Electronic traffic monitor

• US 7,580,547 B2
• Filed: 10/24/2006
• Issued: 08/25/2009
• Est. Priority Date: 10/24/2006
• Status: Active Grant

• Current Assignee: Iteris Incorporated
• Original Assignee: Iteris Incorporated
• Inventors: Benhammou, Daniel Jacques
• Primary Examiner(s): Strege; John B
• Application Number:
  US11/552,242
• Publication Number:
  US 20080095403A1
• Time in Patent Office: 1,036 Days
• Field of Search: 382/103, 382/104, 382/107, 382/168, 382/194, 382/300, 348/143, 348148-149, 340/907
• US Class Current: 382/103
• CPC Class Codes: G06T 2207/30236 Traffic on road, railway or crossing
• CPC Class Codes: G06T 7/20 Analysis of motion motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51
• CPC Class Codes: G06T 7/80 Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
• CPC Class Codes: G06V 20/54 of traffic, e.g. cars on the road, trains or boats

Abstract

A method of calibrating a video image to the scene imaged is provided for which a vehicle image transit can be associated with the transit of the real-world vehicle. A path is determined by a tracking point for at least one vehicle image. A measuring vehicle image is selected, wherein the measuring vehicle image is associated with a known dimension of the real-world measuring vehicle. At a point on the path, collocated with the tracking point, a scaling factor is determined such that a measuring vehicle image dimension is associated with a real-world dimension. A number of scaling factors can then be assembled into a scaling map allowing vehicles traversing the video scene to be measured in time, by the video frame rate, and distance, by the distance of their images and scaled according to the scaling map. Other embodiments are also disclosed.

Claims

1. A method, comprising:

using a computer to carry out the steps of;

receiving a video image of a road surface imaged and of a vehicle moving across the road surface;

determining a tracking point of a vehicle image, of the vehicle, captured in at least two frames of the video image;

deriving a formula descriptive of a curve containing the tracking points;

determining the vehicle image is a measuring vehicle by;

obtaining a number of target vehicle images captured by the video image;

identifying a segment of the target vehicle images by a function, wherein the function identifies at least one measuring vehicle image from the target vehicle images; and

identifying the vehicle image as a measuring vehicle upon the vehicle image matching the at least one measuring vehicle image; and

upon determining the vehicle image is the measuring vehicle, deriving a scaling map comprising a scaling factor associating a vehicle image dimension, in the direction of travel, to a real-world vehicle dimension at a point collocated with the vehicle image in one frame of the video image.

2. The method of claim 1, further comprising,

segmenting the curve into segments representing roadway segments of a known length with distances determined in accord with the scaling map;

and

determining the progression of a target vehicle by determining the progression of the target vehicle'"'"'s image relative to at least one segment.

3. The method of claim 1, wherein function identifies the mean vehicle image size of the number of target vehicle images.

4. The method of claim 1, wherein the tracking point is a centroid.

5. The method of claim 1, further comprising:

receiving a number of additional vehicle images;

determining tracking points for the number of additional vehicle images, each of the number of additional vehicle images contributing a tracking point for at least two frames of the video image; and

determined a number of curves for each path of the number of additional vehicle images.

6. The method of claim 1, further comprising, measuring the transit of a subject vehicle by measuring the transit of the subject vehicle in a number of frames of the video image in accord with distances defined by the scaling map.

7. A system for gathering traffic information, comprising:

a video receiver to receive a video image of a road surface and vehicles moving across the road surface;

a vehicle recognition process to identify a vehicle image in the video image and calculate a tracking point for the vehicle image;

a vehicle path calculator to derive a formula descriptive of a path of the tracking point through a number of video image frames;

a vehicle selection processor for selecting a measuring vehicle by matching a candidate vehicle image to a measurement vehicle image criteria determined by a function, wherein the measurement vehicle image criteria is associated with a known dimension of a measurement vehicle;

a scale calculator to define a scaling map comprising scaling factors for at least one point on the path and equating a dimension of the measuring vehicle image to a real-world dimension of the measuring vehicle for the one point; and

a vehicle measurement process to determine the progression of a target vehicle from the progression of a target vehicle image along the road surface image in accord with distances defined by the scaling map.

8. The system of claim 7, wherein:

the vehicle path calculator further derives a number of segments representing known lengths of road surface, the lengths being determined in accord with distances defined by the scaling map; and

the vehicle measurement process determining the progression of the target vehicle relative to at least one segment.

9. The system of claim 7, wherein the function determines the mean vehicle image size of a number of additional vehicle images.

10. The system of claim 7, wherein the dimension of the measuring vehicle image is measured orthogonal to the direction of travel.

11. The system of claim 7, wherein determining the progression of a target vehicle includes determining time of a portion of the progression from the frame rate of the video image.

12. The system of claim 7, further comprising an alert processor to create a notification if the progression of a defined number of vehicle images are statistical outliers.

13. A computer-readable medium having computer-executable instructions for performing the steps comprising:

receiving a video image of a road surface imaged and of a vehicle moving across the road surface;

determining a tracking point of a vehicle image, of the vehicle, captured in at least two frames of the video image;

deriving a formula descriptive of a curve containing the tracking points;

determining the vehicle image is a measuring vehicle by;

obtaining a number of target vehicle images captured by the video image;

identifying a segment of the target vehicle images by a function, wherein the function identifies at least one measuring vehicle image from the target vehicle images; and

identifying the vehicle image as a measuring vehicle upon the vehicle image matching the at least one measuring vehicle image; and

upon determining the vehicle image is a measuring vehicle, deriving a scaling map comprising a scaling factor associating a vehicle image dimension, in the direction of travel, to a real-world vehicle dimension at a point collocated with the vehicle image in one frame of the video image.

14. The computer-readable medium of claim 13 having further computer-executable instructions for performing the steps of:

segmenting the curve into segments representing roadway segments of a known length with distances determined in accord with the scaling map; and

determining the progression of a target vehicle by determining the progression of the target vehicle'"'"'s image relative to at least one segment.

15. The computer-readable medium of claim 13 wherein the function identifies the mean vehicle image size of the number of target vehicle images.

16. The computer-readable medium of claim 13, wherein the tracking point is a centroid.

17. The computer-readable medium of claim 13 having further computer-executable instructions for performing the steps of:

receiving a number of additional vehicle images;

determining tracking points for the number of additional vehicle images, each of the number of additional vehicle images contributing a tracking point for at least two frames of the video image; and

determined a number of curves for each path of the number of additional vehicle images.