Patent US5850475A successfully refreshed from Google Patents.
US5850475A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method and apparatus for dividing image area
Assignee: Fujifilm Business Innovation Corp
•
Filing: 1997-10-15
•
Granted: 1998-12-15
•
Expires: 2017-10-15
Abstract
Abstract In a method and apparatus for dividing an image area according to the present invention, when data of a quantized multi-gradational image are to be divided into regions according to a texture of the image, the quantized multi-gradational image data are divided into plural rectangular subareas and boundary areas. Each of the rectangular subareas and a boundary area in the periphery of the subarea are combined to form one set. The area dividing is conducted on each of the sets. In the area dividing, the subarea is subjected one time to a dividing process, and the boundary area is subjected two or more times to a dividing process.
Inventors
- 👤 Atsushi Kasao
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 17:12
CPC Classifications
G; G06; G06T; G06T7/00; G06T7/10; G06T7/11; G06T7/143; G06T7/60; G06T7/66; G06V; G06V10/00; G06V10/40; G06T2207/00; G06T2207/20; G06T2207/20021
Claims Analysis
Total Claims: 10
Claim 1
Independent Claim
129 words
A method for segmenting quantized multi-gradational image data into areas corresponding to image texture comprising the steps of: creating in said quantized multi-gradational image data a plurality of rectangular subareas and a plurality of rectangular boundary areas located between adjacent subareas; dividing said image data in each of said rectangular subareas and rectangular boundary areas into subareas corresponding to an image texture; recalculating a closest area center of pixels included in said divided area bounded by an unnatural outline, wherein said closest area center is an area center having the shortest distance between a pixel in a divided area with an inconsistency and area centers stored in said divided area; relocating said pixels included in said divided area bounded by an unnatural outline; and relocating said closest area center.
Claim 2
Independent Claim
129 words
A method for segmenting quantized multi-gradational image data into areas corresponding to image texture comprising the steps of: creating in said quantized multi-gradational image data a plurality of rectangular subareas and a plurality of rectangular boundary areas located between adjacent subareas; dividing said image data in each of said rectangular subareas and rectangular boundary areas into subareas corresponding to an image texture; recalculating a closest area center of pixels included in said divided area bounded by an unnatural outline, wherein said closest area center is an area center having the shortest distance between a pixel in a divided area with an inconsistency and area centers stored in said divided area; relocating said pixels included in said divided area bounded by an unnatural outline; and relocating said closest area center.
Claim 3
Dependent Claim
32 words
The method of claim 1 wherein the step of dividing further includes the substep of: using computed information from a previously divided rectangular boundary area adjacent a rectangular subarea currently being divided.
Claim 4
Dependent Claim
52 words
The method of claim 2, wherein the step of using computed information further includes the substep of: computing information from three pixel color elements, an average value of x and y coordinates specifying a location of data in said quantized multi-gradational image, and a total number of pixels in said divided area.
Claim 5
Dependent Claim
50 words
The method of claim 1, wherein said dividing step further includes the substep of: computing a K-means clustering using a five dimensional space including three dimensions corresponding to the three color elements of data in said quantized multi-gradational image and two dimensions represented by x and y in the image.
Claim 6
Dependent Claim
51 words
The method of claim 4, wherein said step of computing a K-means clustering includes the substep of: computing a distance measure incorporating a coefficient representing a ratio of contribution rates of said two dimensions represented by x and y in the image and the contribution rates of said three color elements.
Claim 7
Dependent Claim
36 words
The method of claim 5, wherein said substep of computing a distance measure includes the substep of: computing a distance between a pixel in one of said rectangular subareas and a center of said divided subarea.
Claim 8
Dependent Claim
119 words
The method of claim 1, wherein said dividing step further includes the substep of: selecting distance measures Ln and L Max according to the following conditions Condition 1 Ln<lx×5, Ln<ly×5, and Ln={(Lx×Ly)/(Ni×π)}1/2, and Condition 2 L.sub.Max <(lx)×6, L.sub.Max <(ly)×6, and L.sub.Max =Max(Lj), where Lx and Ly are lengths of said image in a lateral direction and a vertical direction respectively, Ni indicates the number of centers located in said divided area, lx and ly indicate a horizontal width and a vertical length respectively of said boundary area respectively, and Lj is a distance between a center of said divided area and a pixel which, among pixels in said divided area, is farthest from said center of said divided area.
Claim 9
Dependent Claim
75 words
The method of claim 7, further comprising the steps of: calculating a distance from a first divided area center to other divided area centers; listing divided areas having the shortest calculated distance; calculating a similarity value comparing color and image texture among said first divided area and said listed divided areas; and integrating said first divided area with at least one of said listed divided areas having said similarity value greater than a reference value.
Claim 10
Dependent Claim
75 words
The method of claim 1, further comprising the steps of: calculating a distance from a first divided area center to other divided area centers; listing divided areas having the shortest calculated distance; calculating a similarity value comparing color and image texture among said first divided area and said listed divided areas; and integrating said first divided area with at least one of said listed divided areas having said similarity value greater than a reference value.
References Cited (Backward Citations)
US4831659A
Cited
US4764971A
Cited
US4944023A
Cited
US5073962A
Cited
US5311307A
Cited
US5185808A
Cited
US5319720A
Cited
US5519789A
Cited
US5710840A
Cited
US5604822A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2001-12-14 | 2002-06-15 | 2002-12-15 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2005-12-14 | 2006-06-15 | 2006-12-15 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2009-12-14 | 2010-06-15 | 2010-12-15 | $7700.00 | Window Closed / Lapsed |