Patent US6963663B1 successfully refreshed from Google Patents.
← Back to Patent Explorer
US6963663B1 Expired / Lapsed US ⏱️ 0.0 years remaining

Image processing for image correction

Assignee: Minolta Co Ltd
•
Filing: 2000-06-28
•
Granted: 2005-11-08
•
Expires: 2020-06-28
📄 Original PDF 🌐 Google Patents 🔄 Sync Now

Abstract

Abstract Characteristics of a plurality of skin types have been classified, and skin areas are detected from image data according to the classification. The skin areas are extracted in two steps, first coarsely and next finely. Then, skin areas can be extracted at high speed. For each of the skin areas image data are corrected according to the characteristics of the skin area. Then, skin areas are corrected according to the skin type. Even when an image includes a plurality of persons having different skin characteristics, correction appropriate for each skin type is possible.

Inventors

  • 👤 Hiroki Yoshida

Bibliographic Details

Application Number —
Jurisdiction US
Data Source google_patents
Last Synchronized 2026-11-09 16:34
CPC Classifications
H; H04; H04N; H04N1/00; H04N1/46; H04N1/56; H04N1/60; H04N1/62; H04N1/628; G; G06; G06T; G06T7/00; G06T7/90; G06V; G06V10/00; G06V10/40; G06V10/56; G06T2207/00; G06T2207/10; G06T2207/10024; G06T2207/20; G06T2207/20021; G06T2207/30; G06T2207/30196; G06T2207/30201

Claims Analysis

Total Claims: 26
Claim 1
Independent Claim 98 words
An image processor, wherein characteristics of a plurality of skin colors are classified beforehand, comprising: an area extractor which extracts skin areas from input image data a decider which decides, among a plurality of different classifications, in which classification each of the extracted skin areas belong; and an image corrector which corrects image data of each of the skin areas extracted by said area extractor according to a correction method provided beforehand that corrects each one of the plurality of different classified skin areas differently according to which of the plurality of different classifications the skin area belongs.
Claim 2
Dependent Claim 98 words
An image processor, wherein characteristics of a plurality of skin colors are classified beforehand, comprising: an area extractor which extracts skin areas from input image data a decider which decides, among a plurality of different classifications, in which classification each of the extracted skin areas belong; and an image corrector which corrects image data of each of the skin areas extracted by said area extractor according to a correction method provided beforehand that corrects each one of the plurality of different classified skin areas differently according to which of the plurality of different classifications the skin area belongs.
Claim 3
Dependent Claim 38 words
The image processor according to claim 1 , wherein said area extractor determines the characteristic of skin according to average, dispersion and center of gravity obtained from a histogram of image data of each of the skin areas.
Claim 4
Dependent Claim 95 words
The image processor according to claim 1 , wherein said image corrector prepares a histogram of hue for each of the skin areas extracted by said image extractor, modifies the histogram to shift the center of gravity thereof to come close to center of gravity of a memory color of the skin color according to the characteristic of skin color of the each of the skin areas, generates a correction curve for correcting the image data according to the modified histogram, and corrects the image data in the skin area according to the correction curve.
Claim 5
Dependent Claim 49 words
The image processor according to claim 3 , wherein said image corrector prepares a histogram of chroma for each the skin areas extracted by said image extractor, modifies the histogram according to the characteristic of skin, and generates a correction curve for the image data in the skin area.
Claim 6
Dependent Claim 64 words
The image processor according to claim 4 , wherein when the skin areas are extracted from the input image data, said area extractor extracts skin candidate areas from image data, converts color specification system of the image data for each of the skin candidate areas and extracts skin areas from the skin candidate areas based on the image data in the color specification system.
Claim 7
Dependent Claim 66 words
The image processor according to claim 5 , wherein said area extractor comprises a divider which divides the image data into rectangle areas, and an area integrator which integrates the rectangle areas for each of the characteristic when the skin areas are extracted, wherein said area integrator integrates adjacent pixels if difference in hue between adjacent pixels is within a threshold determined for each skin type.
Claim 8
Dependent Claim 46 words
The image processor according to claim 1 , wherein said image corrector prepares a histogram of chroma for each of skin areas extracted by said image extractor, modifies the histogram according to the characteristic of skin, and generates a correction curve for correcting the image data.
Claim 9
Dependent Claim 64 words
The image processor according to claim 1 , wherein when the skin areas are extracted from the input image data, said area extractor extracts skin candidate areas from image data, converts color specification system of the image data for each of the skin candidate areas and extracts skin areas from the skin candidate areas based on the image data in the color specification system.
Claim 10
Dependent Claim 66 words
The image processor according to claim 1 , wherein said area extractor comprises a divider which divides the image data into rectangle areas, and an area integrator which integrates the rectangle areas for each of the characteristic when the skin areas are extracted, wherein said area integrator integrates adjacent pixels if difference in hue between adjacent pixels is within a threshold determined for each skin type.
Claim 11
Dependent Claim 92 words
An image processing method, wherein characteristics of a plurality of skin colors are classified beforehand, the method comprising the steps of: extracting skin areas from input image data; deciding which, among a plurality of different classifications, each of the extracted skin areas belongs; and correcting image data of each of the extracted skin areas according to one of a plurality of different correction methods provided beforehand that corrects each one of the plurality of different classified skin areas differently according to which of the plurality of different classifications the skin area belongs.
Claim 12
Dependent Claim 92 words
An image processing method, wherein characteristics of a plurality of skin colors are classified beforehand, the method comprising the steps of: extracting skin areas from input image data; deciding which, among a plurality of different classifications, each of the extracted skin areas belongs; and correcting image data of each of the extracted skin areas according to one of a plurality of different correction methods provided beforehand that corrects each one of the plurality of different classified skin areas differently according to which of the plurality of different classifications the skin area belongs.
Claim 13
Dependent Claim 97 words
The image processing method according to claim 10 , wherein said image correction step comprises the steps of: preparing a histogram of hue of each of the extracted skin areas; changing the histogram to shift the center of gravity thereof to come close to center of gravity of a memory color of the skin color according to the characteristic of skin color of the each of the skin areas; generating a correction curve for correcting the image data according to the modified histogram; and correcting the image data in the skin area according to the correction curve.
Claim 14
Dependent Claim 96 words
A computer-readable storage medium storing a computer program comprising the steps of: providing a plurality of classifications of characteristics, each classification corresponding to a plurality of skin colors; extracting skin areas from input image data; deciding which classification, among the plurality of different classifications, each of the extracted skin areas belong; and correcting image data of each of the extracted skin areas according to a correction method provided beforehand that corrects each one of the plurality of different classified skin areas differently according to which of the plurality of different classifications the skin area belongs to.
Claim 15
Dependent Claim 96 words
A computer-readable storage medium storing a computer program comprising the steps of: providing a plurality of classifications of characteristics, each classification corresponding to a plurality of skin colors; extracting skin areas from input image data; deciding which classification, among the plurality of different classifications, each of the extracted skin areas belong; and correcting image data of each of the extracted skin areas according to a correction method provided beforehand that corrects each one of the plurality of different classified skin areas differently according to which of the plurality of different classifications the skin area belongs to.
Claim 16
Dependent Claim 39 words
The computer-readable storage medium according to claim 12 , wherein in said extracting step, the characteristic of skin is determined according to average, dispersion and center of gravity obtained from a histogram of image data of the skin area.
Claim 17
Dependent Claim 92 words
The computer-readable storage medium according to claim 12 , wherein in said correcting step, a histogram of hue of the extracted skin area is prepared, the histogram is modified to shift the center of gravity thereof to come close to center of gravity of a memory color of the skin color according to the characteristic of skin color of each of the skin areas, a correction curve for image data is generated according to the modified histogram, and the image data in the skin area is corrected according to the correction curve.
Claim 18
Dependent Claim 47 words
The computer-readable storage medium according to claim 14 , wherein in said correcting step, a histogram of chroma of each of the extracted skin areas is prepared, the histogram is extended according to the characteristic of skin, and a correction curve for the image data is generated.
Claim 19
Dependent Claim 66 words
The computer-readable storage medium according to claim 15 , wherein when the skin areas are extracted from the input image data, skin candidate areas are extracted from image data, color specification system of the image data is converted for each of the skin candidate areas and skin areas are extracted from the extracted skin candidate areas based on the image data in the color specification system.
Claim 20
Dependent Claim 60 words
The computer-readable storage medium according to claim 16 , wherein in said extracting step, the image data are divided into rectangle areas, and the rectangle areas are integrated for each of the characteristic when the skin areas are extracted, wherein adjacent pixels are integrated if difference in hue between adjacent pixels is within a threshold determined for each skin type.
Claim 21
Dependent Claim 47 words
The computer-readable storage medium according to claim 12 , wherein in said correcting step, a histogram of chroma of each of the extracted skin areas is prepared, the histogram is modified according to the characteristic of skin, and a correction curve for the image data is generated.
Claim 22
Dependent Claim 66 words
The computer-readable storage medium according to claim 12 , wherein when the skin areas are extracted from the input image data, skin candidate areas are extracted from image data, color specification system of the image data is converted for each of the skin candidate areas and skin areas are extracted from the extracted skin candidate areas based on the image data in the color specification system.
Claim 23
Dependent Claim 60 words
The computer-readable storage medium according to claim 12 , wherein in said extracting step, the image data are divided into rectangle areas, and the rectangle areas are integrated for each of the characteristic when the skin areas are extracted, wherein adjacent pixels are integrated if difference in hue between adjacent pixels is within a threshold determined for each skin type.
Claim 24
Dependent Claim 81 words
An image processor comprising: an area extractor which extracts skin areas from input image data; a classifier which classifies each of the skin areas into one of a plurality of skin types according to characteristics of a plurality of skin colors; and an image corrector which differently corrects image data of each of the plurality of different classifications of the skin areas classified in the same skin type according to which of the plurality of different classifications the skin area belongs.
Claim 25
Dependent Claim 81 words
An image processor comprising: an area extractor which extracts skin areas from input image data; a classifier which classifies each of the skin areas into one of a plurality of skin types according to characteristics of a plurality of skin colors; and an image corrector which differently corrects image data of each of the plurality of different classifications of the skin areas classified in the same skin type according to which of the plurality of different classifications the skin area belongs.
Claim 26
Dependent Claim 38 words
The image processor according to claim 1 , wherein said area extractor firstly extracts skin candidate areas from image data of an entire area and secondly extracts skin areas from the image data in said skin candidate areas.

References Cited (Backward Citations)

US5075788A Cited
US6072526A Cited
US5309228A Cited
US5497431A Cited
US6332041B1 Cited
US6088137A Cited
US6034759A Cited
US6272239B1 Cited
US6549644B1 Cited

USPTO Annuity & Maintenance Windows

Annuity Stage Window Opens Payment Deadline Surcharge Period Until Estimated Fee Status
3.5 Years (Year 4) 2008-11-07 2009-05-09 2009-11-08 $2000.00 Window Closed / Lapsed
7.5 Years (Year 8) 2012-11-07 2013-05-09 2013-11-08 $3760.00 Window Closed / Lapsed
11.5 Years (Year 12) 2016-11-07 2017-05-09 2017-11-08 $7700.00 Window Closed / Lapsed

Assigned Categories & Tags

No categories assigned to this patent yet.

Assign Category Tag