Adaptive visual occupant detection and classification system

• US 8,560,179 B2
• Filed: 02/20/2004
• Issued: 10/15/2013
• Est. Priority Date: 02/20/2003
• Status: Expired due to Fees

• Current Assignee: Intelligent Mechatronic Systems Inc. (Trak Global Group Ltd.)
• Original Assignee: Intelligent Mechatronic Systems Inc. (Trak Global Group Ltd.)
• Inventors: Basir, Otman Adam, Bullock, David, Breza, Emil
• Primary Examiner(s): To, Tuan C.
• Application Number:
  US10/783,390
• Publication Number:
  US 20040186642A1
• Time in Patent Office: 3,525 Days
• Field of Search: 701/45, 701/1, 382/115, 382/190, 382/224, 382/159, 382/220, 382/104, 382/154, 382/181, 348/143, 280/735, 180/271, 345/598, 345/599, 345/614, 345/644, 345/698, 706/20
• US Class Current: 701/45
• CPC Class Codes: B60N 2/002 Seats provided with an occupancy detection means mounted therein or thereon detection means responsive to presence or absence of children B60N2/266
• CPC Class Codes: B60N 2/286 forward facing B60N2/2866 takes precedence
• CPC Class Codes: B60N 2/2863 backward facing
• CPC Class Codes: B60R 21/01538 for image processing, e.g. cameras or sensor arrays
• CPC Class Codes: B60R 21/01542 detecting passenger motion
• CPC Class Codes: G06V 40/10 Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands

Abstract

A vision-based vehicle occupant detection and classification system includes a camera mounted in the vehicle interior to capture low resolution images of the seating area. Image processing algorithms account for lighting, motion, and other phenomena. A spatial-feature matrix numerically describes the content of the image. This descriptor is the result of a number of digital filters being run against a set of sub-images, derived from pre-defined window regions in the original image. This spatial-feature matrix is used as an input to an expert classifier function, which classifies this image as best representing a scenario in which the seat is (i) empty, (ii) occupied by an adult, (iii) occupied by a child, (iv) occupied by a rear-facing infant seat (RFIS), (v) occupied by a front-facing infant seat (FFIS), or (vi) occupied by an undetermined object.

Claims

1. A method for classifying an occupant including the steps of:

a) capturing an image of an occupant area in a vehicle;

b) dividing the image into a plurality of subimages of predetermined spatial regions;

c) generating a plurality of low-level descriptors from each of the plurality of subimages;

d) analyzing the low-level descriptors; and

e) classifying an occupant in the occupant area based upon step d)into one of a plurality of classifications, wherein the classifications include;

infant seat.

2. The method of claim 1 wherein said step d) further includes the step of analyzing the low-level descriptors based upon a set of training data.

3. The method of claim 2 further including the step of creating the set of training data by capturing a plurality of images of known occupant classifications of the occupant area.

4. The method of claim 1 wherein said steps d) and e) are performed using a neural network.

5. The method of claim 1 wherein said step d) is based upon system parameters including an orientation or a location from which the image is captured relative to the occupant area.

6. The method of claim 5 further including the step of:

f) altering the orientation or the location from which the image is captured and adjusting the system parameters.

7. The method of claim 6 wherein said step f) further includes the step of entering physical data representing a physical orientation and location of the occupant area.