Method and device for detecting elliptical structures in an image

• US 10,013,619 B2
• Filed: 03/25/2016
• Issued: 07/03/2018
• Est. Priority Date: 03/26/2015
• Status: Active Grant

• Current Assignee: HL Mando Corporation (Halla Holdings Corporation)
• Original Assignee: Mando Corporation
• Inventors: Abhau, Jochen
• Primary Examiner(s): Couso, Yon
• Application Number:
  US15/081,838
• Publication Number:
  US 20160283806A1
• Time in Patent Office: 830 Days
• CPC Class Codes: G06T 2207/20061 Hough transform
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06V 10/48 by mapping characteristic values of the pattern into a parameter space, e.g. Hough transformation
• CPC Class Codes: G06V 20/582 of traffic signs

Abstract

The invention relates to method of detecting elliptical structures (10) in an image (9), comprising:

• • detecting circular arc-shaped structures (11) in the image (9) using a circle Hough transform (CHT) of the image (9), wherein a radius and a center point (12) are determined for each circular arc-shaped structure (11),
  • identifying pairs of circular arc-shaped structures (11) consisting of two of the detected circular arc-shaped structures (11) with substantially equal radii,
  • defining, for each one of these pairs, a search area (14) within the image (9) depending on the center points (12) of the respective pair of circular arc-shaped structures (11),
  • searching in the search area (14) defined for any given pair of circular arc-shaped structures (11), for a pair of edges (16) connecting these two circular arc-shaped structures (11). The invention further relates to a device for detecting elliptical structures in an image.

Claims

1. A method of providing assistance to a driver of a vehicle, the method comprising:

producing images of a surrounding of the vehicle by means of at least one camera;

detecting elliptical structures in the images by detecting circular arc-shaped structures in the image by means of a circle Hough transform (CHT) of the image, wherein a radius and a center point are determined for each circular arc-shaped structure, identifying pairs of circular arc-shaped structures consisting of two of the detected circular arc-shaped structures with substantially equal radii, defining, for each of the pairs of circular arc-shaped structures, a search area within the image depending on the center points of the respective pair of circular arc-shaped structures, searching, for each of the pairs of circular arc-shaped structures, in the search area defined for the pair of circular arc-shaped structures, for a pair of edges connecting the two circular arc-shaped structures of this pair of circular arc-shaped structures, and identifying, if a pair of edges connecting the two circular arc-shaped structures of a pair of circular arc-shaped structures is found, an area enclosed by this pair of edges and this pair of circular arc-shaped structures as an elliptical structure; and

identifying circular traffic signs using information about detected elliptical structures.

2. The method of claim 1, wherein the step of detecting circular arc-shaped structures includes the steps of:

defining a circle Hough parameter space and a circle Hough accumulator, wherein each parameter point of the circle Hough parameter space defines a circle in the image, and wherein the circle Hough accumulator assigns a value to each parameter point,

identifying image points of the image which vote for one or more of the circles defined by the parameter points,

increasing the values of the circle Hough accumulator depending on the number of image points which are identified as voting for the circles defined by the respective parameter points,

detecting parameter points in the circle Hough parameter space at which the circle Hough accumulator exceeds a predefined threshold value and/or at which the circle Hough accumulator has a local maximum, wherein the detected circular arc-shaped structures are represented by the detected parameter points.

3. The method of claim 2, wherein a detected circular arc-shaped structure which is not included in any of the identified pairs of circular arc-shaped structures is identified as a complete circle if the value of the circle Hough accumulator at the parameter point representing this detected circular arc-shaped structure exceeds a predefined threshold.

4. The method of claim 1, wherein, for each of the identified pairs of circular arc-shaped structures, the search area of the pair of circular arc-shaped structures consists of two subareas which are separated from each other by a straight connection line connecting the two center points of this pair of circular arc-shaped structures.

5. The method of claim 4, wherein, for each of the identified pairs of circular arc-shaped structures, the step of searching for a pair of edges is restricted to searching for a pair of edges for which each one of the two subareas defined for this pair of circular arc-shaped structures includes exactly one of the two edges of this pair of edges.

6. The method of claim 1, including the step of:

pre-processing the image, before calculating the circle Hough transform (CHT) of the image, by applying a gradient extraction and/or an edge detection algorithm to the image to determine gradients and/or to detect edges in the image and

using a result of the pre-processing step for calculating the circle Hough transform (CHT) of the image.

7. The method of claim 6, wherein the step of searching the pairs of edges in the defined search areas includes:

assessing edges of the image which have been detected by means of the edge detection algorithm applied to the image before calculating the circle Hough transform of the image.

8. The method of claim 1, wherein the step of searching for the pairs of edges includes applying an edge detection algorithm exclusively to the defined search areas after calculating the circle Hough transform of the image.

9. The method of claim 1, wherein the method further includes, for at least one of the pairs of edges which has been found for one of the identified pairs of circular arc-shaped structures, at least one of the steps of:

evaluating whether the edges are substantially straight,

comparing the lengths of the two edges with a distance between the center points of the respective pair of circular arc-shaped structures.

10. The method of claim 1, the method further comprises providing information about identified circular traffic signs to the driver by means of a driver information unit.

11. A driver assistance system, comprising:

at least one camera for producing images of a surrounding of a vehicle,

a device for detecting elliptical structures in an image, comprising a data processing unit detecting circular arc-shaped structures in the image by means of a circle Hough transform (CHT) of the image, a radius and a center point being determined for each circular arc-shaped structure, identifying pairs of circular arc-shaped structures consisting of two of the detected circular arc-shaped structures with substantially equal radii, defining, for each of the pairs of circular arc-shaped structures, a search area within the image depending on the center points of the respective pair of circular arc-shaped structures, searching, for each of the pairs of circular arc-shaped structures, in the search area defined for the pair of circular arc-shaped structures, for a pair of edges connecting the two circular arc-shaped structures of this pair of circular arc-shaped structures, and identifying, if a pair of edges connecting the two circular arc-shaped structures of a pair of circular arc-shaped structures is found, an area enclosed by this pair of edges and this pair of circular arc-shaped structures as an elliptical structure, wherein the data processing unit of the device is coupled to the at least one camera to receive the images from the camera,

a driver information unit coupled to the data processing unit and providing information on identified circular traffic signs to the driver.

12. The system of claim 11, wherein the data processing unit of the device:

detects circular arc-shaped structures in the image by means of a circle Hough transform (CHT) of the image, wherein a radius and a center point are determined for each circular arc-shaped structure,

identifies pairs of circular arc-shaped structures consisting of two of the detected circular arc-shaped structures with substantially equal radii,

defines, for each of the pairs of circular arc-shaped structures, a search area within the image depending on the center points of the respective pair of circular arc-shaped structures,

searches, for each of the pairs of circular arc-shaped structures, in the search area defined for the pair of circular arc-shaped structures, for a pair of edges connecting the two circular arc-shaped structures of this pair of circular arc-shaped structures, and

identifies, if a pair of edges connecting the two circular arc-shaped structures of a pair of circular arc-shaped structures is found, an area enclosed by this pair of edges and this pair of circular arc-shaped structures as an elliptical structure.

13. A vehicle comprising the driver assistance system of claim 11, wherein the at least one camera is installed in the vehicle to produce images of a surrounding of the vehicle.