Patent US5930393A successfully refreshed from Google Patents.
US5930393A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method and apparatus for enhancing degraded document images
Assignee: Alcatel Lucent SAS
•
Filing: 1997-08-11
•
Granted: 1999-07-27
•
Expires: 2017-08-11
Abstract
Abstract A technique for the enhancement of degraded document images to improve their display quality characteristics and image recognition accuracy. Images believed to be representative of the same symbol which occur in different positions over a image source (e.g., a facsimile page) are clustered together. Using the symbols within a particular cluster, an average character image outline for that cluster of symbols is derived and thereafter used to refine the matching of symbols within the cluster and to determine a final representative symbol for that cluster. The final representative symbols from the various resulting clusters are then used to replace all matching images throughout the image source. Advantageously, the display quality and recognition accuracy of the image source is enhanced after application of the present invention due to the resulting improvement of the images in the image source.
Inventors
- 👤 Tin Kam Ho
- 👤 John D. Hobby
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-08 14:49
CPC Classifications
G; G06; G06V; G06V30/00; G06V30/10; G06V30/24; G06V30/242
Claims Analysis
Total Claims: 28
Claim 1
Independent Claim
120 words
A method for enhancing the image quality of a degraded image source having a plurlity of images, the method comprising the steps of: clustering particular images, the particular images being representative of a same symbol occurring in different positions across the degraded image source, from the plurality of images of the degraded image source into a cluster, the clustering being performed as a function of sorting the plurality of images using an image measure to form an initial cluster, the image measure describing particular quantitative characteristics of a single image type; generating a representative symbol for the cluster from the particular images within the cluster; and replacing at least one image from the plurality of images with the representative symbol.
Claim 2
Independent Claim
120 words
A method for enhancing the image quality of a degraded image source having a plurlity of images, the method comprising the steps of: clustering particular images, the particular images being representative of a same symbol occurring in different positions across the degraded image source, from the plurality of images of the degraded image source into a cluster, the clustering being performed as a function of sorting the plurality of images using an image measure to form an initial cluster, the image measure describing particular quantitative characteristics of a single image type; generating a representative symbol for the cluster from the particular images within the cluster; and replacing at least one image from the plurality of images with the representative symbol.
Claim 3
Dependent Claim
12 words
The method of claim 1 wherein the clustering step includes character segmentation.
Claim 4
Dependent Claim
79 words
The method of claim 1 wherein the clustering step includes the steps of: identifying a plurality of equivalence classes in the initial cluster, at least one equivalence class being identified as a function of a image similarity relation between at least two images of the plurality of images, and using the identified equivalence classes to form a secondary cluster from the initial cluster; and using a plurality of feature vectors derived from the secondary cluster to form the cluster.
Claim 5
Dependent Claim
29 words
The method of claim 3 wherein the initial cluster includes more images than the secondary cluster and the cluster, and the secondary cluster includes more images than the cluster.
Claim 6
Dependent Claim
15 words
The method of claim 2 wherein the representative symbol is an average character image outline.
Claim 7
Dependent Claim
16 words
The method of claim 5 wherein the average character image outline is formed using bitmap averaging.
Claim 8
Independent Claim
172 words
A method for enhancing the image quality of a degraded image source having a plurality of images, the method comprising the steps of: generating a plurality of clusters from the plurality of images, each cluster comprising particular ones of the images which are selected on the basis of being representative of a same symbol occurring in different positions across the degraded image source as determined as a function of sorting the plurality of images using a series of image measures to form an initial cluster, each image measure of the series describing particular quantitative characteristics of a single image type; generating a plurality of representative symbols from the plurality of clusters, each one of the representative symbols corresponding to a single one of the clusters; comparing at least one of the representative symbols to at least one other representative symbol, and if a match occurs, merging the clusters whose corresponding representative symbols match; and replacing at least one image of the plurality of images with at least one of the representative symbols.
Claim 9
Independent Claim
172 words
A method for enhancing the image quality of a degraded image source having a plurality of images, the method comprising the steps of: generating a plurality of clusters from the plurality of images, each cluster comprising particular ones of the images which are selected on the basis of being representative of a same symbol occurring in different positions across the degraded image source as determined as a function of sorting the plurality of images using a series of image measures to form an initial cluster, each image measure of the series describing particular quantitative characteristics of a single image type; generating a plurality of representative symbols from the plurality of clusters, each one of the representative symbols corresponding to a single one of the clusters; comparing at least one of the representative symbols to at least one other representative symbol, and if a match occurs, merging the clusters whose corresponding representative symbols match; and replacing at least one image of the plurality of images with at least one of the representative symbols.
Claim 10
Dependent Claim
94 words
The method of claim 7 wherein in the generating a plurality of clusters step each cluster of the plurality of clusters is generated by the further steps of: identifying a plurality of equivalence classes in the initial cluster at least one equivalence class being identified as a function of a image similarity relation between at least two images of the plurality of images, and using the identified equivalence classes to form a secondary cluster from the initial cluster; and using a plurality of feature vectors derived from the secondary cluster to form the cluster.
Claim 11
Dependent Claim
31 words
The method of claim 7 further comprising the steps of: identifying singleton clusters from the plurality of clusters; and merging the singleton clusters with non-singleton clusters from the plurality of clusters.
Claim 12
Dependent Claim
42 words
The method of claim 8 including the further step of: comparing the particular ones of the images within the cluster with the representative symbol corresponding to the cluster and, if a mismatch occurs, removing the mismatched particular one image from the cluster.
Claim 13
Dependent Claim
19 words
The method of claim 7 wherein the representative symbol is an average character image outline derived through bitmap averaging.
Claim 14
Dependent Claim
17 words
The method of claim 8 wherein the plurality of equivalence classes are determined using a comparison function.
Claim 15
Dependent Claim
16 words
The method of claim 1 or 7 wherein the degraded image source is a printed document.
Claim 16
Dependent Claim
29 words
The method of claim 8 wherein at least one of the feature vectors is selected from the group consisting of: histogram, contour, pixel correlation, subsamples and stroke detection distribution.
Claim 17
Dependent Claim
161 words
An optical character recognition apparatus for enhancing the image quality of a degraded image source the optical character recognition apparatus comprising: a selector for selecting a first plurality of images from the degraded image source; a image symbol generator for clustering particular images, the particular images being representative of a same symbol occurring in different positions across the degraded image source, from the first plurality of images into a cluster, the clustering of the particular images being preformed as a function of sorting the first plurality of images using an image measure to form an initial cluster, the image measure describing particular quantitative characteristics of a single image type, and generating a representative symbol for the cluster from the particular images within the cluster; a image replacer for replacing at least one image from the first plurality of images with the representative symbol thereby creating a second plurality of images; and an image classifier for classifying the second plurality of images.
Claim 18
Dependent Claim
161 words
An optical character recognition apparatus for enhancing the image quality of a degraded image source the optical character recognition apparatus comprising: a selector for selecting a first plurality of images from the degraded image source; a image symbol generator for clustering particular images, the particular images being representative of a same symbol occurring in different positions across the degraded image source, from the first plurality of images into a cluster, the clustering of the particular images being preformed as a function of sorting the first plurality of images using an image measure to form an initial cluster, the image measure describing particular quantitative characteristics of a single image type, and generating a representative symbol for the cluster from the particular images within the cluster; a image replacer for replacing at least one image from the first plurality of images with the representative symbol thereby creating a second plurality of images; and an image classifier for classifying the second plurality of images.
Claim 19
Dependent Claim
77 words
The apparatus of claim 15 wherein the image symbol generator identifies a plurality of equivalence classes in the initial cluster, at least one equivalence class being identified as a function of a image similarity relation between at least two images of the first plurality of images, and uses the identified equivalence classes to form a secondary cluster from the initial cluster, and uses a plurality of feature vectors derived from the secondary cluster to form the cluster.
Claim 20
Dependent Claim
15 words
The apparatus of claim 15 wherein the representative symbol is an average character image outline.
Claim 21
Dependent Claim
16 words
The apparatus of claim 17 wherein the average character image outline is derived through bitmap averaging.
Claim 22
Dependent Claim
14 words
The apparatus of claim 15 wherein the degraded image source is a printed document.
Claim 23
Dependent Claim
22 words
The apparatus of claim 15 wherein the second plurality of images has a higher image resolution than the first plurality of images.
Claim 24
Independent Claim
139 words
An apparatus for enhancing the image quality of a degraded image source having a plurality of images, the apparatus comprising: a selector for selecting the plurality of images from the degraded image source; a image symbol generator for clustering particular images, the particular images being representative of a same symbol occurring in different positions across the degraded image source, from the plurality of images into a cluster, the clustering of the particular images being performed as a function sorting the plurality of images using a series image measures, each image measure of the series describing particular quantitative characteristics of a single image type, and generating a representative symbol for the cluster from the particular images within the cluster; and a image replacer for replacing at least one of the images from the plurality of images with the representative symbol.
Claim 25
Independent Claim
139 words
An apparatus for enhancing the image quality of a degraded image source having a plurality of images, the apparatus comprising: a selector for selecting the plurality of images from the degraded image source; a image symbol generator for clustering particular images, the particular images being representative of a same symbol occurring in different positions across the degraded image source, from the plurality of images into a cluster, the clustering of the particular images being performed as a function sorting the plurality of images using a series image measures, each image measure of the series describing particular quantitative characteristics of a single image type, and generating a representative symbol for the cluster from the particular images within the cluster; and a image replacer for replacing at least one of the images from the plurality of images with the representative symbol.
Claim 26
Dependent Claim
39 words
The apparatus of claim 21 wherein, prior to the replacing, the image replacer performs a comparison between a centroid of the representative symbol and a centroid of the at least one of the images from the plurality of images.
Claim 27
Dependent Claim
19 words
The apparatus of claim 21 wherein the representative symbol is an average character image outline formed by bitmap averaging.
Claim 28
Dependent Claim
14 words
The apparatus of claim 23 wherein the degraded image source is a printed document.
References Cited (Backward Citations)
US5657397A
Cited
US5315668A
Cited
US5524065A
Cited
US5822465A
Cited
US5550931A
Cited
US5825925A
Cited
US5577135A
Cited
US5757964A
Cited
US5818952A
Cited
US5764799A
Cited
US5818965A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2002-07-26 | 2003-01-25 | 2003-07-27 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2006-07-26 | 2007-01-25 | 2007-07-27 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2010-07-26 | 2011-01-25 | 2011-07-27 | $7700.00 | Window Closed / Lapsed |