Method and system for vehicle classification

• US 9,239,955 B2
• Filed: 03/14/2012
• Issued: 01/19/2016
• Est. Priority Date: 03/14/2011
• Status: Active Grant

• Current Assignee: Regents of the University of California (University of California)
• Original Assignee: Regents of the University of California (University of California)
• Inventors: Bhanu, Bir, Thakoor, Ninad
• Primary Examiner(s): ANSARI, TAHMINA N
• Application Number:
  US14/005,102
• Publication Number:
  US 20140037142A1
• Time in Patent Office: 1,406 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 2218/12 Classification; Matching
• CPC Class Codes: G06V 10/98 Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns
• CPC Class Codes: G06V 20/54 of traffic, e.g. cars on the road, trains or boats
• CPC Class Codes: G06V 30/2504 Coarse or fine approaches, e.g. resolution of ambiguities or multiscale approaches
• CPC Class Codes: G08G 1/015 with provision for distinguishing between two or more types of vehicles, e.g. between motor-cars and cycles
• CPC Class Codes: G08G 1/04 using optical or ultrasonic detectors

Abstract

A method and system of vehicle classification, and more particularly to a method and system called hierarchical vehicle classification system using a video and/or video image, a method and system of vehicle classification using a vehicle ground clearance measurement system, and method and system for classification of passenger vehicles and measuring their properties, and more particularly to capturing a vehicle traveling along a road from a single camera and classifying the vehicle into a vehicle class.

Claims

1. A method for classification of vehicles comprising:

(a) calibrating a transformation angle between a camera image and a road surface;

(b) detecting moving objects on the road surface in the camera image;

(c) establishing a vertical axis of symmetry for each of the moving objects;

(d) detecting dominant horizontal edges intersecting with the vertical axis of symmetry for each of the moving objects;

(e) tracking the dominant horizontal edges for a plurality of consecutive video frames;

(f) projecting each of the dominant horizontal edges onto a ground plane;

(g) treating each pair of adjacent projected edges as a projected surface;

(h) calculating changes in the projected surface; and

(i) calculating distinct structural signatures from the changes in the projected surfaces.

2. The method of claim 1, further comprising:

classifying the distinct structural signatures in a class based on the distinct structural signatures and wherein the moving object is a motor vehicle and the class includes one of the following;

sedan, pickup, minivan and sport utility vehicle (SUV).

3. The method of claim 2, comprising:

separating asymmetric parts of the moving objects and classifying them as non-vehicles, and after separating asymmetric parts of the moving objects and classifying them as non-vehicles, projecting a lower edge of the blob onto the road surface using the transformation in classification step (a), and wherein the projected size is a width of the vehicle.

4. The method of claim 1, further comprising after treating each pair of adjacent projected edges as a projected surface, calculating a speed of a vehicle using a location of a lowest projected surface at at least two different times.

5. The method of claim 2, further comprising classifying passenger vehicles on a road surface.

6. The method of claim 2, further comprising allocating parking garage and/or parking lot spaces based on a vehicle classification.

7. The method of claim 2, further comprising calculating vehicle speed for law enforcement.

8. The method of claim 2, further comprising ensuring limited and/or controlled access to vehicles for security related applications.

9. A system for classification of vehicles comprising:

a camera for capturing images of at least one moving object; and

a computer processing unit, which performs the following steps;

(a) calibrating a transformation angle between a camera image and a road surface;

(b) detecting moving objects on the road surface in the camera image;

(c) establishing a vertical axis of symmetry for each of the moving objects;

(d) detecting dominant horizontal edges intersecting with the vertical axis of symmetry for each of the moving objects;

(e) tracking the dominant horizontal edges for a plurality of consecutive video frames;

(f) projecting each of the dominant horizontal edges onto a ground plane;

(g) treating each pair of adjacent projected edges as a projected surface;

(h) calculating changes in the projected surface; and

(i) calculating distinct structural signatures from the changes in the projected surfaces.

10. The system of claim 9, wherein the computer processing unit further comprises a memory arrangement, a processing unit and an optional display unit for displaying data and/or classification of the distinct structures in the class.

11. The system of claim 9, wherein the camera captures video images.

12. The system of claim 9, further comprising classifying the distinct structural signatures in a class based on the distinct structural signatures.

13. The system of claim 12, further comprising separating asymmetric parts of the moving objects and classifying them as non-vehicles.

14. The system of claim 13, further comprising after separating asymmetric parts of the moving objects and classifying them as non-vehicles, projecting a lower edge of the blob onto the road surface using the transformation in classification step (a).

15. A computer program product comprising a non-transitory computer usable medium having a computer readable code embodied therein for classification of passenger vehicles and measuring their properties from a rear view video frame, the computer readable program code is configured to execute a process, which includes the steps of:

(a) calibrating a transformation angle between a camera image and a road surface;

(b) detecting moving objects on the road surface in the camera image;

(c) establishing a vertical axis of symmetry for each of the moving objects;

(d) detecting dominant horizontal edges intersecting with the vertical axis of symmetry for each of the moving objects;

(e) tracking the dominant horizontal edges for a plurality of consecutive video frames;

(f) projecting each of the dominant horizontal edges onto a ground plane;

(g) treating each pair of adjacent projected edges as a projected surface;

(h) calculating changes in the projected surface; and

(i) calculating distinct structural signatures from the changes in the projected surfaces.

16. The product of claim 15, further comprising classifying the distinct structural signatures in a class based on the distinct structural signatures.

17. The product of claim 16, wherein the moving object is a motor vehicle and the class includes one of the following sedan, pickup, minivan and sport utility vehicle (SUV).

18. The product of claim 17, further comprising separating asymmetric parts of the moving objects and classifying them as non-vehicles.

19. The product of claim 18, further comprising after separating asymmetric parts of the moving objects and classifying them as non-vehicles, projecting a lower edge of the blob onto the road surface using the transformation in classification step (a).