Method and system for seatbelt detection using adaptive histogram normalization

Patent metadata

Field	Value
Publication number	US-11798296-B2
Application number	US-202117557370-A
Country	US
Kind code	B2
Filing date	Dec 21, 2021
Priority date	Dec 21, 2021
Publication date	Oct 24, 2023
Grant date	Oct 24, 2023

Abstract

A method for detecting seatbelt positioning comprises: capturing, by a camera, a source image including a plurality of pixels each having a corresponding brightness level; determining a range of brightness levels of a group of the plurality of pixels located within at least one region of the source image; generating an adjusted image by adjusting the corresponding brightness levels of the group of the pixels located within the region of the source image based on the range of the brightness levels of the group of the plurality of pixels located within the at least one region of the source image; converting the adjusted image to a black-and-white image; and scanning across the black-and-white image to detect a plurality of transitions between black and white segments corresponding to a predetermined pattern of the seatbelt, and using detections of the plurality of transitions to indicate a detection of the seatbelt.

First claim

The invention claimed is:

1. A method for detecting seatbelt positioning, comprising:
capturing, by a camera, a source image of an occupant, the source image including a plurality of pixels each having a corresponding brightness level;
determining a range of brightness levels of a group of the plurality of pixels located within at least one region of the source image;
generating an adjusted image by adjusting the corresponding brightness levels of the group of the plurality of pixels located within the at least one region of the source image based on the range of the brightness levels of the group of the plurality of pixels located within the at least one region of the source image;
converting the adjusted image to a black-and-white image; and
scanning across the black-and-white image to detect a plurality of transitions between black and white segments corresponding to a predetermined pattern of the seatbelt, and using detections of the plurality of transitions to indicate a detection of the seatbelt.

2. The method of claim 1 , wherein the at least one region includes a plurality of regions, and wherein the adjusted image includes the plurality of regions each adjusted independently.

3. The method of claim 1 , wherein capturing the source image of the occupant includes capturing the source image in near infrared (NIR).

4. The method of claim 1 , wherein the region of the source image has at least one dimension approximately equal to a width of the seatbelt in the source image.

5. The method of claim 1 , wherein converting the adjusted image to the black-and-white image includes comparing the brightness level of each of the pixels with a threshold value.

6. The method of claim 1 , wherein adjusting the brightness levels of the pixels within the region comprises multiplying the brightness level of each of the pixels by a scale factor, the scale factor based on the range of the brightness levels of the pixels within the region.

7. The method of claim 6 , wherein the scale factor is based on a difference between a low brightness value and a high brightness value, with each of the low brightness value and the high brightness value based on the brightness levels of the pixels within the region.

8. The method of claim 7 , wherein the low brightness value is greater than a lowest brightness level of the pixels within the region.

9. The method of claim 7 , wherein the high brightness value is lower than a highest brightness level of the pixels within the region.

10. The method of claim 1 , wherein adjusting the brightness levels of the pixels within the region includes offsetting the brightness level of each of the pixels based on a low brightness value, the low brightness value based on the brightness levels of the pixels within the region.

11. A system for detecting seatbelt positioning, comprising:
a seatbelt having a predetermined pattern;
a camera configured to capture a source image of an occupant wearing the seatbelt, the source image including a plurality of pixels each having a corresponding brightness level;
a processor in communication with the camera; and
a controller in communication with the camera and configured to:
determine a range of brightness levels of a group of the plurality of pixels located within at least one region of the source image;
generate an adjusted image by adjusting the corresponding brightness levels of the group of the plurality of pixels located within the at least one region of the source image based on the range of the brightness levels of the group of the plurality of pixels located within the at least one region of the source image;
convert the adjusted image to a black-and-white image; and
determine a position of the seatbelt based on detecting transitions in the black-and-white image corresponding to the predetermined pattern of the seatbelt.

12. The system of claim 11 , wherein the at least one region includes a plurality of regions, and wherein the adjusted image includes the plurality of regions each adjusted independently.

13. The system of claim 11 , wherein capturing the source image of the occupant includes capturing the source image in near infrared (NIR).

14. The system of claim 11 , wherein the region of the source image has at least one dimension approximately equal to a width of the seatbelt in the source image.

15. The system of claim 11 , wherein converting the adjusted image to the black-and-white image includes the controller comparing the brightness level of each of the pixels with a threshold value.

16. The system of claim 11 , wherein adjusting the brightness levels of the pixels within the region includes the controller multiplying the brightness level of each of the pixels by a scale factor, the scale factor based on the range of the brightness levels of the pixels within the region.

17. The system of claim 16 , wherein the scale factor is based on a difference between a low brightness value and a high brightness value, with each of the low brightness value and the high brightness value based on the brightness levels of the pixels within the region.

18. The system of claim 17 , wherein the low brightness value is greater than a lowest brightness level of the pixels within the region.

19. The system of claim 17 , wherein the high brightness value is lower than a highest brightness level of the pixels within the region.

20. The system of claim 11 , wherein adjusting the brightness levels of the pixels within the region includes the controller offsetting the brightness level of each of the pixels based on a low brightness value, the low brightness value based on the brightness levels of the pixels within the region.

Assignees

• Veoneer Us Inc
• Veoneer Us Llc

Inventors

• Pleune Mitchell
• Chung Caroline
• Naik Afrah

Classifications

• G06V20/593 Primary

  Recognising seat occupancy · CPC title

• G06T5/40

  using histogram techniques · CPC title

• G06T7/70

  Determining position or orientation of objects or cameras (camera calibration G06T7/80) · CPC title

• G06V10/60

  relating to illumination properties, e.g. using a reflectance or lighting model · CPC title

• G06V40/103

  Static body considered as a whole, e.g. static pedestrian or occupant recognition · CPC title