in Proc. of ICASSP'93, pp. V113-V116, 1993.
"Direct gray-scale extraction of features for character recognition," by L. Wang et al., IEEE Trans. on PAMI, vol. 15, Oct. 1993.
"A frame-synchronous Network Search Algorithm for Connected Word Recognition" by C. Lee et al., IEEE trans. Acouts. Speech Signal Processing, vol. ASSP-37, pp. 1649-1658, Nov. 1989.
"Recognition of Isolated Digits Using Hidden markov Models with Continuous Mixture Densities" by L. R. Rabiner et al., AT&T Tech. J. vol. 64, pp. 1211-1222, Jul./Aug. 1985.
"Minimum Error Thresholding" by J. Kittler et al., Pattern Recognition, vol. 19, pp. 41-47, 1986.
Agazzi Oscar Ernesto
Kuo Shyh-Shiaw
Yen Chinching
Bartholomew Steven R.
Couso Yon
Lucent Technologies - Inc.
Patel Jayanti K.
LandOfFree
Degraded gray-scale document recognition using pseudo two-dimens does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Degraded gray-scale document recognition using pseudo two-dimens, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Degraded gray-scale document recognition using pseudo two-dimens will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1861757