Patent US2830919A successfully refreshed from Google Patents.
US6038340A
Expired / Lapsed
US
⏱️ 0.0 years remaining
System and method for detecting the black and white points of a color image
Assignee: Seiko Epson Corp
•
Filing: 1996-11-08
•
Granted: 2000-03-14
•
Expires: 2016-11-08
Abstract
Abstract A system for detecting image black and white points for a digital image. The system includes image partitioning routines for partitioning a digital image into image blocks, a pixel counter for determining whether the image block contains sufficient black and white pixels to classify the block as having relatively black text on a relatively white background, block validity testing routines for determining whether the selected image block is valid by determining whether the number of relatively black pixels for the selected image block is greater than a predetermined black threshold and whether the number of relatively white pixels for the selected image block is greater than a predetermined white threshold, and pixel clustering routines for arranging the relatively black pixels of a valid block into black clusters and the relatively white pixels of a valid block into white clusters and for assigning one of the black cluster centroids based on a first predetermined set of rules as the image black point and one of the white cluster centroids based on a second predetermined set of rules as the image white point.
Inventors
- 👤 Hakan Ancin
- 👤 Anoop K. Bhattacharjya
- 👤 Joseph S. Shu
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-29 07:26
CPC Classifications
H; H04; H04N; H04N1/00; H04N1/46; G; G06; G06V; G06V10/00; G06V10/20; G06V10/28; H04N1/56
Claims Analysis
Total Claims: 25
Claim 1
Independent Claim
354 words
A computer-implemented method for obtaining a black point and a white point for a digital image, comprising the steps of: partitioning the digital image into image blocks; classifying each of the image blocks as valid if a given image block has a total number of relatively black pixels exceeding a first predetermined threshold and a total number of relatively white pixels exceeding a second predetermined threshold, wherein a relatively black pixel is defined as a pixel whose color component values are each less than a predetermined black pixel threshold and a relatively white pixel is defined as a pixel whose color component values are each greater than a predetermined white pixel threshold; clustering the relatively black pixels of the valid image blocks into at least one black cluster and the relatively white pixels of the valid image blocks into at least one white cluster, each black cluster having a black cluster centroid and each white cluster having a white cluster centroid, wherein a current centroid of a first black cluster is set to color component values of a first relatively black pixel, a current centroid of a first white cluster is set to color component values of a first relatively white pixel, and color component values of a selected untested pixel are compared with the color component values of each of the current centroids of the first black and white clusters to determine if the selected pixel belongs to the first black cluster, the first white cluster or a new cluster, wherein as a new pixel is added to an existing cluster new centroid color component values are calculated for the existing cluster based on the existing cluster's current centroid color component values and color component values of the new pixel, there being at least three color component values for each pixel and each centroid; and selecting one of the black cluster centroids according to a first predetermined set of rules to be a black point of the digital image and one of the white cluster centroids according to a second predetermined set of rules to be a white point of the digital image.
Claim 2
Dependent Claim
354 words
A computer-implemented method for obtaining a black point and a white point for a digital image, comprising the steps of: partitioning the digital image into image blocks; classifying each of the image blocks as valid if a given image block has a total number of relatively black pixels exceeding a first predetermined threshold and a total number of relatively white pixels exceeding a second predetermined threshold, wherein a relatively black pixel is defined as a pixel whose color component values are each less than a predetermined black pixel threshold and a relatively white pixel is defined as a pixel whose color component values are each greater than a predetermined white pixel threshold; clustering the relatively black pixels of the valid image blocks into at least one black cluster and the relatively white pixels of the valid image blocks into at least one white cluster, each black cluster having a black cluster centroid and each white cluster having a white cluster centroid, wherein a current centroid of a first black cluster is set to color component values of a first relatively black pixel, a current centroid of a first white cluster is set to color component values of a first relatively white pixel, and color component values of a selected untested pixel are compared with the color component values of each of the current centroids of the first black and white clusters to determine if the selected pixel belongs to the first black cluster, the first white cluster or a new cluster, wherein as a new pixel is added to an existing cluster new centroid color component values are calculated for the existing cluster based on the existing cluster's current centroid color component values and color component values of the new pixel, there being at least three color component values for each pixel and each centroid; and selecting one of the black cluster centroids according to a first predetermined set of rules to be a black point of the digital image and one of the white cluster centroids according to a second predetermined set of rules to be a white point of the digital image.
Claim 3
Dependent Claim
19 words
The method of claim 1, wherein the step of partitioning includes partitioning the digital image into equal-sized image blocks.
Claim 4
Dependent Claim
49 words
The method of claim 1, wherein a relatively black pixel is defined as a pixel whose RGB values are each less than the predetermined black pixel threshold and a relatively white pixel is defined as a pixel whose RGB values are each greater than the predetermined white pixel threshold.
Claim 5
Dependent Claim
16 words
The method of claim 1, wherein the step of clustering includes applying a centroid averaging algorithm.
Claim 6
Dependent Claim
102 words
The method of claim 4, wherein the step of clustering includes establishing new cluster centroid RGB values based on adding a selected pixel by applying the equations C.sub.i.sup.R =[(N*C.sub.i-1.sup.R)+R]/(N+1); C.sub.i.sup.G =[(N*C.sub.i-1.sup.G)+G]/(N+1); C.sub.i.sup.B =[(N*C.sub.i-1.sup.B)+B]/(N+1); where C i R , C i G and C i B denote the new cluster centroid RGB values; C i-1 R , C i-1 G and C i-1 B denote the previous cluster centroid RGB values exclusive of the selected pixel; N denotes the number of pixels in the cluster exclusive of the selected pixel; and R, G and B denote the RGB values for the selected pixel.
Claim 7
Dependent Claim
50 words
The method of claim 1, wherein the step of selecting includes selecting the black cluster centroid of the black cluster having a greatest number of pixels as the black point and selecting the white cluster centroid of the white cluster having a greatest number of pixels as the white point.
Claim 8
Dependent Claim
361 words
A computer-based method of detecting an image black point and an image white point for a digital image, comprising the steps of: partitioning the digital image into image blocks; determining whether each of the image blocks is valid, by determining whether a total number of relatively black pixels in a given image block is greater than a first predetermined threshold and whether a total number of relatively white pixels in the given image block is greater than a second predetermined threshold, wherein a relatively black pixel is defined as a pixel whose color component values are each less than a predetermined black pixel threshold and a relatively white pixel is defined as a pixel whose color component values are each greater than a predetermined white pixel threshold; arranging the relatively black pixels of the valid image blocks into black clusters and the relatively white pixels of the valid image blocks into white clusters, each of the black clusters having a black cluster centroid and each of the white clusters having a white cluster centroid, wherein a current centroid of a first black cluster is set to color component values of a first relatively black pixel, a current centroid of a first white cluster is set to color component values of a first relatively white pixel, and color component values of a selected untested pixel are compared with the color component values of each of the current centroids of the first black and white clusters to determine if the selected pixel belongs to the first black cluster, the first white cluster or a new cluster, wherein as a new pixel is added to an existing cluster new centroid color component values are calculated for the existing cluster based on the existing cluster's current centroid color component values and color component values of the new pixel, there being at least three color component values for each pixel and each centroid; and selecting one of the black cluster centroids based on a first predetermined set of rules as the black point and one of the white cluster centroids based on a second predetermined set of rules as the white point of the digital image.
Claim 9
Dependent Claim
361 words
A computer-based method of detecting an image black point and an image white point for a digital image, comprising the steps of: partitioning the digital image into image blocks; determining whether each of the image blocks is valid, by determining whether a total number of relatively black pixels in a given image block is greater than a first predetermined threshold and whether a total number of relatively white pixels in the given image block is greater than a second predetermined threshold, wherein a relatively black pixel is defined as a pixel whose color component values are each less than a predetermined black pixel threshold and a relatively white pixel is defined as a pixel whose color component values are each greater than a predetermined white pixel threshold; arranging the relatively black pixels of the valid image blocks into black clusters and the relatively white pixels of the valid image blocks into white clusters, each of the black clusters having a black cluster centroid and each of the white clusters having a white cluster centroid, wherein a current centroid of a first black cluster is set to color component values of a first relatively black pixel, a current centroid of a first white cluster is set to color component values of a first relatively white pixel, and color component values of a selected untested pixel are compared with the color component values of each of the current centroids of the first black and white clusters to determine if the selected pixel belongs to the first black cluster, the first white cluster or a new cluster, wherein as a new pixel is added to an existing cluster new centroid color component values are calculated for the existing cluster based on the existing cluster's current centroid color component values and color component values of the new pixel, there being at least three color component values for each pixel and each centroid; and selecting one of the black cluster centroids based on a first predetermined set of rules as the black point and one of the white cluster centroids based on a second predetermined set of rules as the white point of the digital image.
Claim 10
Dependent Claim
19 words
The method of claim 7, wherein the step of partitioning includes partitioning the digital image into equal-sized image blocks.
Claim 11
Dependent Claim
49 words
The method of claim 7, wherein a relatively black pixel is defined as a pixel whose RGB values are each less than the predetermined black pixel threshold and a relatively white pixel is defined as a pixel whose RGB values are each greater than the predetermined white pixel threshold.
Claim 12
Dependent Claim
16 words
The method of claim 7, wherein the step of arranging includes applying a centroid averaging algorithm.
Claim 13
Dependent Claim
102 words
The method of claim 10, wherein the step of arranging includes establishing new cluster centroid RGB values based on adding a selected pixel by applying the equations C.sub.i.sup.R =[(N*C.sub.i-1.sup.R)+R]/(N+1); C.sub.i.sup.G =[(N*C.sub.i-1.sup.G)+G]/(N+1); C.sub.i.sup.B =[(N*C.sub.i-1.sup.B)+B]/(N+1); where C i R , C i G and C i B denote the new cluster centroid RGB values; C i-1 R , C i-1 G and C i-1 B denote the previous cluster centroid RGB values exclusive of the selected pixel; N denotes the number of pixels in the cluster exclusive of the selected pixel; and R, G and B denote the RGB values for the selected pixel.
Claim 14
Dependent Claim
46 words
The method of claim 7, wherein the step of selecting includes selecting the centroid of the black cluster having a greatest number of pixels as the black point and selecting the centroid of the white cluster having a greatest number of pixels as the white point.
Claim 15
Independent Claim
361 words
A system for detecting respective black and white points for a digital image, comprising: means for partitioning the digital image into image blocks; means for determining whether each of the image blocks is valid, by determining whether a total number of relatively black pixels in a given image block is greater than a first predetermined threshold and whether a total number of relatively white pixels in the given image block is greater than a second predetermined threshold, wherein a relatively black pixel is defined as a pixel whose color component values are each less than a predetermined black pixel threshold and a relatively white pixel is defined as a pixel whose color component values are each greater than a predetermined white pixel threshold; and means for arranging the relatively black pixels of the valid image blocks into black clusters and the relatively white pixels of the valid image blocks into white clusters, each of the black clusters having a black cluster centroid and each of the white clusters having a white cluster centroid, wherein a current centroid of a first black cluster is set to color component values of a first relatively black pixel, a current centroid of a first white cluster is set to color component values of a first relatively white pixel, and color component values of a selected untested pixel are compared with the color component values of each of the current centroids of the first black and white clusters to determine if the selected pixel belongs to the first black cluster, the first white cluster or a new cluster, wherein as a new pixel is added to an existing cluster new centroid color component values are calculated for the existing cluster based on the existing cluster's current centroid color component values and color component values of the new pixel, there being at least three color component values for each pixel and each centroid, and for selecting one of the black cluster centroids based on a first predetermined set of rules as the black point and one of the white cluster centroids based on a second predetermined set of rules as the white point of the digital image.
Claim 16
Independent Claim
361 words
A system for detecting respective black and white points for a digital image, comprising: means for partitioning the digital image into image blocks; means for determining whether each of the image blocks is valid, by determining whether a total number of relatively black pixels in a given image block is greater than a first predetermined threshold and whether a total number of relatively white pixels in the given image block is greater than a second predetermined threshold, wherein a relatively black pixel is defined as a pixel whose color component values are each less than a predetermined black pixel threshold and a relatively white pixel is defined as a pixel whose color component values are each greater than a predetermined white pixel threshold; and means for arranging the relatively black pixels of the valid image blocks into black clusters and the relatively white pixels of the valid image blocks into white clusters, each of the black clusters having a black cluster centroid and each of the white clusters having a white cluster centroid, wherein a current centroid of a first black cluster is set to color component values of a first relatively black pixel, a current centroid of a first white cluster is set to color component values of a first relatively white pixel, and color component values of a selected untested pixel are compared with the color component values of each of the current centroids of the first black and white clusters to determine if the selected pixel belongs to the first black cluster, the first white cluster or a new cluster, wherein as a new pixel is added to an existing cluster new centroid color component values are calculated for the existing cluster based on the existing cluster's current centroid color component values and color component values of the new pixel, there being at least three color component values for each pixel and each centroid, and for selecting one of the black cluster centroids based on a first predetermined set of rules as the black point and one of the white cluster centroids based on a second predetermined set of rules as the white point of the digital image.
Claim 17
Dependent Claim
360 words
A computer-readable storage medium storing program code for causing a computer to perform the steps of: partitioning the digital image into image blocks; determining whether each of the image blocks is valid, by determining whether a total number of relatively black pixels in a given image block is greater than a first predetermined threshold and whether a total number of relatively white pixels in the given image block is greater than a second predetermined threshold, wherein a relatively black pixel is defined as a pixel whose color component values are each less than a predetermined black pixel threshold and a relatively white pixel is defined as a pixel whose color component values are each greater than a predetermined white pixel threshold; arranging the relatively black pixels of the valid image blocks into black clusters and the relatively white pixels of the valid image blocks into white clusters for each valid image block, each of the black clusters having a black cluster centroid and each of the white clusters having a white cluster centroid, wherein a current centroid of a first black cluster is set to color component values of a first relatively black pixel, a current centroid of a first white cluster is set to color component values of a first relatively white pixel, and color component values of a selected untested pixel are compared with the color component values of each of the current centroids of the first black and white clusters to determine if the selected pixel belongs to the first black cluster, the first white cluster or a new cluster, wherein as a new pixel is added to an existing cluster new centroid color component values are calculated for the existing cluster based on the existing cluster's current centroid color component values and color component values of the new pixel, there being at least three color component values for each pixel and each centroid; and selecting one of the black cluster centroids based on a first predetermined set of rules as the black point and one of the white cluster centroids based on a second predetermined set of rules as the white point of the digital image.
Claim 18
Dependent Claim
360 words
A computer-readable storage medium storing program code for causing a computer to perform the steps of: partitioning the digital image into image blocks; determining whether each of the image blocks is valid, by determining whether a total number of relatively black pixels in a given image block is greater than a first predetermined threshold and whether a total number of relatively white pixels in the given image block is greater than a second predetermined threshold, wherein a relatively black pixel is defined as a pixel whose color component values are each less than a predetermined black pixel threshold and a relatively white pixel is defined as a pixel whose color component values are each greater than a predetermined white pixel threshold; arranging the relatively black pixels of the valid image blocks into black clusters and the relatively white pixels of the valid image blocks into white clusters for each valid image block, each of the black clusters having a black cluster centroid and each of the white clusters having a white cluster centroid, wherein a current centroid of a first black cluster is set to color component values of a first relatively black pixel, a current centroid of a first white cluster is set to color component values of a first relatively white pixel, and color component values of a selected untested pixel are compared with the color component values of each of the current centroids of the first black and white clusters to determine if the selected pixel belongs to the first black cluster, the first white cluster or a new cluster, wherein as a new pixel is added to an existing cluster new centroid color component values are calculated for the existing cluster based on the existing cluster's current centroid color component values and color component values of the new pixel, there being at least three color component values for each pixel and each centroid; and selecting one of the black cluster centroids based on a first predetermined set of rules as the black point and one of the white cluster centroids based on a second predetermined set of rules as the white point of the digital image.
Claim 19
Independent Claim
459 words
A system for detecting an image black point and an image white point for a digital image, comprising: a memory storing a digital image and image threshold data; an image partitioning element coupled to the memory for partitioning the digital image into image blocks; a pixel counter coupled to the image partitioning element for counting relatively black pixels based on the threshold data and relatively white pixels based on the threshold data in each of the image blocks, and accumulating a total number of relatively black pixels for each of the image blocks and a total number of relatively white pixels for each of the image blocks; a block validity testing element coupled to the pixel counter for determining whether each of the image blocks is valid by determining whether the total number of relatively black pixels in a given image block is greater than a first predetermined threshold and the total number of relatively white pixels in the given image block is greater than a second predetermined threshold, wherein a relatively black pixel is defined as a pixel whose color component values are each less than a predetermined black pixel threshold and a relatively white pixel is defined as a pixel whose color component values are each greater than a predetermined white pixel threshold; and a pixel clustering element coupled to the block validity testing element for arranging the relatively black pixels of the valid blocks into black clusters and the relatively white pixels of the valid blocks into white clusters, each of the black clusters having a black cluster centroid and each of the white clusters having a white cluster centroid, wherein a current centroid of a first black cluster is set to color component values of a first relatively black pixel, a current centroid of a first white cluster is set to color component values of a first relatively white pixel, and color component values of a selected untested pixel are compared with the color component values of each of the current centroids of the first black and white clusters to determine if the selected pixel belongs to the first black cluster, the first white cluster or a new cluster, wherein as a new pixel is added to an existing cluster new centroid color component values are calculated for the existing cluster based on the existing cluster's current centroid color component values and color component values of the new pixel, there being at least three color component values for each pixel and each centroid, and for selecting one of the black cluster centroids based on a first predetermined set of rules as the image black point and one of the white cluster centroids based on a second predetermined set of rules as the image white point.
Claim 20
Independent Claim
459 words
A system for detecting an image black point and an image white point for a digital image, comprising: a memory storing a digital image and image threshold data; an image partitioning element coupled to the memory for partitioning the digital image into image blocks; a pixel counter coupled to the image partitioning element for counting relatively black pixels based on the threshold data and relatively white pixels based on the threshold data in each of the image blocks, and accumulating a total number of relatively black pixels for each of the image blocks and a total number of relatively white pixels for each of the image blocks; a block validity testing element coupled to the pixel counter for determining whether each of the image blocks is valid by determining whether the total number of relatively black pixels in a given image block is greater than a first predetermined threshold and the total number of relatively white pixels in the given image block is greater than a second predetermined threshold, wherein a relatively black pixel is defined as a pixel whose color component values are each less than a predetermined black pixel threshold and a relatively white pixel is defined as a pixel whose color component values are each greater than a predetermined white pixel threshold; and a pixel clustering element coupled to the block validity testing element for arranging the relatively black pixels of the valid blocks into black clusters and the relatively white pixels of the valid blocks into white clusters, each of the black clusters having a black cluster centroid and each of the white clusters having a white cluster centroid, wherein a current centroid of a first black cluster is set to color component values of a first relatively black pixel, a current centroid of a first white cluster is set to color component values of a first relatively white pixel, and color component values of a selected untested pixel are compared with the color component values of each of the current centroids of the first black and white clusters to determine if the selected pixel belongs to the first black cluster, the first white cluster or a new cluster, wherein as a new pixel is added to an existing cluster new centroid color component values are calculated for the existing cluster based on the existing cluster's current centroid color component values and color component values of the new pixel, there being at least three color component values for each pixel and each centroid, and for selecting one of the black cluster centroids based on a first predetermined set of rules as the image black point and one of the white cluster centroids based on a second predetermined set of rules as the image white point.
Claim 21
Dependent Claim
28 words
The system of claim 15, wherein the pixel clustering element uses a centroid averaging algorithm to arrange the black and white pixels into black and white clusters, respectively.
Claim 22
Dependent Claim
424 words
A scanner for scanning an image from a medium, comprising: optical scanner electronics for capturing optical information of the image from the medium and converting the optical information to a digital image; and an image processing unit coupled to the optical scanner electronics and including memory, said image processing unit comprising a black and white point detector having image partitioning routines for partitioning the digital image into image blocks; a pixel counter for counting relatively black pixels and relatively white pixels in each of the image blocks; block validity testing routines for determining whether each image block is valid by determining whether a total number of relatively black pixels in a given image block is greater than a first predetermined threshold and a total number of relatively white pixels in the given image block is greater than a second predetermined threshold, wherein a relatively black pixel is defined as a pixel whose color component values are each less than a predetermined black pixel threshold and a relatively white pixel is defined as a pixel whose color component values are each greater than a predetermined white pixel threshold; and pixel clustering routines for arranging the relatively black pixels of the valid image blocks into black clusters and the relatively white pixels of the valid image blocks into white clusters, each of the black clusters having a black cluster centroid and each of the white clusters having a white cluster centroid, wherein a current centroid of a first black cluster is set to color component values of a first relatively black pixel, a current centroid of a first white cluster is set to color component values of a first relatively white pixel, and color component values of a selected untested pixel are compared with the color component values of each of the current centroids of the first black and white clusters to determine if the selected pixel belongs to the first black cluster, the first white cluster or a new cluster, wherein as a new pixel is added to an existing cluster new centroid color component values are calculated for the existing cluster based on the existing cluster's current centroid color component values and color component values of the new pixel, there being at least three color component values for each pixel and each centroid, and for selecting one of the black cluster centroids based on first predetermined set of rules as the image black point and one of the white cluster centroids based on a second predetermined set of rules as the image white point.
Claim 23
Dependent Claim
424 words
A scanner for scanning an image from a medium, comprising: optical scanner electronics for capturing optical information of the image from the medium and converting the optical information to a digital image; and an image processing unit coupled to the optical scanner electronics and including memory, said image processing unit comprising a black and white point detector having image partitioning routines for partitioning the digital image into image blocks; a pixel counter for counting relatively black pixels and relatively white pixels in each of the image blocks; block validity testing routines for determining whether each image block is valid by determining whether a total number of relatively black pixels in a given image block is greater than a first predetermined threshold and a total number of relatively white pixels in the given image block is greater than a second predetermined threshold, wherein a relatively black pixel is defined as a pixel whose color component values are each less than a predetermined black pixel threshold and a relatively white pixel is defined as a pixel whose color component values are each greater than a predetermined white pixel threshold; and pixel clustering routines for arranging the relatively black pixels of the valid image blocks into black clusters and the relatively white pixels of the valid image blocks into white clusters, each of the black clusters having a black cluster centroid and each of the white clusters having a white cluster centroid, wherein a current centroid of a first black cluster is set to color component values of a first relatively black pixel, a current centroid of a first white cluster is set to color component values of a first relatively white pixel, and color component values of a selected untested pixel are compared with the color component values of each of the current centroids of the first black and white clusters to determine if the selected pixel belongs to the first black cluster, the first white cluster or a new cluster, wherein as a new pixel is added to an existing cluster new centroid color component values are calculated for the existing cluster based on the existing cluster's current centroid color component values and color component values of the new pixel, there being at least three color component values for each pixel and each centroid, and for selecting one of the black cluster centroids based on first predetermined set of rules as the image black point and one of the white cluster centroids based on a second predetermined set of rules as the image white point.
Claim 24
Dependent Claim
396 words
An article of manufacture, comprising a computer usable medium having readable code means embodied therein for detecting an image black point and an image white point for a digital image, the computer readable code program means comprising computer readable program code means for partitioning the digital image into image blocks; computer readable program code means for determining whether each of the image blocks is valid, by determining whether a total number of relatively black pixels in a given image block is greater than a first predetermined threshold and whether a total number of relatively white pixels in the given image block is greater than a second predetermined threshold, wherein a relatively black pixel is defined as a pixel whose color component values are each less than a predetermined black pixel threshold and a relatively white pixel is defined as a pixel whose color component values are each greater than a predetermined white pixel threshold; and computer readable program code means for arranging the relatively black pixels of the valid image blocks into black clusters and the relatively white pixels of the valid image blocks into white clusters, each of the black clusters having a black cluster centroid and each of the white clusters having a white cluster centroid, wherein a current centroid of a first black cluster is set to color component values of a first relatively black pixel, a current centroid of a first white cluster is set to color component values of a first relatively white pixel, and color component values of a selected untested pixel are compared with the color component values of each of the current centroids of the first black and white clusters to determine if the selected pixel belongs to the first black cluster, the first white cluster or a new cluster, wherein as a new pixel is added to an existing cluster new centroid color component values are calculated for the existing cluster based on the existing cluster's current centroid color component values and color component values of the new pixel, there being at least three color component values for each pixel and each centroid, and for selecting one of the black cluster centroids based on a first predetermined set of rules as the black point and one of the white cluster centroids based on a second predetermined set of rules as the white point of the digital image.
Claim 25
Dependent Claim
396 words
An article of manufacture, comprising a computer usable medium having readable code means embodied therein for detecting an image black point and an image white point for a digital image, the computer readable code program means comprising computer readable program code means for partitioning the digital image into image blocks; computer readable program code means for determining whether each of the image blocks is valid, by determining whether a total number of relatively black pixels in a given image block is greater than a first predetermined threshold and whether a total number of relatively white pixels in the given image block is greater than a second predetermined threshold, wherein a relatively black pixel is defined as a pixel whose color component values are each less than a predetermined black pixel threshold and a relatively white pixel is defined as a pixel whose color component values are each greater than a predetermined white pixel threshold; and computer readable program code means for arranging the relatively black pixels of the valid image blocks into black clusters and the relatively white pixels of the valid image blocks into white clusters, each of the black clusters having a black cluster centroid and each of the white clusters having a white cluster centroid, wherein a current centroid of a first black cluster is set to color component values of a first relatively black pixel, a current centroid of a first white cluster is set to color component values of a first relatively white pixel, and color component values of a selected untested pixel are compared with the color component values of each of the current centroids of the first black and white clusters to determine if the selected pixel belongs to the first black cluster, the first white cluster or a new cluster, wherein as a new pixel is added to an existing cluster new centroid color component values are calculated for the existing cluster based on the existing cluster's current centroid color component values and color component values of the new pixel, there being at least three color component values for each pixel and each centroid, and for selecting one of the black cluster centroids based on a first predetermined set of rules as the black point and one of the white cluster centroids based on a second predetermined set of rules as the white point of the digital image.
References Cited (Backward Citations)
US4097847A
Cited
GB1541578A
Cited
US4513442A
Cited
EP0188193A2
Cited
US5068910A
Cited
EP0371389A2
Cited
US5420938A
Cited
US5201014A
Cited
US5289296A
Cited
US5617485A
Cited
US5287204A
Cited
US5280367A
Cited
US5179599A
Cited
US5400153A
Cited
US5282061A
Cited
US5375175A
Cited
US5703959A
Cited
US5596655A
Cited
US5550931A
Cited
US5566245A
Cited
US5371615A
Cited
US5581370A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2003-03-14 | 2003-09-13 | 2004-03-14 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2007-03-14 | 2007-09-13 | 2008-03-14 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2011-03-14 | 2011-09-13 | 2012-03-14 | $7700.00 | Window Closed / Lapsed |