VEHICLE VISION SYSTEM WITH PEDESTRIAN DETECTION

• US 20160180158A1
• Filed: 12/15/2015
• Published: 06/23/2016
• Est. Priority Date: 12/18/2014
• Status: Abandoned Application

• Current Assignee: Magna Electronics Incorporated (Magna International Incorporated)
• Original Assignee: Magna Electronics Incorporated (Magna International Incorporated)
• Inventors: Phan, Xai, Gupta, Nikhil
• Application Number:
  US14/969,528
• Publication Number:
  US 20160180158A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06T 5/80 Geometric correction
• CPC Class Codes: G06V 20/58 Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
• CPC Class Codes: G06V 40/10 Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands

Abstract

A vision system of a vehicle includes a camera configured to be disposed at a vehicle so as to have a field of view exterior of the vehicle. An image processor is operable to process image data captured by the camera. The image processor, responsive to image processing of captured image data when the camera is disposed at the vehicle, is operable to determine the presence of a pedestrian in the field of view of the camera. The vision system may utilize a linear classifier to identify areas where a pedestrian is more likely to be present for enhanced processing of those areas. The vision system may utilize a Histogram Intersection Kernel classifier for enhanced processing of identified areas. The vision system may utilize detection area filtering to reduce processing of image data representative of areas of the field of view where a pedestrian is not likely to be present.

Claims

1. A vision system of a vehicle, said vision system comprising:

a camera configured to be disposed at a vehicle so as to have a field of view exterior of the vehicle;

wherein said camera comprises a pixelated imaging array having a plurality of photosensing elements;

an image processor operable to process image data captured by said camera;

wherein said image processor, responsive to image processing of captured image data and with said camera disposed at the vehicle, is operable to determine the presence of a pedestrian in the field of view of said camera; and

wherein said vision system utilizes a linear classifier to identify areas for further processing of areas where a pedestrian is likely to be present, and wherein said image processor enhances processing of captured image data representative of the identified areas.

2. The vision system of claim 1, wherein said vision system utilizes a Histogram Intersection Kernel classifier to enhance processing of the identified areas.

3. The vision system of claim 1, wherein said vision system utilizes detection area filtering to reduce processing of image data representative of areas of the field of view where a pedestrian is not likely to be present.

4. The vision system of claim 1, wherein said camera comprises a fish-eye lens and wherein said vision system processes captured image data to correct for lens distortion.

5. The vision system of claim 4, wherein said vision system utilizes a Kalman filter to process the corrected image data to determine the presence of a pedestrian in the field of view of said camera.

6. The vision system of claim 1, wherein said vision system utilizes at least one of (i) a distance estimation algorithm and (ii) a trajectory estimation algorithm.

7. The vision system of claim 1, wherein, when processing captured frames of image data of various sizes, said vision system processes image data of smaller frames before processing image data of larger frames.

8. The vision system of claim 1, wherein said image processor processes captured image data from a bottom of the frame to a top of the frame to determine a pedestrian that is closest to the vehicle first.

9. The vision system of claim 8, wherein, responsive to detection of a pedestrian, said system does not process image data above the detected pedestrian.

10. A vision system of a vehicle, said vision system comprising:

a camera configured to be disposed at a vehicle so as to have a field of view exterior of the vehicle;

wherein said camera comprises a pixelated imaging array having a plurality of photosensing elements;

an image processor operable to process image data captured by said camera;

wherein said image processor, responsive to image processing of captured image data and with said camera disposed at the vehicle, is operable to determine the presence of a pedestrian in the field of view of said camera; and

wherein said vision system utilizes a Histogram Intersection Kernel classifier to enhance processing of areas where a pedestrian is likely to be present.

11. The vision system of claim 10, wherein said vision system utilizes detection area filtering to reduce processing of image data representative of areas of the field of view where a pedestrian is not likely to be present.

12. The vision system of claim 10, wherein said camera comprises a fish-eye lens and wherein said vision system processes captured image data to correct for lens distortion.

13. The vision system of claim 12, wherein said vision system utilizes a Kalman filter to process the corrected image data to determine the presence of a pedestrian in the field of view of said camera.

14. The vision system of claim 13, wherein said vision system utilizes at least one of (i) a distance estimation algorithm and (ii) a trajectory estimation algorithm.

15. The vision system of claim 10, wherein said vision system utilizes at least one of (i) a distance estimation algorithm and (ii) a trajectory estimation algorithm.

16. A vision system of a vehicle, said vision system comprising:

a camera configured to be disposed at a vehicle so as to have a field of view exterior of the vehicle;

wherein said camera comprises a pixelated imaging array having a plurality of photosensing elements;

an image processor operable to process image data captured by said camera;

wherein said image processor, responsive to image processing of captured image data and with said camera disposed at the vehicle, is operable to determine the presence of a pedestrian in the field of view of said camera;

wherein said vision system utilizes a linear classifier to identify areas for further processing of areas where a pedestrian is likely to be present;

wherein said vision system utilizes a Histogram Intersection Kernel classifier to enhance processing of captured image data representative of the identified areas; and

wherein said vision system utilizes at least one of (i) a distance estimation algorithm and (ii) a trajectory estimation algorithm.

17. The vision system of claim 16, wherein said vision system utilizes detection area filtering to reduce processing of image data representative of areas of the field of view where a pedestrian is not likely to be present.

18. The vision system of claim 16, wherein said camera comprises a fish-eye lens and wherein said vision system processes captured image data to correct for lens distortion.

19. The vision system of claim 16, wherein said vision system utilizes a Kalman filter to process the corrected image data to determine the presence of a pedestrian in the field of view of said camera.

20. The vision system of claim 16, wherein said image processor processes captured image data from a bottom of the frame to a top of the frame to determine a pedestrian that is closest to the vehicle first, and wherein, responsive to detection of a pedestrian, said system does not process image data above the detected pedestrian.