Method and system for detecting traffic information

• US 7,869,935 B2
• Filed: 03/23/2007
• Issued: 01/11/2011
• Est. Priority Date: 03/23/2006
• Status: Expired due to Fees

• Current Assignee: Keysight Technologies, Inc.
• Original Assignee: Agilent Technologies, Inc.
• Inventors: Ma, Xin-yu, Jia, Huixing
• Primary Examiner(s): Camby; Richard M.
• Application Number:
  US11/726,920
• Publication Number:
  US 20070225895A1
• Time in Patent Office: 1,390 Days
• Field of Search: 701/117, 701/118, 701/119, 701/300, 340/907, 340/909, 340/937, 348/135, 348/139
• US Class Current: 701/117
• CPC Class Codes: G06V 20/54 of traffic, e.g. cars on the road, trains or boats
• CPC Class Codes: G08G 1/04 using optical or ultrasonic detectors
• CPC Class Codes: G08G 1/052 with provision for determining speed or overspeed speed measuring in general G01P

Abstract

A traffic information sensor system includes an image acquisition device to generate video images of a traffic scene. The system also includes an image detection device to extract traffic information from the video images in connection with the lane position information of the traffic scene. The system also includes an information transfer device to transmit the extracted traffic information instead of the video images to a traffic management center. A traffic information detection method and system is also described.

Claims

1. A traffic information sensor system, comprising:

an image acquisition device to generate video images of a traffic scene;

an image detection device to extract traffic information from the video images in connection with lane position information of the traffic scene; and

an information transfer device to transmit the extracted traffic information instead of the video images to a traffic management center.

2. The traffic information sensor system of claim 1, wherein the image detection device further comprises a lane localization module to detect and obtain the traffic lane position information from the video images.

3. The traffic information sensor system of claim 2, wherein the lane localization module further comprises

an activity map generation sub-module to generate an activity map by accumulating changes among the video images;

and

a lane discrimination module to obtain the lane position information by detecting local maximum and minimum values in the activity map.

4. The traffic information sensor system of claim 2, wherein the image detection device further comprises a vehicle counting module to count the number of vehicles within an interested lane, as part of the traffic information, of the traffic scene by

arranging a plurality of virtual lines at different positions on the lane, and

detecting the number of vehicles passing the virtual lines by detecting pixel changes at the virtual lines in the video images in order to obtain the vehicle count of the lane.

5. The traffic information sensor system of claim 4, wherein the pixel changes at the virtual line are determined by calculating an average value of several frames of images as an initial background of the lane and comparing the detected images with the initial background.

6. The traffic information sensor system of claim 4, wherein during night time, vehicles passing the virtual line are detected by detecting pixels whose luminance is higher than a predetermined threshold in the virtual line.

7. The traffic information sensor system of claim 2, wherein the image detection device further comprises an average vehicle speed calculation module to calculate average vehicle speed within an interested lane of the traffic information by

arranging a virtual line on the lane, and

detecting the time vehicles take to pass the virtual line by detecting pixel changes at the virtual line in the video images in order to obtain the average vehicle speed.

8. The traffic information sensor system of claim 7, wherein the pixel changes at the virtual line are determined by calculating an average value of several frames of images as an initial background of the lane and comparing the detected images with the initial background.

9. The traffic information sensor system of claim 7, wherein during night time, vehicles passing the virtual line are detected by detecting pixels whose luminance is higher than a predetermined threshold in the virtual line.

10. A traffic information detection system, comprising:

A plurality of traffic information sensors, each comprising;

an image acquisition device to generate video images of a traffic scene;

an image detection device to extract traffic information from the video images in connection with lane position information of the traffic scene; and

an information transfer device to transmit the extracted traffic information instead of the video images;

a communications network connected to the information transfer device of each of the sensors to transmit the extracted traffic information to a traffic management center.

11. The traffic information detection system of claim 10, wherein the image detection device further comprises a lane localization module to detect and obtain the traffic lane position information from the video images.

12. The traffic information detection system of claim 11, wherein the lane localization module further comprises

an activity map generation sub-module to generate an activity map by accumulating changes among the video images;

and

a lane discrimination module to obtain the lane position information by detecting local maximum and minimum values in the activity map.

13. The traffic information detection system of claim 11, wherein the image detection device further comprises a vehicle counting module to count the number of vehicles within an interested lane, as part of the traffic information, of the traffic scene by

arranging a plurality of virtual lines at different positions on the lane, and

detecting the number of vehicles passing the virtual lines by detecting pixel changes at the virtual lines in the video images in order to obtain the vehicle count of the lane.

14. The traffic information detection system of claim 13, wherein the pixel changes at the virtual line are determined by calculating an average value of several frames of images as an initial background of the lane and comparing the detected images with the initial background.

15. The traffic information detection system of claim 13, wherein during night time, vehicles passing the virtual line are detected by detecting pixels whose luminance is higher than a predetermined threshold in the virtual line.

16. The traffic information detection system of claim 11, wherein the image detection device further comprises an average vehicle speed calculation module to calculate average vehicle speed within an interested lane of the traffic information by

arranging a virtual line on the lane, and

detecting the time vehicles take to pass the virtual line by detecting pixel changes at the virtual line in the video images in order to obtain the average vehicle speed.

17. The traffic information detection system of claim 16, wherein the pixel changes at the virtual line are determined by calculating an average value of several frames of images as an initial background of the lane and comparing the detected images with the initial background.

18. The traffic information detection system of claim 16, wherein during night time, vehicles passing the virtual line are detected by detecting pixels whose luminance is higher than a predetermined threshold in the virtual line.

19. A traffic information detection system, comprising:

generating video images of a traffic scene;

extracting traffic information from the video images in connection with lane position information of the traffic scene; and

transmit the extracted traffic information instead of the video images to a traffic management center.