Occupant sensor
First Claim
1. A method of sensing an occupant in a vehicle, comprising:
- a. providing for acquiring a first three-dimensional image of a scene from a first viewing perspective;
b. providing for segmenting said first three-dimensional image so as to identify a region-of-interest in said first three-dimensional image;
c. providing for forming a second three-dimensional image by removing a portion of said first three-dimensional image that is outside of said region-of-interest; and
d. providing for classifying a scenario responsive to an image content of said second three-dimensional image, wherein said image content is represented by a plurality of three-dimensional features selected from a volumetric central moment, a centroid, a volume, a volumetric roundness, a ratio of radii, a volume of a bounding cube, a ratio of volumes and a percentage of volume occupied.
6 Assignments
0 Petitions
Accused Products
Abstract
An occupant sensor incorporates a 3-D imaging system that acquires a 3-D image of an object. The image is segmented to remove unwanted portions and to identify a region-of-interest, and the content thereof is classified responsive to a plurality of 3-D features. In one embodiment, the 3-D image is transformed to a second 3-D image from a second viewing perspective. A two-dimensional projection of the second 3-D image is classified, and a presence, size and position of occupant can be identified from features thereof. A safety restraint system is controlled responsive to the detected scenario, including the presence, position and size of an occupant.
-
Citations
16 Claims
-
1. A method of sensing an occupant in a vehicle, comprising:
-
a. providing for acquiring a first three-dimensional image of a scene from a first viewing perspective;
b. providing for segmenting said first three-dimensional image so as to identify a region-of-interest in said first three-dimensional image;
c. providing for forming a second three-dimensional image by removing a portion of said first three-dimensional image that is outside of said region-of-interest; and
d. providing for classifying a scenario responsive to an image content of said second three-dimensional image, wherein said image content is represented by a plurality of three-dimensional features selected from a volumetric central moment, a centroid, a volume, a volumetric roundness, a ratio of radii, a volume of a bounding cube, a ratio of volumes and a percentage of volume occupied. - View Dependent Claims (2, 3, 4, 5, 6, 7)
-
-
8. A method of sensing an occupant in a vehicle, comprising:
-
a. providing for acquiring a first three-dimensional image of a scene from a first viewing perspective;
b. providing for transforming said first three-dimensional image to a second three-dimensional image from a second viewing perspective;
c. providing for segmenting either said first three-dimensional image prior to said transforming operation, or said second three-dimensional image, so as to identify a region-of-interest in said first or said second three-dimensional image;
d. providing for forming a third three-dimensional image by removing a portion of said first or said second three-dimensional image that is outside of said region-of-interest; and
e. providing for classifying a scenario responsive to an image content of said third three-dimensional image, wherein said image content comprises a two-dimensional representation of said third three-dimensional image. - View Dependent Claims (9, 10, 11, 12, 13, 14, 15)
-
-
16. A method of sensing an occupant in a vehicle, comprising:
-
a. providing for acquiring a first three-dimensional image of a scene from a first viewing perspective;
b. providing for segmenting said first three-dimensional image so as to identify a region-of-interest in said first three-dimensional image;
c. providing for forming a second three-dimensional image by removing a portion of said first three-dimensional image that is outside of said region-of-interest;
d. providing for generating a first representation of a plurality of points within said region-of-interest;
e. comparing said representation with a plurality of a priori representations, wherein each a priori representation represents an object to be identified at a particular position and orientation, and if a difference between said first representation and one of said a priori representations is less than a threshold, then indicating that first three-dimensional image contains an image of said object.
-
Specification