System to determine distance to a lead vehicle

• US 20060212215A1
• Filed: 03/15/2005
• Published: 09/21/2006
• Est. Priority Date: 03/15/2005
• Status: Active Grant

• Current Assignee: Visteon Global Technologies Incorporated (Visteon Corporation)
• Original Assignee: Visteon Global Technologies Incorporated (Visteon Corporation)
• Inventors: Koulinitch, Anatoli S.
• Granted Patent: US 7,231,288 B2
• Time in Patent Office: Days
• US Class Current: 701/205
• CPC Class Codes: G01S 11/12 using electromagnetic waves other than radio waves
• CPC Class Codes: G06T 2207/20016 Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
• CPC Class Codes: G06T 2207/30261 Obstacle
• CPC Class Codes: G06T 7/564 from contours
• CPC Class Codes: G06V 2201/08 Detecting or categorising vehicles

Abstract

A system implemented in a vehicle to detect a lead vehicle includes a camera which captures an image of the rear of the lead vehicle and an image processing system that receives image data from the camera. The image processing system is implemented with an algorithm that includes a preprocessing layer, an image segmentation layer, and a detection layer to analyze the image data to determine the relative distance between the vehicle and the lead vehicle.

Claims

1. A system implemented in a vehicle to detect a lead vehicle comprising:

a camera which captures an image of the rear of the lead vehicle; and

an image processing system that receives image data from the camera, the image processing system being implemented with an algorithm that includes a preprocessing layer, an image segmentation layer, and a detection layer to analyze the image data to determine the relative distance between the vehicle and the lead vehicle.

2. The system of claim 1 wherein the preprocessing layer reduces noise in the image data and generates vertical and horizontal edge maps.

3. The system of claim 2 wherein the preprocessing layer includes a Gausses filter and a sharpening filter.

4. The system of claim 2 wherein the preprocessing layer includes a Sobel filter.

5. The system of claim 1 wherein the detection layer includes a coarse level and a refine level.

6. The system of claim 5 wherein the coarse level is a hypothesis generation level that uses information of vertical and horizontal edges to hypothesize the location of the lead vehicle.

7. The system of claim 6 wherein the refining level is a hypothesis verification level that determines the true existence of the lead vehicle to verify the distance to the lead vehicle.

8. The system of claim 7 wherein the true existence of the lead vehicle is tested by data consistency analysis using different dependencies between vehicle image allocation and vehicle dimensions.

9. A method to detect a lead vehicle in front of a vehicle comprising:

capturing an image of the rear of the lead vehicle with a camera; and

processing image data from the camera with an algorithm that includes a preprocessing layer, an image segmentation layer, and a detection layer to determine the relative distance between the vehicle and the lead vehicle.

10. The method of claim 9 wherein the preprocessing layer reduces noise in the image data and generates vertical and horizontal edge maps.

11. The method of claim 10 wherein the preprocessing layer includes a Gausses filter and a sharpening filter.

12. The method of claim 10 wherein the preprocessing layer includes a Sobel filter.

13. The method of claim 9 wherein the detection layer includes a coarse level and a refine level.

14. The method of claim 13 wherein the coarse level is a hypothesis generation level that uses information of vertical and horizontal edges to hypothesize the location of the lead vehicle.

15. The method of claim 13 wherein the refining level is a hypothesis verification level that determines the true existence of the lead vehicle to verify the distance to the lead vehicle.

16. The method of claim 15 wherein the true existence of the lead vehicle is tested by data consistency analysis using different dependencies between vehicle image allocation and vehicle dimensions.