US7218759B1
Expired / Lapsed
US
⏱️ 0.0 years remaining
Face detection in digital images
Assignee: Canon Inc
•
Filing: 1999-12-08
•
Granted: 2007-05-15
•
Expires: 2019-12-08
Abstract
Abstract A method and apparatus for the detection of faces in a digital image captured under a variety of lighting conditions are provided. Rather than subjecting an entire image to computationally intensive face detection analysis, the image is subjected to computationally simply analysis to identify candidate pixels likely to be of skin colour. Only those pixels having such color are subject to computationally intensive face detection analysis. The pixels are selected by first selecting a color distribution model based on an image capture condition provided with the color digial image. Then, the color of the pixels is tested using the distribution model to determine those pixels having predominantly skin color. Preferably, the distribution model is selected from among a number of different distribution models with each of the distribution models associated with a different set of lighting conditions. In this way, a face detection process may accommodate a variety of lighting conditions associated with the capture of an image using a computationally simple analysis.
Inventors
- 👤 Edwin Ho
- 👤 Alison Joan Lennon
- 👤 Andrew Peter Bradley
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 12:01
CPC Classifications
G; G06; G06V; G06V40/00; G06V40/10; G06V40/16; G06V40/161; G06V40/162
Claims Analysis
Total Claims: 47
Claim 1
Independent Claim
96 words
A method of detecting a face in a color digital image formed of a plurality of pixels, said method comprising the steps of: selecting a color distribution model from a plurality of color distribution models, the selection being dependent on at least one image capture condition provided with the color digital image; testing the color of the plurality of pixels to determine those pixels having predominantly skin color, the testing utilizing the selected color distribution model; and executing a facial feature analysis based on the pixels determined as having predominantly skin color in the testing step.
Claim 2
Independent Claim
96 words
A method of detecting a face in a color digital image formed of a plurality of pixels, said method comprising the steps of: selecting a color distribution model from a plurality of color distribution models, the selection being dependent on at least one image capture condition provided with the color digital image; testing the color of the plurality of pixels to determine those pixels having predominantly skin color, the testing utilizing the selected color distribution model; and executing a facial feature analysis based on the pixels determined as having predominantly skin color in the testing step.
Claim 3
Dependent Claim
26 words
A method according to claim 1 , wherein the at least one image capture condition is acquired at a time the color digital image is captured.
Claim 4
Dependent Claim
36 words
A method according to claim 2 , wherein the color digital image is encoded according to a predetermined format and the at least one image capture condition is represented as meta-data associated with the predetermined format.
Claim 5
Dependent Claim
26 words
A method according to claim 1 wherein the at least one image capture condition comprises lighting conditions at a time the color digital image was captured.
Claim 6
Dependent Claim
80 words
A method according to claim 1 , wherein said testing step comprises the sub-step, preceding the testing, of dividing the color digital image into a plurality of regions, each region comprising a plurality of pixels, wherein the testing is performed on pixels within each region to determine those ones of the plurality of regions that are predominantly skin color, and said executing step comprises performing the facial feature analysis on only those regions determined to be predominantly of skin color.
Claim 7
Dependent Claim
43 words
A method according to claim 5 , wherein one of the plurality of regions is determined to be predominantly skin color if more than a predetermined percentage of the total number of pixels in the one region are classified as being skin color.
Claim 8
Dependent Claim
20 words
A method according to claim 5 , wherein the plurality of regions are geometrically divided from the color digital image.
Claim 9
Dependent Claim
20 words
A method according to claim 5 , wherein the plurality of regions are formed of pixels having substantially homogenous color.
Claim 10
Dependent Claim
23 words
A method according to claim 8 , wherein the plurality of regions are formed using a region growing method based upon color differences.
Claim 11
Dependent Claim
16 words
A method according to claim 5 , wherein said executing step is independent of face color.
Claim 12
Dependent Claim
22 words
A method according to claim 1 , wherein the selected color distribution model has been generated using previously sampled facial image data.
Claim 13
Dependent Claim
20 words
A method according to claim 11 , wherein the color distribution model is generated for a particular image capture device.
Claim 14
Dependent Claim
19 words
A method according to claim 11 , wherein separate color distribution models are generated for different image capture conditions.
Claim 15
Dependent Claim
48 words
A method according to claim 13 , wherein the at least one image capture condition comprises lighting conditions at a time the color digital image was captured and separate color models are generated for different lighting conditions at a time the previously sampled facial image data was captured.
Claim 16
Dependent Claim
28 words
A method according to claim 14 , wherein separate color distribution models are generated for groups of images taken with a flash and images taken without a flash.
Claim 17
Dependent Claim
24 words
A method according to claim 14 , wherein separate color distribution models are generated for groups of images taken indoors and images taken outdoors.
Claim 18
Dependent Claim
22 words
A method according to claim 11 , wherein each color distribution model is represented as a frequency histogram of color representation vectors.
Claim 19
Dependent Claim
35 words
A method according to claim 17 , wherein each of the plurality of pixels is classified as being skin color if the frequency of the color representation vector corresponding thereto exceeds a predetermined threshold frequency.
Claim 20
Dependent Claim
22 words
A method according to claim 11 , wherein each color distribution model is represented as a probability distribution of color representation vectors.
Claim 21
Dependent Claim
35 words
A method according to claim 19 , wherein each of the plurality of pixels is classified as being skin color if the probability of the color representation vector corresponding thereto exceeds a predetermined probability threshold.
Claim 22
Dependent Claim
22 words
A method according to claim 11 , wherein each color distribution model is represented as a binary map of color representation vectors.
Claim 23
Dependent Claim
34 words
A method according to claim 17 , 19 or 21 , wherein the color representation vectors are derived from perceptual color space values of predetermined skin-color pixels in the previously sampled facial image data.
Claim 24
Dependent Claim
35 words
A method according to claim 17 , 19 or 21 , wherein color representation vectors contain chromatic color values derived from those RGB values of predetermined skin-color pixels in the previously sampled facial image data.
Claim 25
Dependent Claim
29 words
A method according to claim 21 , wherein the binary map comprises a percentage of the skin color pixels that were identified in the previously sampled facial image data.
Claim 26
Dependent Claim
32 words
A method according to claim 24 , wherein one of the plurality of pixels is classified as being skin color if the color representation vector corresponding thereto occurs within the binary map.
Claim 27
Dependent Claim
31 words
A method according to claim 1 , wherein said testing step for determining whether a pixel has predominantly skin color, the threshold selected using the at least one image capture condition.
Claim 28
Independent Claim
101 words
An apparatus for detecting a face in a color digital image formed of a plurality of pixels, said apparatus comprising: means for selecting a color distribution model from a plurality of color distribution models, the selection being dependent on at least one image capture condition provided with the color digital image; means for testing the color of the plurality of pixels to determine those pixels having predominantly skin color, said testing means utilizing the selected color distribution model; and means for executing a facial feature analysis based on the pixels determined as having predominantly skin color by the means for testing.
Claim 29
Independent Claim
101 words
An apparatus for detecting a face in a color digital image formed of a plurality of pixels, said apparatus comprising: means for selecting a color distribution model from a plurality of color distribution models, the selection being dependent on at least one image capture condition provided with the color digital image; means for testing the color of the plurality of pixels to determine those pixels having predominantly skin color, said testing means utilizing the selected color distribution model; and means for executing a facial feature analysis based on the pixels determined as having predominantly skin color by the means for testing.
Claim 30
Dependent Claim
26 words
An apparatus according to claim 27 , wherein the at least one image capture condition is acquired at a time the color digital image is captured.
Claim 31
Dependent Claim
36 words
An apparatus according to claim 28 , wherein the color digital image is encoded according to a predetermined format and the at least one image capture condition is represented as meta-data associated with the predetermined format.
Claim 32
Dependent Claim
27 words
An apparatus according to claim 27 , wherein the at least one image capture condition comprises lighting conditions at a time the color digital image was captured.
Claim 33
Dependent Claim
82 words
An apparatus according to claim 27 , wherein said means for testing comprises means for dividing the color digital image into a plurality of regions, each region comprising a plurality of pixels, wherein said means for testing operates on pixels within each region to determine those ones of the plurality of regions that are predominantly skin color and said means for executing causes the further facial feature analysis to be performed on only those regions determined to be predominantly of skin color.
Claim 34
Dependent Claim
108 words
A computer readable medium incorporating a computer program product for detecting a face in a color digital image formed of a plurality of pixels, said computer program product comprising: code for selecting a color distribution model from a plurality of color distribution models, the selection being dependent on at least one image capture condition provided with the color digital image; code for testing the color of the plurality of pixels to determine those pixels having predominantly skin color, the testing utilizing the selected color distribution model; and code for executing facial feature analysis based on the pixels determined as having predominantly skin color by the code for testing.
Claim 35
Dependent Claim
108 words
A computer readable medium incorporating a computer program product for detecting a face in a color digital image formed of a plurality of pixels, said computer program product comprising: code for selecting a color distribution model from a plurality of color distribution models, the selection being dependent on at least one image capture condition provided with the color digital image; code for testing the color of the plurality of pixels to determine those pixels having predominantly skin color, the testing utilizing the selected color distribution model; and code for executing facial feature analysis based on the pixels determined as having predominantly skin color by the code for testing.
Claim 36
Dependent Claim
28 words
A computer readable medium according to claim 32 , wherein the at least one image capture condition is acquired at a time the color digital image is captured.
Claim 37
Dependent Claim
38 words
A computer readable medium according to claim 33 , wherein the color digital image is encoded according to a predetermined format and the at least one image capture condition is represented as meta-data associated with the predetermined format.
Claim 38
Dependent Claim
29 words
A computer readable medium according to claim 32 , wherein the at least one image capture condition comprises lighting conditions at a time the color digital image was captured.
Claim 39
Dependent Claim
83 words
A computer readable medium according to claim 32 , wherein said code for testing comprises code for dividing the color digital image into a plurality of regions, each region comprising a plurality of pixels, wherein said code for testing operates on pixels within each region to determine those ones of the plurality of regions that are predominantly skin color, and said code for executing causes the facial feature analysis to be performed on only those regions determined to be predominantly of skin color.
Claim 40
Independent Claim
99 words
A method of detecting a face in a color digital image formed of a plurality of pixels, and method comprising the step of: testing the color of the plurality of pixels to determine those pixels having predominantly skin color, the testing utilizing at least one threshold dependent on the least one image capture condition provided with the color digital image, said image capture condition being expressed as metadata and not depending on a brightness of the digital image data; and executing a facial feature analysis based on the pixels determined as having predominantly skin color in the testing step.
Claim 41
Independent Claim
99 words
A method of detecting a face in a color digital image formed of a plurality of pixels, and method comprising the step of: testing the color of the plurality of pixels to determine those pixels having predominantly skin color, the testing utilizing at least one threshold dependent on the least one image capture condition provided with the color digital image, said image capture condition being expressed as metadata and not depending on a brightness of the digital image data; and executing a facial feature analysis based on the pixels determined as having predominantly skin color in the testing step.
Claim 42
Independent Claim
98 words
A method of detecting a face in a color digital image formed of a plurality of pixels, said method comprising the step of: selecting a threshold from a plurality of thresholds, the selection being dependent on at least one image capture condition provided with the color digital image, said image capture condition being expressed as metadata; testing the color of the plurality of pixels to determine those pixels having predominantly skin color, the testing utilizing the selected threshold; and executing a facial feature analysis based on the pixels determined as having predominantly skin color in the testing step.
Claim 43
Independent Claim
98 words
A method of detecting a face in a color digital image formed of a plurality of pixels, said method comprising the step of: selecting a threshold from a plurality of thresholds, the selection being dependent on at least one image capture condition provided with the color digital image, said image capture condition being expressed as metadata; testing the color of the plurality of pixels to determine those pixels having predominantly skin color, the testing utilizing the selected threshold; and executing a facial feature analysis based on the pixels determined as having predominantly skin color in the testing step.
Claim 44
Dependent Claim
27 words
A method according to claim 38 , wherein the at least one image capture condition comprises lighting conditions at a time the color digital image was captured.
Claim 45
Dependent Claim
37 words
A method according to claim 38 , wherein said threshold comprises a color distribution model selected from a plurality of color distribution, models, the selected color distribution model having been generated using previously sampled facial image data.
Claim 46
Dependent Claim
114 words
A computer readable medium having a computer program recorded thereon and executable to make a computer detect a face in a color digital image formed of a plurality of pixels, said program comprising: code for selecting a threshold from a plurality of thresholds, the selection being dependent on at least one image capture condition provided with the color digital image, said image capture condition being expressed as metadata; code for testing the color of the plurality of pixels to determine those pixels having predominantly skin color, the testing utilizing the selected threshold; and code for executing a fascial feature analysis based on the pixels determined as having predominantly skin color in the testing step.
Claim 47
Dependent Claim
114 words
A computer readable medium having a computer program recorded thereon and executable to make a computer detect a face in a color digital image formed of a plurality of pixels, said program comprising: code for selecting a threshold from a plurality of thresholds, the selection being dependent on at least one image capture condition provided with the color digital image, said image capture condition being expressed as metadata; code for testing the color of the plurality of pixels to determine those pixels having predominantly skin color, the testing utilizing the selected threshold; and code for executing a fascial feature analysis based on the pixels determined as having predominantly skin color in the testing step.
References Cited (Backward Citations)
US4203671A
Cited
US5488429A
Cited
US5430809A
Cited
US5812193A
Cited
US5557688A
Cited
US5748776A
Cited
US5715377A
Cited
US5912980A
Cited
US5781665A
Cited
US5802220A
Cited
US6028960A
Cited
EP0836326A2
Cited
US5864630A
Cited
US6188777B1
Cited
US6292574B1
Cited
US6148092A
Cited
US6332033B1
Cited
US6678407B1
Cited
US6072496A
Cited
US6633655B1
Cited
US6263113B1
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2010-05-14 | 2010-11-13 | 2011-05-15 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2014-05-14 | 2014-11-13 | 2015-05-15 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2018-05-14 | 2018-11-13 | 2019-05-15 | $7700.00 | Window Closed / Lapsed |