Occupant information detection system, occupant restraint system, and vehicle

• US 7,630,804 B2
• Filed: 07/19/2006
• Issued: 12/08/2009
• Est. Priority Date: 07/19/2005
• Status: Expired due to Fees

• Current Assignee: Joyson Safety Systems Japan K.K.
• Original Assignee: Takata Corporation (Ningbo Junsheng Investment Group Co., Ltd.)
• Inventors: Aoki, Hiroshi, Yokoo, Masato, Hakomori, Yuu
• Primary Examiner(s): Hellner; Mark
• Application Number:
  US11/458,505
• Publication Number:
  US 20080021616A1
• Time in Patent Office: 1,238 Days
• Field of Search: 701/45, 701/46, 180/271, 382/140, 382/154, 340/576, 340/573.1, 340/575
• US Class Current: 701/45
• CPC Class Codes: B60R 21/0152 using strain gauges
• CPC Class Codes: B60R 21/01538 for image processing, e.g. cameras or sensor arrays
• CPC Class Codes: B60R 21/01552 detecting position of specific human body parts, e.g. face, eyes or hands
• CPC Class Codes: G06V 20/64 Three-dimensional objects
• CPC Class Codes: G06V 40/103 Static body considered as a whole, e.g. static pedestrian or occupant recognition

Abstract

A technology is provided which is effective for precisely detecting information about a vehicle occupant in a vehicle seat. In one form, an occupant restraint system to be installed in a vehicle is adapted to associate occupant surface profile information relating to a specific body region of a vehicle occupant, which is obtained by a camera for obtaining an image relating to a single view point, with previously stored body information so as to derive body information about the vehicle occupant and varies the occupant restraining mode of an occupant restraint device according to the derived body information.

Claims

1. In combination, a vehicle and a detection apparatus for detecting information about a vehicle occupant, the detection apparatus comprising:

a cabin of the vehicle having a seat for the vehicle occupant;

a detection device mounted in the cabin for obtaining an image of the vehicle occupant and the vehicle seat on which the occupant is to be seated;

a storage device for storing information relating to potential vehicle occupants seated on the vehicle seat;

a controller configured to derive positional information from the obtained image of a predetermined one of a plurality of body regions of the vehicle occupant by comparing the obtained image with the stored information relating to the potential vehicle occupants, the controller further being configured to derive positional information from the obtained image of a predetermined reference region of the vehicle seat; and

a body size estimation unit of the controller that is configured to derive body size information of the vehicle occupant including at least one of sitting height, chest thickness, and shoulder width of the vehicle occupant based on the derived positional information of the one body region and the derived positional information of the predetermined reference region of the vehicle seat.

2. The detection apparatus of claim 1, wherein, the stored information relating to potential vehicle occupants includes positional relations of body regions of human bodies, and the controller is configured to derive the positional information of the body region by comparing a positional relation between adjacent body regions in the obtained image to the positional relation of stored body regions.

3. The detection apparatus of claim 1, wherein the predetermined one of the body regions is a neck region.

4. The detection apparatus of claim 1, wherein the controller includes a seat information extraction unit that determines positioning of the vehicle seat reference region from the obtained image.

5. The detection apparatus of claim 1, in combination with an occupant protection device, and the controller is operable to selectively actuate the occupant protection device based on the vehicle occupant'"'"'s body size information.

6. The detection apparatus of claim 1, wherein the controller is configured to determine the positional information relating to the vehicle occupant by requiring a predetermined degree of accuracy of association between the obtained image and the stored information relating to the potential vehicle occupant.

7. A method of determining information regarding a vehicle occupant, the method comprising:

storing information regarding configurations of individual body regions including neck and adjacent shoulder and jaw regions and positional relations between individual ones of the body regions and body regions adjacent thereto of potential vehicle occupants;

obtaining an image of the vehicle occupant in the vehicle;

selecting a region of the vehicle occupant image determined as corresponding to only the neck region of the vehicle occupant based on an initial comparison of the selected image region with the stored neck region configurations;

subsequently comparing positional relations between the selected image region determined as corresponding to only the neck region and adjacent image regions with the stored information regarding the positional relations between the stored neck region configurations and the body regions adjacent thereto including the shoulder and jaw regions when the initial comparison results in multiple ones of the stored individual body regions as being a potential match for only the neck region so that the initial comparison generally is inconclusive and so that the subsequent comparison optimizes the accuracy of matching the selected image region to only the neck region; and

deriving body size information regarding the vehicle occupant based on information relating to the selected image region corresponding to only the neck region via the initial comparison and, when the initial comparison is generally inconclusive, the subsequent comparison.

8. The method of claim 7, further comprising selectively controlling a restraint device at least partially dependent on the derived body size information regarding the vehicle occupant.

9. The method of claim 7, further comprising applying predetermined criteria relating to the accuracy of the obtained image of the vehicle occupant, or the similarity between the obtained image and the stored information, and the body size information regarding the vehicle occupant is further derived by screening out the unreliable image regions based on the predetermined criteria.

10. The method of claim 7, further comprising obtaining positional information regarding a vehicle reference region;

and

the body size information of the vehicle occupant is derived based on the information relating to the image region matched to the selected body region of the vehicle occupant and the positional information regarding the vehicle reference region.

11. The method of claim 10, wherein the vehicle reference region is a vehicle seat, and the positional information regarding the vehicle seat is obtained from a seat position detector.

12. The method of claim 10, further comprising selectively controlling a restraint device based on the estimated size of the vehicle occupant.

13. The method of claim 7, further comprising selectively controlling a vehicle system using the derived body size information regarding the vehicle occupant.

14. The method of claim 7, further comprising obtaining positional information regarding a vehicle seat;

the body size information of the vehicle occupant is derived based on the information relating to the image region matched to the selected body region of the vehicle occupant and the positional information regarding the vehicle seat; and

selectively controlling a restraint device based on the estimated size of the vehicle occupant.

15. The method of claim 14, further comprising detecting an emergency condition of the vehicle, wherein the restraint device is further selectively controlled based on the detected emergency.