Vehicle counting and emission estimation

• US 9,489,581 B2
• Filed: 08/11/2014
• Issued: 11/08/2016
• Est. Priority Date: 08/11/2014
• Status: Expired due to Fees

• Current Assignee: Piece Future Pte., Ltd.
• Original Assignee: Nokia Technologies Oy (Nokia Corporation)
• Inventors: Chen, Xin, Huang, Xinyu
• Primary Examiner(s): Dunphy, David F
• Application Number:
  US14/456,404
• Publication Number:
  US 20160042234A1
• Time in Patent Office: 820 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06T 2207/30236 Traffic on road, railway or crossing
• CPC Class Codes: G06T 2207/30242 Counting objects in image
• CPC Class Codes: G06T 7/40 Analysis of texture depth or shape recovery from texture G06T7/529
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
• CPC Class Codes: G06V 20/54 of traffic, e.g. cars on the road, trains or boats
• CPC Class Codes: G06V 2201/08 Detecting or categorising vehicles
• CPC Class Codes: G06V 30/422 Technical drawings; Geographical maps
• CPC Class Codes: G08G 1/00 Traffic control systems for road vehicles arrangement of road signs or traffic signals E01F9/00 ; automatic vehicle control B62D
• CPC Class Codes: G08G 1/0116 from roadside infrastructure, e.g. beacons
• CPC Class Codes: G08G 1/0133 for classifying traffic situation
• CPC Class Codes: G08G 1/0141 for traffic information dissemination
• CPC Class Codes: G08G 1/04 using optical or ultrasonic detectors
• CPC Class Codes: G08G 1/065 by counting the vehicles in a section of the road or in a parking area, i.e. comparing incoming count with outgoing count

Abstract

A computing device, for example, receives image data of vehicles on a path captured by a camera. The image is analyzed using a low level feature extraction on the image data. The computing device estimates exhaust levels produced by the vehicles based on the low level feature extraction or based on vehicle classifications and quantities determined from the low level feature extraction.

Claims

1. A method comprising:

receiving image data of vehicles on a path;

performing, using a processor, low level feature extraction on the image data;

selecting a regression strategy based on a traffic measurement for the path, wherein the regression strategy includes the low level feature extraction; and

estimating exhaust levels produced by the vehicles based on the low level feature extraction,

wherein inputs to the regression strategy includes only the low level feature extraction when the traffic measurement is a low level.

2. The method of claim 1, wherein the low level feature extraction determines a dimension of a foreground image segment, a dimension of an object in the foreground image segment, a texture coarseness for the foreground image segment, or a combination thereof.

3. The method of claim 1, wherein performing low level feature extraction on the image data comprises:

identifying, from the low level feature extraction, a quantity for a first vehicle class from the image data of vehicles on the path; and

identifying, from the low level feature extraction and the quantity for the first vehicle class, a quantity for a second vehicle class from the image data of vehicles on the path.

4. The method of claim 3, wherein the first vehicle class includes a first size of vehicles and second vehicle class includes a second size of vehicles.

5. The method of claim 3, wherein performing low level feature extraction on the image data further comprises:

identifying, from the low level feature extraction and the quantity for the second vehicle class, a quantity for a third vehicle class from the image data of vehicles on the path.

6. The method of claim 3, wherein estimating exhaust levels produced by the vehicles based on the low level feature extraction comprising:

determining exhaust levels based on the quantity for the first vehicle class and the quantity for the second vehicle class.

7. The method of claim 1, wherein estimating exhaust produced by the vehicles based on the low level feature extraction comprises:

determining exhaust levels based on a vehicle count.

8. The method of claim 1, further comprising:

overlaying graphical indicia for the exhaust levels of the path on geographic data.

9. The method of claim 1, wherein when the regression strategy includes a vehicle count when the traffic measurement is a high level.

10. A method comprising:

receiving image data of vehicles on a path;

performing, using a processor, low level feature extraction on the image data; and

estimating exhaust levels produced by the vehicles based on the low level feature extraction,

wherein performing low level feature extraction on the image data comprises;

identifying, from the low level feature extraction, a quantity for a first vehicle class from the image data of vehicles on the path; and

identifying, from the low level feature extraction and the quantity for the first vehicle class, a quantity for a second vehicle class from the image data of vehicles on the path; and

concatenating data for the quantity for the first vehicle class to a feature vector for the low level feature extraction.

11. An apparatus comprising:

at least one processor; and

at least one memory including computer program code for one or more programs;

the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least perform;

receiving image data of vehicles on a path;

performing low level feature extraction on the image data;

identifying, based on the low level feature extraction, a quantity for a first vehicle class from the image data of vehicles on the path;

identifying, based on the low level feature extraction and the quantity for the first vehicle class, a quantity for a second vehicle class from the image data of vehicles on the path; and

concatenating data for the first vehicle class and data for the second vehicle class to define a feature vector.

12. The apparatus of claim 11, wherein the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least perform:

estimating exhaust levels produced by the vehicles based on the quantity for the first vehicle class and the quantity for second vehicle class.

13. The apparatus of claim 12, wherein the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least perform:

overlaying graphical indicia for the exhaust levels of the path on geographic data.

14. The apparatus of claim 11, wherein the low level feature extraction include a dimension of a foreground image segment, a dimension of an object in the foreground image segment, a texture coarseness for the foreground image segment, or a combination thereof.

15. The apparatus of claim 11, wherein the first vehicle class includes a first size of vehicles and second vehicle class includes a second size of vehicles.

16. The apparatus of claim 11, wherein the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least perform:

identifying, based on the low level feature extraction and the quantity for the second vehicle class, a quantity for a third vehicle class from the image data of vehicles on the path.

17. The apparatus of claim 11, wherein the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least perform:

selecting a regression strategy based on a traffic measurement for the path, wherein the regression strategy includes the low level feature extraction.

18. A non-transitory computer readable medium including instructions that when executed on a computer are operable to:

receiving image data of vehicles on a path;

selecting a regression strategy from a plurality of regression strategies based on a traffic measurement for the path, wherein the traffic measurement includes a traffic speed or a traffic level, wherein the regression strategy includes a low level feature extraction; and

estimating exhaust levels for the vehicles based on the regression strategy.

19. The non-transitory computer readable medium of claim 18, wherein the regression strategy includes a vehicle count when the traffic measurement is a high level.

20. The non-transitory computer readable medium of claim 18, wherein the vehicle count is determined based on:

identifying, based on the low level feature extraction, a quantity for a first vehicle class from the image data of vehicles on the path; and

identifying, based on the low level feature extraction and the quantity for the first vehicle class, a quantity for a second vehicle class from the image data of vehicles on the path.

21. A non-transitory computer readable medium including instructions that when executed on a computer are operable to:

receiving image data of vehicles on a path;

selecting a regression strategy based on a traffic measurement for the path, wherein the regression strategy includes a low level feature extraction; and

estimating exhaust levels for the vehicles based on the regression strategy, wherein inputs to the regression strategy include only the low level feature extraction when the traffic measurement is a low level.