Back to results
US 9,805,251 B2Granted
Systems and methods for yaw estimation
AssigneeAmrita Vishwa Vidyapeetham
InventorsAthinarayanan Sankaranarayanan, Kamal Bijlani
Life of this patent
Estimating the term…
Abstract
Systems and methods of automatic detection of a facial feature are disclosed. Moreover, methods and systems of yaw estimation of a human head based on a geometrical model are also disclosed.
Claims3
An operator inattention monitoring system, comprising:
- a processing device configured to determine an attention level of the operator, said device comprising a processor, memory and at least one communication interface;
- wherein the processing device is configured to obtain frontal images of the operator's head over a period of time, wherein the images comprise the head, the neck, and the face of the operator;
- and
- wherein the processing device is configured to analyze a plurality of yaw angles to determine the attention level of the operator;
- wherein the processor is configured to:
- compute the plurality of yaw angles of the operator's head from the obtained images by
- a) modeling the head as an ellipse in vertical projection;
- b) fixing the center of rotation of the ellipse as the center of the neck;
- c) truncating the ellipse along the major axis representing a region of the face to an arc of ±60°;
- d) mapping a center of the face as the projection of a nose;
- e) locating boundaries of the face on the ellipse by identifying coordinates on the elliptic arc subtending ±60°;
- and
- f) determining head rotation using position of the nose and the boundaries of the face;
- wherein the ellipse is described by the equation
- x
- 2
- r
- 2
- +
- (
- y
- +
- δ
-
-
- r
- +
- α
-
-
- r
- )
- 2
- (
- β
-
-
- r
- )
- 2
- =
- 1
- ,
- wherein the parametric form of ellipse is x=r cos t, y=βr sin t−αr, for tε[0, 2π], wherein, r is half of the frontal face width, α is a factor representing the extent of truncation, δ is a factor representing the distance of the pivot point from the center of the ellipse, β is the ratio of the major to the minor axes.
41 more elements
An operator inattention monitoring system, comprising:
- a processing device configured to determine an attention level of the operator, said device comprising a processor, memory and at least one communication interface;
- wherein the processing device is configured to obtain frontal images of the operator's head over a period of time, wherein the images comprise the head, the neck, and the face of the operator;
- and
- wherein the processing device is configured to analyze a plurality of yaw angles to determine the attention level of the operator;
- wherein the processor is configured to:
- compute the plurality of yaw angles of the operator's head from the obtained images by
- a) modeling the head as an ellipse in vertical projection;
- b) fixing the center of rotation of the ellipse as the center of the neck;
- c) truncating the ellipse along the major axis representing a region of the face to an arc of ±60°;
- d) mapping a center of the face as the projection of a nose;
- e) locating boundaries of the face on the ellipse by identifying coordinates on the elliptic arc subtending ±60°;
- and
- f) determining head rotation using position of the nose and the boundaries of the face;
- wherein the ellipse rotated in clockwise and counter-clockwise direction by a yaw angle θ from boundaries of ellipse comprises:
- a. face center, xc=±(α+β)r sin θ
- b. face boundaries comprising left and right boundaries,
- x
- l
- =
- ±
- (
- -
- r
-
- 1
- +
- (
- β
- 2
- -
- γ
- b
- 2
- -
- 1
- )
-
- sin
- 2
-
- θ
- +
- δ
-
-
- r
-
-
- sin
-
-
- θ
- +
- α
-
-
- r
-
-
- sin
-
-
- θ
- )
- x
- r
- =
- ±
- (
- r
-
- 1
- +
- (
- β
- 2
- -
- γ
- b
- 2
- -
- 1
- )
-
- sin
- 2
-
- θ
- +
- δ
-
-
- r
-
-
- sin
-
-
- θ
- +
- α
-
-
- r
-
-
- sin
-
-
- θ
- )
- c. yaw angle,
- θ
- =
- sin
- -
- 1
-
- {
- x
- c
- -
- (
- x
- l
- +
- x
- r
- 2
- )
- (
- β
- +
- γ
- c
- )
-
- r
- }
- d. half face-width,
- r
- =
- ±
- (
- (
- x
- r
- -
- x
- l
- )
- 2
- -
- (
- β
- 2
- -
- γ
- b
- 2
- -
- 1
- )
- β
- 2
-
- {
- 2
-
- x
- c
- -
- (
- x
- l
- +
- x
- r
- )
- }
- 2
- )
- .
180 more elements
Dependencies are inferred from the claim text.