Object detection and ranging method

• US 8,391,557 B2
• Filed: 09/12/2008
• Issued: 03/05/2013
• Est. Priority Date: 09/14/2007
• Status: Active Grant

• Current Assignee: Magna International Incorporated
• Original Assignee: Magna International Incorporated
• Inventors: Burtch, Matthew T.
• Primary Examiner(s): Fitzpatrick, Atiba O
• Application Number:
  US12/677,586
• Publication Number:
  US 20100260382A1
• Time in Patent Office: 1,635 Days
• Field of Search: 382100-107, 340907-915, 340933-940, 348113-115, 701/1, 701/23, 701/301
• US Class Current: 382/104
• CPC Class Codes: B60T 7/22 initiated by contact of vehicle, e.g. bumper, with an external object, e.g. another vehicle , or by means of contactless obstacle detectors mounted on the vehicle
• CPC Class Codes: B60W 2420/403 Image sensing, e.g. optical camera
• CPC Class Codes: B60W 30/06 Automatic manoeuvring for parking
• CPC Class Codes: B60W 30/095 Predicting travel path or likelihood of collision
• CPC Class Codes: B60W 40/02 related to ambient conditions
• CPC Class Codes: G06T 2207/10028 Range image; Depth image; 3D point clouds
• CPC Class Codes: G06T 2207/20021 Dividing image into blocks, subimages or windows
• CPC Class Codes: G06T 2207/30264 Parking
• CPC Class Codes: G06T 7/11 Region-based segmentation

Abstract

The present invention relates to a method for object detection and ranging within a vehicles rearward field of interest which includes an algorithm to translate images provided by an imaging device. An image device provides images to a processor which divides the images into groups of rows of pixels. The rows are processed by the algorithm which includes assigning each pixel in the rows to an object. The translation of the image from outside of the vehicle is provided to the vehicle operator and includes the tracking of location and dimensions of multiple objects within the viewing range.

Claims

1. An object detection and ranging method for a motor vehicle, comprising the steps of:

providing an imaging device;

providing a data processor for interpreting a captured image of a field of interest provided by the imaging device;

processing rows of pixels;

comparing pixel data collected by said data processor to a pixel map of one or more surfaces to filter out and discard said surfaces; and

reporting a real world translation to a vehicle operator, wherein said real world translation includes at least object dimensions and distance from the vehicle for all objects in said field of interest;

wherein processing said rows further comprises a morphological erosion filter for filtering out and discarding pixel noise.

2. The object detection and ranging method for a motor vehicle of claim 1, wherein said imaging device is a three dimensional imaging camera.

3. The object detection and ranging method for a motor vehicle of claim 1, further comprising the step of said data processor collecting a frame of said captured image and breaking data from said collected frame into groups of rows of pixels for processing said rows of pixels to operably assign each respective pixel in a row to one or more objects in said field of interest.

4. The object detection and ranging method for a motor vehicle of claim 1, wherein said data processor processes rows of pixels using first and second processing steps that are point based operations to filter out pixel data.

5. The object detection and ranging method for a motor vehicle of claim 1, wherein said data processor processes rows of pixels using third and fourth processing steps that are spatial operations performed on each respective pixel of interest in said row to assign said pixel of interest to an object in said field of interest.

6. The object detection and ranging method for a motor vehicle of claim 1, further comprising the step of selectively merging objects determined during one or more sweeps of said field of interest by said imaging device to be in such proximity with one another as to be part of the same object.

7. The object detection and ranging method for a motor vehicle of claim 1, wherein said reported real world translation is selected from the group consisting of object location, geometric centroid, velocity vector, dimensions, size, distance from the vehicle, and combinations thereof.

8. An object detection and ranging method for a motor vehicle, comprising the steps of:

providing an imaging device to selectively capture an image of a field of interest;

providing a data processor to operably collect a frame of said captured image;

breaking data from said collected frame into groups of rows of pixels;

processing said rows of pixels to operably assign each respective pixel in a row to one or more objects in the field of interest until every row of pixels is processed;

generating a real world translation of said one or more objects; and

reporting said real world translation to a vehicle operator, wherein said real world translation includes at least object dimensions and distance from the vehicle for all objects in said field of interest;

wherein processing said rows further comprises a morphological erosion filter for filtering out and discarding pixel noise.

9. The object detection and ranging method for a motor vehicle of claim 8, wherein said imaging device is a three dimensional imaging camera.

10. The object detection and ranging method for a motor vehicle of claim 8, wherein processing said rows further comprises first and second processing steps that are point based operations to filter pixel data.

11. The object detection and ranging method for a motor vehicle of claim 8, wherein processing said rows further comprises first and second processing steps, wherein said second processing step compares pixel data collected by said data processor to a pixel map of one or more surfaces to filter out and discard said surfaces.

12. The object detection and ranging method for a motor vehicle of claim 8, wherein processing said rows further comprises third and fourth processing steps that are spatial operations performed on each respective pixel of interest in said row to assign said pixel of interest to an object in said field of interest.

13. The object detection and ranging method for a motor vehicle of claim 8, further comprising the step of selectively merging objects determined during one or more sweeps of said field of interest by said imaging device to be in such proximity with one another as to be part of the same object.

14. The object detection and ranging method for a motor vehicle of claim 8, wherein said reported real world translation is selected from the group consisting of object location, geometric centroid, velocity vector, dimensions, size, distance from the vehicle, and combinations thereof.

18. The object detection and ranging method for a motor vehicle of claim 13, further comprising selectively merging objects first determined to be separate objects during one sweep of said field of interest by said imaging device into a single object determined in one or more subsequent sweeps to be in such proximity with one another as to be part of the same object.

15. An object detection and ranging method for a motor vehicle, comprising the steps of:

providing an imaging device to selectively capture an image of a field of interest;

providing a data processor to operably collect a frame of said captured image;

breaking data from said collected frame into groups of rows of pixels;

processing said rows of pixels to operably assign each respective pixel in a row to one or more objects in the field of interest until every row of pixels is processed, wherein said processing said rows includes comparing pixel data collected by said data processor to a pixel map of one or more surfaces to filter out and discard said surfaces;

merging objects determined to be in such proximity with one another as to be part of the same object

generating a real world translation of said one or more objects; and

reporting said real world translation to a vehicle operator, wherein said real world translation includes at least object dimensions and distance from the vehicle for all objects in said field of interest,

wherein processing said rows further comprises a morphological erosion filter for filtering out and discarding pixel noise.

16. The object detection and ranging method for a motor vehicle of claim 15, wherein said imaging device is a three dimensional imaging camera.

17. The object detection and ranging method for a motor vehicle of claim 15, wherein processing said rows further comprises third and fourth processing steps that are spatial operations performed on each respective pixel of interest in said row to assign said pixel of interest to an object in said field of interest.

19. The object detection and ranging method for a motor vehicle of claim 15, wherein said reported real world translation is selected from the group consisting of object location, geometric centroid, velocity vector, dimensions, size, distance from the vehicle, and combinations thereof.