US6711286B1
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method for blond-hair-pixel removal in image skin-color detection
Assignee: Monument Peak Ventures LLC
•
Filing: 2000-10-20
•
Granted: 2004-03-23
•
Expires: 2020-10-20
Abstract
Abstract A computer vision/image processing method of removing blond hair color pixels in digital image skin detection for a variety of imaging related applications, such as redeye defects detection, employs a combination of skin detectors operating in a generalized RGB space in combination with a hue space derived from the original image space to detect skin pixels and blond hair pixels within the skin pixels.
Inventors
- 👤 Shoupu Chen
- 👤 Lawrence A. Ray
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 12:02
CPC Classifications
G; G06; G06T; G06T7/00; G06T7/10; G06T7/11; G06V; G06V10/00; G06V10/20; G06V10/26; G06V10/273; G06V10/28; G06V10/40; G06V10/56; G06T2207/00; G06T2207/30; G06T2207/30196; G06T2207/30201
Claims Analysis
Total Claims: 18
Claim 1
Independent Claim
135 words
A method for removing blond hair color pixels during skin color detection in a digital RGB image, the method comprising the steps of: (a) converting the image from an RGB space to a generalized RGB space, thereby obtaining a gRGB image; (b) detecting skin color pixels within the GRGB image; (c) forming a first skin color image mask based on the detected skin color pixels; (d) generating a masked GRGB image using the first skin color image mask; (e) converting the masked gRGB image to a hue image; (f) separating blond hair color and the associated skin color within the hue image; (g) forming a second skin color image mask based on the separated skin color pixels in the hue image; and (h) generating a masked RGB image using the second skin color image mask.
Claim 2
Independent Claim
135 words
A method for removing blond hair color pixels during skin color detection in a digital RGB image, the method comprising the steps of: (a) converting the image from an RGB space to a generalized RGB space, thereby obtaining a gRGB image; (b) detecting skin color pixels within the GRGB image; (c) forming a first skin color image mask based on the detected skin color pixels; (d) generating a masked GRGB image using the first skin color image mask; (e) converting the masked gRGB image to a hue image; (f) separating blond hair color and the associated skin color within the hue image; (g) forming a second skin color image mask based on the separated skin color pixels in the hue image; and (h) generating a masked RGB image using the second skin color image mask.
Claim 3
Dependent Claim
24 words
A method as claimed in claim 1 wherein the step (b) of detecting skin color pixels within the gRGB image employs an ellipse classifier.
Claim 4
Dependent Claim
24 words
A method as claimed in claim 2 wherein the ellipse classifier classifies skin color pixels and non-skin color pixels in the generalized RG space.
Claim 5
Dependent Claim
59 words
A method as claimed in claim 3 wherein the ellipse classification comprises the steps of: (a) calculating a value for each of the pixels using the generalized R and G components; (b) performing the calculation in step (a) using ellipse parameters, and (c) classifying skin color pixels or non-skin color pixels based on the value calculated in step (a).
Claim 6
Dependent Claim
21 words
A method as claimed in claim 1 , wherein the conversion in step (e) is performed on the masked gRGB image.
Claim 7
Dependent Claim
32 words
A method as claimed in claim 1 , wherein the separation of blond hair color pixel and the associated skin color pixel in step (f) is performed in one dimensional hue space.
Claim 8
Dependent Claim
38 words
A method as claimed in claim 6 , wherein the separation of blond hair color pixel and the associated skin color pixel in step (f) is performed by means of a histogram metric in one dimensional hue space.
Claim 9
Dependent Claim
12 words
A computer program product employing the method claimed in claim 1 .
Claim 10
Independent Claim
108 words
A method for removing blond hair color pixels during skin color detection in an original digital image space, the method comprising the steps of: (a) converting the original digital image space to a normalized space, thereby obtaining a normalized image; (b) detecting skin color pixels within the normalized image; (c) generating a masked normalized image using the detected skin color pixels; (d) converting the masked normalized image to a one-dimensional image space; (e) separating blond hair color and the associated skin color within the one-dimensional image space; and (f) utilizing the separated pixels in the one-dimensional space to identify blond hair color pixels in the original image space.
Claim 11
Independent Claim
108 words
A method for removing blond hair color pixels during skin color detection in an original digital image space, the method comprising the steps of: (a) converting the original digital image space to a normalized space, thereby obtaining a normalized image; (b) detecting skin color pixels within the normalized image; (c) generating a masked normalized image using the detected skin color pixels; (d) converting the masked normalized image to a one-dimensional image space; (e) separating blond hair color and the associated skin color within the one-dimensional image space; and (f) utilizing the separated pixels in the one-dimensional space to identify blond hair color pixels in the original image space.
Claim 12
Dependent Claim
50 words
The method as claimed in claim 9 wherein step (f) comprises the steps of: (f1) forming a skin color image mask based on the separated skin color pixels in the one-dimensional image space; and (f2) generating a masked image in the original image space using the skin color image mask.
Claim 13
Dependent Claim
16 words
The method as claimed in claim 9 wherein the one-dimensional image space is a hue image.
Claim 14
Dependent Claim
17 words
The method as claimed in claim 9 wherein the original digital image space is an RGB space.
Claim 15
Dependent Claim
16 words
The method as claimed in claim 12 wherein the normalized space is a generalized RGB space.
Claim 16
Dependent Claim
12 words
A computer program product employing the method claimed in claim 9 .
Claim 17
Independent Claim
58 words
A system for detecting blond hair color pixels in an original digital image space, said system comprising a combination of skin detectors operating in a generalized RGB image space and a in sole combination with a one-dimensional hue image space derived from the original image space to detect skin pixels and blond hair pixels within the skin pixels.
Claim 18
Independent Claim
58 words
A system for detecting blond hair color pixels in an original digital image space, said system comprising a combination of skin detectors operating in a generalized RGB image space and a in sole combination with a one-dimensional hue image space derived from the original image space to detect skin pixels and blond hair pixels within the skin pixels.
References Cited (Backward Citations)
US3778541A
Cited
US5128711A
Cited
US6249317B1
Cited
US6072526A
Cited
US5907315A
Cited
US5748764A
Cited
US5940530A
Cited
US5850463A
Cited
US6343141B1
Cited
US6292574B1
Cited
US6611613B1
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2007-03-23 | 2007-09-22 | 2008-03-23 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2011-03-23 | 2011-09-22 | 2012-03-23 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2015-03-23 | 2015-09-22 | 2016-03-23 | $7700.00 | Window Closed / Lapsed |