CLEAR PATH DETECTION USING AN EXAMPLE-BASED APPROACH

• US 20100097458A1
• Filed: 10/19/2009
• Published: 04/22/2010
• Est. Priority Date: 04/24/2008
• Status: Active Grant

• Current Assignee: Carnegie Mellon University, GM Global Technology Operations LLC (General Motors Company)
• Original Assignee: Carnegie Mellon University, GM Global Technology Operations Incorporated (General Motors Company)
• Inventors: Wu, Qi, Chen, Tsuhan, Zhang, Wende
• Granted Patent: US 8,803,966 B2
• Time in Patent Office: Days
• US Class Current: 348/119
• CPC Class Codes: B60W 2420/403 Image sensing, e.g. optical camera
• CPC Class Codes: G06T 7/74 involving reference images or patches
• 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 20/588 Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road

Abstract

A method for detecting a clear path of travel for a vehicle using a current image generated by a camera includes defining an exemplary clear path for each of a plurality of sample images, identifying features within each of the plurality of sample images, monitoring the current image generated by the camera, identifying features within the current image, matching the current image to at least one of the sample images based upon the identified features within the current image and the identified features within the plurality of sample images, determining a clear path of travel based upon the matching and the exemplary clear path for each of the matched sample images, and utilizing the clear path of travel to navigate the vehicle.

Claims

1. A method for detecting a clear path of travel for a vehicle using a current image generated by a camera, the method comprising:

defining an exemplary clear path for each of a plurality of sample images;

identifying features within each of the plurality of sample images;

monitoring the current image generated by the camera;

identifying features within the current image;

matching the current image to at least one of the sample images based upon the identified features within the current image and the identified features within the plurality of sample images;

determining a clear path of travel based upon the matching and the exemplary clear path for each of the matched sample images; and

utilizing the clear path of travel to navigate the vehicle.

2. The method of claim 1, wherein identifying features within each of the plurality of sample images is executed offline.

3. The method of claim 1, wherein identifying features within the current image is executed in real-time.

4. The method of claim 1, wherein each of the plurality of sample images comprises different combination of geometric layouts, lighting conditions, camera perspectives, traffic conditions, and camera lens radial distortion.

5. The method of claim 1, wherein identifying features within each of the plurality of sample images and wherein identifying features within the current image include using a filter bank.

6. The method of claim 5, wherein the filter bank comprises one of a Leung-Malik filter bank and a Gabor filter bank.

7. The method of claim 1, further comprising:

determining a feature value for each of the plurality of sample images based upon the identified features within each of the sample images; and

determining a feature value for the current image based upon the identified features within the current image;

wherein matching the current image to at least one of the sample images based upon the identified features within the current image and the identified features within the plurality of sample images comprises

comparing the feature value for the current image to the feature value for each of the plurality of sample images, and

selecting a predetermined number of matched sample images having feature values in nearest spatial proximity to the feature value of the current image.

8. The method of claim 7, wherein spatial proximity between the feature value of a sample image and the feature value of the current image is given by:

9. The method of claim 1, wherein determining the clear path of travel based upon the matching and the exemplary clear path for each of the matched sample images comprises:

inferring the clear path of travel based upon the defined exemplary clear path for each of the matched sample images.

10. The method of claim 9, wherein inferring the clear path of travel based upon the defined exemplary clear path for each of the matched sample images comprises:

fusing each clear path for the matched sample images into a single clear path;

associating image area locations on the current image with the single clear path; and

inferring the clear path of travel using the associated image area locations.

11. A method for detecting a clear path of travel for a vehicle using a current image area generated by a camera, the method comprising:

defining an exemplary clear path for each of a plurality of sample image areas;

identifying features within each of the plurality of sample image areas;

monitoring the current image area generated by the camera;

identifying features within the current image area;

matching the current image area to at least one of the sample image areas based upon the identified features within the current image area and the identified features within the plurality of sample image areas;

determining a clear path of travel based upon the matching and the exemplary clear path for each of the matched sample image areas; and

utilizing the clear path of travel to navigate the vehicle.

12. The method of claim 11, wherein identifying features within each of the plurality of sample image areas is executed offline.

13. The method of claim 11, wherein identifying features within each of the plurality of sample image areas and wherein identifying features within the current image area include using a filter bank.

14. The method of claim 11, further comprising:

determining a feature value for each of the plurality of sample image areas based upon the identified features within each of the sample image areas; and

determining a feature value for the current image area based upon the identified features within the current image area;

wherein matching the current image area to at least one of the sample image areas based upon the identified features within the current image area and the identified features within the plurality of sample image areas comprises

comparing the feature value for the current image area to the feature value for each of the plurality of sample image areas, and

selecting a predetermined number of matched sample image areas having feature values in nearest spatial proximity to the feature value of the current image area.

15. The method of claim 11, wherein determining the clear path of travel based upon the matching and the exemplary clear path for each of the matched sample image areas comprises:

determining the clear path of travel based upon the defined exemplary clear path for a closest match of the matched sample image areas.

16. The method of claim 11, wherein determining the clear path of travel based upon the matching and the exemplary clear path for each of the matched sample image areas comprises:

inferring the clear path of travel based upon the defined exemplary clear path for each of the matched sample image areas.

17. The method of claim 16 wherein inferring the clear path of travel based upon the defined exemplary clear path for each of the matched sample image areas comprises:

fusing each clear path for the matched sample image areas into a single clear path;

associating image area locations on the current image with the single clear path; and

inferring the clear path of travel using the associated image area locations.

18. A method for detecting a clear path of travel for a vehicle using a current image generated by a camera, the method comprising:

defining an exemplary vanishing point for each of a plurality of sample images;

identifying features within each of the plurality of sample images;

monitoring the current image generated by the camera;

identifying features within the current image;

matching the current image to at least one of the sample images based upon the identified features within the current image and the identified features within the plurality of sample images;

determining a vanishing point for the current image based upon the matching and the exemplary vanishing point for each of the matched sample images;

utilizing the vanishing point for the current image to define a clear path of travel; and

utilizing the clear path of travel to navigate the vehicle.

19. The method of claim 18, wherein determining the vanishing point for the current image based upon the matching and the exemplary vanishing point for each of the matched sample images comprises averaging the locations of the vanishing points for each of the matched sample images.