Homography-based passive vehicle speed measuring

• US 8,238,610 B2
• Filed: 12/18/2009
• Issued: 08/07/2012
• Est. Priority Date: 12/18/2008
• Status: Active Grant

• Current Assignee: University of Central Florida Research Foundation Inc. (State University System of Florida)
• Original Assignee: University of Central Florida Research Foundation Inc. (State University System of Florida)
• Inventors: Liyanage, Janaka Pradeep, Shah, Mubarak Ali
• Primary Examiner(s): Wang, Claire X
• Application Number:
  US12/642,719
• Publication Number:
  US 20100158321A1
• Time in Patent Office: 963 Days
• Field of Search: 382/104, 382/106, 382/107, 382/159, 348/149, 340/436, 340/933, 340/937, 700/1, 700/117
• US Class Current: 382/104
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/20081 Training; Learning
• CPC Class Codes: G06T 2207/30236 Traffic on road, railway or crossing
• CPC Class Codes: G06T 7/194 involving foreground-background segmentation
• CPC Class Codes: G06T 7/73 using feature-based methods

Abstract

An apparatus for passively measuring vehicle speed includes at least one video camera for acquiring images of a roadway upon which at least one moving vehicle travels upon, each of the images comprising a plurality of pixels. A computer processes pixel data associated with the plurality of pixels, including using an adaptive background subtraction model to perform background subtraction on the pixel data to identify a plurality of foreground pixels, extracting a plurality of blobs from the foreground pixels, and rectifying the blobs to form a plurality of rectified blobs using a homography matrix. The homography matrix is obtained by comparing at least one known distance in the roadway with distances between the pixels. Using a planar homography transform, the moving vehicle is identified from the plurality of rectified blobs, wherein the respective ones of the plurality of rectified blobs include vehicle data associated with the moving vehicle. The speed of the moving vehicle is computed from the vehicle data.

Claims

1. An apparatus for passively measuring vehicle speed, comprising:

at least one video camera for acquiring a plurality of images of a roadway upon which at least one moving vehicle travels upon, each of said plurality of images comprising a plurality of pixels;

a computer for processing pixel data associated with said plurality of pixels, said computer having code operable for processing said pixel data, wherein said processing said pixel data comprises;

performing background subtraction using an adaptive background subtraction model on said pixel data to identify a plurality of foreground pixels and to extract a plurality of blobs from said plurality of foreground pixels;

rectifying said plurality of blobs to form a plurality of rectified blobs using a homography matrix, said homography matrix obtained by comparing at least one known distance in said roadway with distances between said plurality of pixels and using a planar homography transform,

identifying said moving vehicle from said plurality of rectified blobs, respective ones of said plurality of rectified blobs including vehicle data associated with said moving vehicle, and

computing a speed of said moving vehicle from said vehicle data.

2. The apparatus of claim 1, wherein said code for processing said pixel data is further operable for data filtering said vehicle data to generate filtered vehicle data, wherein said computing said speed of said moving vehicle utilizes said filtered vehicle data.

3. The apparatus of claim 2, wherein said data filtering comprises RANdom Sample Consensus (RANSAC).

4. The apparatus of claim 2, wherein said data filtering discards ones of said plurality of rectified blobs that do not appear consistently in a predetermined number of consecutive frames.

5. The apparatus of claim 1, wherein said adaptive background subtraction model comprises a Gaussian Mixture Model (GMM).

6. The apparatus of claim 1, wherein said code for processing said pixel data is further operable for determining a geometric length of said moving vehicle.

7. The apparatus of claim 1, wherein said video camera is positioned above said roadway.

8. The apparatus of claim 1, wherein said video camera is positioned alongside said roadway.

9. A method for passively measuring vehicle speed, comprising:

acquiring a plurality of images of a roadway upon which at least one moving vehicle travels upon, each of said images comprising a plurality of pixels;

processing pixel data associated with said plurality of pixels, said computer having code operable for processing said pixel data, said processing comprising;

performing background subtraction using an adaptive background subtraction model on said pixel data to identify a plurality of foreground pixels and to extract a plurality of blobs from said plurality of foreground pixels;

rectifying said plurality of blobs to form a plurality of rectified blobs using a homography matrix, said homography matrix obtained by comparing at least one known distance in said roadway with distances between said plurality of pixels and using a planar homography transform,

identifying at least said moving vehicle from said plurality of rectified blobs, respective ones of said plurality of rectified blobs including vehicle data associated with said moving vehicle, and

computing a speed of said moving vehicle from said vehicle data.

10. The method of claim 9, further comprising data filtering said vehicle data to generate filtered vehicle data, wherein said computing said speed of said moving vehicle utilizes said filtered vehicle data.

11. The method of claim 10, wherein said data filtering comprises RANdom Sample Consensus (RANSAC).

12. The method of claim 10, wherein said data filtering discards ones of said plurality of rectified blobs that do not appear consistently in a predetermined number of consecutive frames.

13. The method of claim 9, wherein said adaptive background subtraction model comprises a Gaussian Mixture Model (GMM).

14. The method of claim 9, wherein said processing further comprises determining a geometric length of said moving vehicle.

15. The method of claim 9, wherein said video camera is positioned above said roadway.

16. The method of claim 9, wherein said video camera is positioned alongside said roadway.