Method of operation for a vision-based occupant sensing system

• US 7,791,670 B2
• Filed: 05/11/2005
• Issued: 09/07/2010
• Est. Priority Date: 05/11/2005
• Status: Active Grant

• Current Assignee: Aptiv Technologies AG
• Original Assignee: Delphi Technologies, Inc. (BorgWarner Incorporated)
• Inventors: Meier, Michael R, Noland, George J., Constable, Robert K, Voreis, Thomas L., Fultz, William W., Taylor, Ronald M.
• Primary Examiner(s): Villecco; John M
• Application Number:
  US11/126,606
• Publication Number:
  US 20060256197A1
• Time in Patent Office: 1,945 Days
• Field of Search: 348/40, 348/335, 348/366, 348/148, 348/362, 348/363, 348/364
• US Class Current: 348/366
• CPC Class Codes: B60R 21/01538 for image processing, e.g. cameras or sensor arrays
• CPC Class Codes: G06V 10/141 Control of illumination
• CPC Class Codes: G06V 40/103 Static body considered as a whole, e.g. static pedestrian or occupant recognition
• CPC Class Codes: H04N 23/70 Circuitry for compensating brightness variation in the scene
• CPC Class Codes: H04N 23/71 Circuitry for evaluating the brightness variation
• CPC Class Codes: H04N 23/72 Combination of two or more compensation controls
• CPC Class Codes: H04N 23/74 by influencing the scene brightness using illuminating means

Abstract

Image processing parameters of an imaging chip in a vision-based occupant sensing system are adjusted for each frame based on ambient illumination responsive information obtained in the idle period preceding that frame. The same sensing data is also used to determine if active illumination is needed to supplement existent ambient illumination during the ensuing image acquisition interval. The inter-frame ambient illumination is detected with an external light sensor or with selected pixels of the imaging chip, and the information is used to calculate and set the gain and integration time of the imaging chip. In applications where the resolution of the imaging chip significantly exceeds the resolution required for occupant sensing, intensity data from one or more otherwise inactive pixels is averaged with the intensity data from the normally active pixels to adapt the sensitivity of the imaging chip to changing ambient illumination.

Claims

1. A method of operation for vision-based occupant sensing in a vehicle cabin with a solid-state imaging chip, comprising the steps of:

producing image data with said imaging chip during periodic image capture intervals that are separated by idle intervals;

sensing ambient illumination in said cabin during said idle intervals;

setting control parameters of said imaging chip for an upcoming image capture interval based on ambient illumination sensed in an idle period just prior to such upcoming image capture interval;

normally acquiring image data from a set of pixels that are distributed over an active area of said imaging chip, each such pixel being part of a cluster of pixels that also includes normally unused pixels; and

based on the sensed ambient illumination, selectively acquiring image data from one or more of the normally unused pixels in said pixel clusters so as to increase sensitivity but not resolution of said imaging chip in said upcoming image capture interval.

2. The method of claim 1, where said control parameters include a gain and an integration time of said imaging chip.

3. The method of claim 1, including the step of actively illuminating said cabin with an active illumination source during said upcoming image capture interval if the ambient illumination sensed in the idle period just prior to such upcoming image capture interval is less than a threshold level.

4. The method of claim 3, where the ambient illumination sensed in the idle period just prior to such upcoming image capture interval is used to control an amount of active illumination during said upcoming image capture interval.

5. The method of claim 1, including the step of:

sensing said ambient illumination in said cabin with an external light sensor.

6. The method of claim 1, including the step of:

acquiring image data from selected pixels of said imaging chip during said idle intervals, and sensing said ambient illumination in said cabin based on such acquired image data.

7. The method of claim 1, including the step of:

in each cluster of pixels, averaging the selectively acquired image data with the normally acquired image data.

8. A method of operation for vision-based occupant sensing in a vehicle cabin with a solid-state imaging chip, comprising the steps of:

producing image data with said imaging chip during periodic image capture intervals that are separated by idle intervals;

sensing ambient illumination in said cabin during said idle intervals;

setting control parameters of said imaging chip for an upcoming image capture interval based on ambient illumination sensed in an idle period just prior to such upcoming image capture interval;

normally acquiring image data from a set of pixels that are distributed over an active area of said imaging chip, each such pixel being part of a cluster of pixels that also includes normally unused pixels;

identifying regions of an image produced from the acquired image data that are persistently underexposed; and

selectively acquiring image data from one or more of the normally unused pixels in pixel clusters corresponding to the identified regions of said image so as to locally increase sensitivity but not resolution of said imaging chip in said upcoming image capture interval.

9. The method of claim 8, including the step of:

in each cluster of pixels corresponding to the identified regions of said image, averaging the selectively acquired image data with the normally acquired image data.