METHOD FOR CALIBRATING VEHICULAR VISION SYSTEM

• US 20140063254A1
• Filed: 11/11/2013
• Published: 03/06/2014
• Est. Priority Date: 03/07/2007
• Status: Active Grant

• Current Assignee: Magna Electronics Incorporated (Magna International Incorporated)
• Original Assignee: Magna International Incorporated
• Inventors: Shi, Bin, Rathi, Ghanshyam, Zibman, Sharon, Perovic, Tom, Gupta, Nikhil, Faraji, Hilda, Zhou, Yong
• Granted Patent: US 9,077,962 B2
• Time in Patent Office: Days
• US Class Current: 348/148
• CPC Class Codes: B60R 21/01538 for image processing, e.g. cameras or sensor arrays
• CPC Class Codes: G06F 18/211 Selection of the most significant subset of features
• CPC Class Codes: G06T 7/80 Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
• CPC Class Codes: G06V 10/50 by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
• CPC Class Codes: G06V 10/52 Scale-space analysis, e.g. wavelet analysis multi-scale boundary representations G06V10/42
• CPC Class Codes: G06V 10/766 using regression, e.g. by projecting features on hyperplanes
• CPC Class Codes: G06V 20/59 inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions
• CPC Class Codes: H04N 17/002 for television cameras

Abstract

A method for calibrating a vehicular vision system includes providing a camera at a vehicle, with the camera having a field of view. Images are captured with the camera and a set of resultant images are acquired for a classification. Information is extracted related to image features in the set of resultant images, and an appropriate subset of coefficients is determined. For each classification, a classification vector of at least one appropriate weight is stored that corresponds to the determined subset of coefficients.

Claims

1. A method for calibrating a vehicular vision system, said method comprising:

providing a camera at a vehicle, said camera having a field of view;

capturing images with said camera;

acquiring a set of resultant images for a classification;

extracting information related to image features in the set of resultant images;

determining an appropriate subset of coefficients; and

for each classification, storing a classification vector of at least one appropriate weight that corresponds to the determined subset of coefficients.

2. The method of claim 1, wherein said camera has a field of view interior of the vehicle.

3. The method of claim 1, wherein the set of resultant images comprises images of a region of interest within the field of view of said camera and wherein the region of interest corresponds to the respective classification for that set of resultant images.

4. The method of claim 1, comprising providing occlusion detection to detect an unacceptable occlusion at least partially occluding the field of view of said camera.

5. The method of claim 1, wherein the acquired resultant images comprise images of at least two regions of interest within the field of view of said camera.

6. The method of claim 5, wherein extracting image features in the set of resultant images is performed on the at least two regions of interest.

7. The method of claim 6, wherein images of the set of resultant images are acquired by obtaining a first image illuminated by ambient light and a second image illuminated by ambient light and a second frequency range of light and subtracting the first image from the second image to obtain the resultant images.

8. The method of claim 1, wherein the determined subset of coefficients is determined by processing sets of coefficients produced from a selection of calibration images and determining a subset of coefficients which acceptably discriminate between defined classifications.

9. The method of claim 1, wherein acquiring a set of resultant images comprises limiting the dynamic range of acquired images to obtain resultant images that comprise at least one region of interest, which encompasses a region of the field of view that is less than the field of view of said camera.

10. A method for calibrating a vehicular vision system, said method comprising:

providing a camera at a vehicle, said camera having a field of view;

capturing images with said camera;

acquiring a set of resultant images for a classification;

wherein the acquired resultant images comprise images of at least two regions of interest within the field of view of said camera;

extracting information related to image features in the set of resultant images;

determining an appropriate subset of coefficients;

for each classification, storing a classification vector of at least one appropriate weight that corresponds to the determined subset of coefficients; and

providing occlusion detection to detect an unacceptable occlusion at least partially occluding the field of view of said camera.

11. The method of claim 10, wherein said camera has a field of view interior of the vehicle.

12. The method of claim 10, wherein extracting image features in the set of resultant images is performed on the at least two regions of interest.

13. The method of claim 12, wherein images of the set of resultant images are acquired by obtaining a first image illuminated by ambient light and a second image illuminated by ambient light and a second frequency range of light and subtracting the first image from the second image to obtain the resultant images.

14. The method of claim 10, wherein the determined subset of coefficients is determined by processing sets of coefficients produced from a selection of calibration images and determining a subset of coefficients which acceptably discriminate between defined classifications.

15. The method of claim 10, wherein acquiring a set of resultant images comprises limiting the dynamic range of acquired images to obtain resultant images that comprise at least one region of interest, which encompasses a region of the field of view that is less than the field of view of said camera.

16. A method for calibrating a vehicular vision system, said method comprising:

providing a camera at a vehicle, said camera having a field of view;

capturing images with said camera;

acquiring a set of resultant images for a classification;

wherein the set of resultant images comprises images of a region of interest within the field of view of said camera and wherein the region of interest corresponds to the respective classification for that set of resultant images;

extracting information related to image features in the set of resultant images;

determining an appropriate subset of coefficients;

for each classification, storing a classification vector of at least one appropriate weight that corresponds to the determined subset of coefficients; and

providing occlusion detection to detect an unacceptable occlusion at least partially occluding the field of view of said camera.

17. The method of claim 16, wherein said camera has a field of view interior of the vehicle.

18. The method of claim 16, wherein images of the set of resultant images are acquired by obtaining a first image illuminated by ambient light and a second image illuminated by ambient light and a second frequency range of light and subtracting the first image from the second image to obtain the resultant images.

19. The method of claim 16, wherein the determined subset of coefficients is determined by processing sets of coefficients produced from a selection of calibration images and determining a subset of coefficients which acceptably discriminate between defined classifications.

20. The method of claim 16, wherein acquiring a set of resultant images comprises limiting the dynamic range of acquired images to obtain resultant images that comprise at least one region of interest, which encompasses a region of the field of view that is less than the field of view of said camera.