Patent US2808908A successfully refreshed from Google Patents.
Patent US6895103B2 successfully refreshed from Google Patents.
Patent US7103218B2 successfully refreshed from Google Patents.
US7103218B2
Expired / Lapsed
US
⏱️ 0.0 years remaining
Image processing methods and apparatus for detecting human eyes, human face, and other objects in an image
Assignee: Canon Inc
•
Filing: 2005-09-27
•
Granted: 2006-09-05
•
Expires: 2025-09-27
Abstract
Abstract For a subset of pixels in an image in which it is desired to detect a human face if one is present, a first variable is derived from the gray-level distribution of the image, and a second variable is derived from a preset reference distribution that is characteristic of the object. The correspondence between the first variable and the second variable is then evaluated over the subset of pixels, and a determination is made as to whether the image contains the object, based on the result of this evaluation.
Inventors
- 👤 Xinwu Chen
- 👤 Xin Ji
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 09:48
CPC Classifications
G; G06; G06T; G06T7/00; G06T7/10; G06T7/12; G06V; G06V40/00; G06V40/10; G06V40/16; G06V40/161; G06T2207/00; G06T2207/20; G06T2207/20112; G06T2207/20132; G06T2207/30; G06T2207/30196; G06T2207/30201
Claims Analysis
Total Claims: 21
Claim 1
Independent Claim
116 words
A method for detecting an object in an image having a gray-level distribution, comprising the steps of: determining a sub-image by detecting a pair of dark areas in the image; selecting an area on the basis of the location of the sub-image; for a subset of pixels in the selected area, deriving a first variable from the gray-level distribution of the image, for that subset of pixels, deriving a second variable from a preset reference distribution, the reference distribution being characteristic of the object; evaluating the correspondence between the first variable and the second variable over the subset of pixels; and determining whether the image contains the object, based on the result of said evaluation step.
Claim 2
Independent Claim
116 words
A method for detecting an object in an image having a gray-level distribution, comprising the steps of: determining a sub-image by detecting a pair of dark areas in the image; selecting an area on the basis of the location of the sub-image; for a subset of pixels in the selected area, deriving a first variable from the gray-level distribution of the image, for that subset of pixels, deriving a second variable from a preset reference distribution, the reference distribution being characteristic of the object; evaluating the correspondence between the first variable and the second variable over the subset of pixels; and determining whether the image contains the object, based on the result of said evaluation step.
Claim 3
Dependent Claim
23 words
The method of claim 1 , wherein the first variable presents the direction of the gradient of the gray-level distribution of the image.
Claim 4
Dependent Claim
20 words
The method of claim 2 , wherein the second variable represents the direction of the gradient of the reference distribution.
Claim 5
Dependent Claim
13 words
The method of claim 3 , wherein the gradient is a discrete gradient.
Claim 6
Dependent Claim
20 words
The method of claim 3 , wherein said evaluation step includes a statistical process performed over the subset of pixels.
Claim 7
Dependent Claim
71 words
The method of claim 3 , wherein said evaluation step includes a statistical processing over the subset of pixels and the statistical process comprises a weighting process, in which the weight assigned to each pixel in the subset is determined on the basis of the gradient of the gray level distribution at the pixel and pixels with larger magnitude of the gradient of the gray level distribution generally have larger weights.
Claim 8
Dependent Claim
44 words
The method of claim 6 , wherein said evaluating step includes both the statistical process using the weighting process and another statistical process without the weighting process and determines whether the image contains the object in accordance with the results of both statistical processes.
Claim 9
Dependent Claim
71 words
The method of claim 3 , wherein said evaluating step includes: for all the pixels in the subset, calculating the average value of the angle between the gradient of the gray-level distribution of the image and the gradient of the reference distribution; comparing the average value with a predetermined value; and determining that the sub-image contains the object if the average value is equal to or smaller than the predetermined value.
Claim 10
Dependent Claim
91 words
The method of claim 3 , wherein said evaluating step includes: for all the pixels in the subset, calculating a weighted average value of the angle between the gradient of the gray-level distribution of the image and the gradient of the reference distribution, wherein a pixel having a greater magnitude of gradient of gray-level distribution generally has a larger weight; comparing the weighted average value with a predetermined value; and determining that the sub-image contains the object if the weighted average value is equal to or smaller than the predetermined value.
Claim 11
Dependent Claim
146 words
The method of claim 3 , wherein said evaluating step includes: for all the pixels in the subset, calculating the average value of the angle between the gradient of the gray-level distribution of the image and the gradient of the reference distribution; for all the pixels in the subset, calculating a weighted average value of the angle between the gradient of the gray-level distribution of the image and gradient of the reference distribution, wherein a pixel having a greater magnitude of gradient of gray-level distribution has a larger weight; comparing the average value with a first predetermined value; comparing the weighted average value with a second predetermined value; and determining that the sub-image contains said object if the average value is equal to or is smaller than the first predetermined value and the weighted average value is equal to or smaller than the second predetermined value.
Claim 12
Dependent Claim
22 words
The method of claim 1 , wherein the selected area is an annular area, which centers around the center of the sub-image.
Claim 13
Dependent Claim
19 words
The method of claim 1 , wherein the subset of pixels includes all the pixels in the selected area.
Claim 14
Dependent Claim
27 words
The method of claim 1 , wherein the sub-image is determined based on the positions of and interval between the centers of the pair of dark areas.
Claim 15
Dependent Claim
57 words
The method of claim 1 , wherein the reference distribution is expressed by z ( x,y )=−( x 2 /a 2 +y 2 /b 2)+ h where h is a constant, a/b equals the ratio of the width of the sub-image to the height of the sub-image, and the origin is at the center of the sub-image.
Claim 16
Independent Claim
115 words
A method for detecting an object in an image having a gray-level distribution, comprising the steps of: determining a sub-image by detecting a pair of dark areas in the image; selecting a subset of pixels in the image based on the location of the sub-image; for pixels of the subset, deriving a first variable from the gray-level distribution of the image; for the subset of pixels, deriving a second variable from a preset reference distribution, the reference distribution being characteristic of the object; evaluating the correspondence between the first variable and the second variable over the subset of pixels; and determining whether the image contains the object based on the result of said evaluation step.
Claim 17
Independent Claim
115 words
A method for detecting an object in an image having a gray-level distribution, comprising the steps of: determining a sub-image by detecting a pair of dark areas in the image; selecting a subset of pixels in the image based on the location of the sub-image; for pixels of the subset, deriving a first variable from the gray-level distribution of the image; for the subset of pixels, deriving a second variable from a preset reference distribution, the reference distribution being characteristic of the object; evaluating the correspondence between the first variable and the second variable over the subset of pixels; and determining whether the image contains the object based on the result of said evaluation step.
Claim 18
Dependent Claim
138 words
A computer-readable storage medium with program code stored therein for detecting an object in an image with a gray-level distribution, said program code comprising: codes for determining a sub-image by detecting a pair of dark areas in the image; codes for selecting an area on the basis of the location of the sub-image: codes for deriving, for a subset of pixels in the selected area, a first variable from the gray-level distribution of the image, codes for deriving, for the subset of pixels, a second variable from a preset reference distribution, the reference distribution being characteristic of the object; codes for evaluating the correspondence between the first variable and the second variable over the subset of pixels; and codes for determining whether the image contains the object based on the result of execution of said codes for evaluating.
Claim 19
Dependent Claim
138 words
A computer-readable storage medium with program code stored therein for detecting an object in an image with a gray-level distribution, said program code comprising: codes for determining a sub-image by detecting a pair of dark areas in the image; codes for selecting an area on the basis of the location of the sub-image: codes for deriving, for a subset of pixels in the selected area, a first variable from the gray-level distribution of the image, codes for deriving, for the subset of pixels, a second variable from a preset reference distribution, the reference distribution being characteristic of the object; codes for evaluating the correspondence between the first variable and the second variable over the subset of pixels; and codes for determining whether the image contains the object based on the result of execution of said codes for evaluating.
Claim 20
Independent Claim
149 words
An apparatus for detecting an object in an image having a gray-level distribution, comprising: a sub-image determination unit adapted to determine a sub-image by detecting a pair of dark areas in the image; a sub-image processor adapted to select a subset of pixels on the basis of the location of the sub-image; a first calculation unit, adapted to derive a first variable from the gray-level distribution for the subset of pixels; a second calculation unit, adapted to derive a second variable from a preset reference distribution for the subset of pixels, the reference distribution being characteristic of the object; a correspondence evaluation processor, adapted to evaluate the correspondence between the first variable and the second variable over the subset of pixels; and a determination unit, adapted to make a determination as to whether the image contains the object based on the evaluation result produced by said correspondence evaluation unit.
Claim 21
Independent Claim
149 words
An apparatus for detecting an object in an image having a gray-level distribution, comprising: a sub-image determination unit adapted to determine a sub-image by detecting a pair of dark areas in the image; a sub-image processor adapted to select a subset of pixels on the basis of the location of the sub-image; a first calculation unit, adapted to derive a first variable from the gray-level distribution for the subset of pixels; a second calculation unit, adapted to derive a second variable from a preset reference distribution for the subset of pixels, the reference distribution being characteristic of the object; a correspondence evaluation processor, adapted to evaluate the correspondence between the first variable and the second variable over the subset of pixels; and a determination unit, adapted to make a determination as to whether the image contains the object based on the evaluation result produced by said correspondence evaluation unit.
References Cited (Backward Citations)
US4958374A
Cited
US5293427A
Cited
US5499303A
Cited
US6115495A
Cited
US5859921A
Cited
US6154566A
Cited
US6072892A
Cited
EP1011064A2
Cited
US6718050B1
Cited
US6130617A
Cited
US6532301B1
Cited
US6876754B1
Cited
US6611622B1
Cited
US20020114495A1
Cited
US6873714B2
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2009-09-04 | 2010-03-06 | 2010-09-05 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2013-09-04 | 2014-03-06 | 2014-09-05 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2017-09-04 | 2018-03-06 | 2018-09-05 | $7700.00 | Window Closed / Lapsed |