Method for road sign recognition

Patent metadata

Field	Value
Publication number	US-8953842-B2
Application number	US-201113696851-A
Country	US
Kind code	B2
Filing date	Apr 28, 2011
Priority date	May 14, 2010
Publication date	Feb 10, 2015
Grant date	Feb 10, 2015

Abstract

For recognizing road signs, a camera captures image data of the surroundings of a vehicle. The image data are analyzed to determine a region that contains a potential road sign. The image region is evaluated by a first classification unit to identify a road sign belonging to a particular class based on a recognized class-specific feature. Then, the brightness or color intensity of at least a portion of the road sign is analyzed along radially extending scanning beams, to determine potential contour points of an information-bearing part of the road sign, which is then extracted and semantically interpreted in a second classification unit to determine the information content thereof.

First claim

The invention claimed is:

1. A method for road sign recognition, in which images of a region surrounding the vehicle are acquired by a camera,
image data of the camera are analyzed, wherein at least one image region that contains a potential road sign is determined,
the image region that contains the potential road sign is submitted to a first classification unit that identifies the potential road sign, on the basis of at least one class-specific feature, as a road sign belonging to a particular class having the class-specific feature,

characterized in that
a brightness or a color intensity of a surface of the road sign is analyzed along radially extending scanning beams starting from a defined starting point within the surface of the road sign, in a brightness or color intensity analysis that comprises a one-dimensional gradient formation of brightness values or color intensity values of pixels of the image data along the scanning beams,
potential contour points of an information-bearing part of the road sign are determined from results of the brightness or color intensity analysis,
a limited region of the surface of the road sign is extracted as an information-bearing part of the road sign that includes all determined potential contour points, and
the extracted limited region is submitted to a second classification unit for semantic interpretation.

2. The method according to claim 1 , characterized in that the first classification unit defines the defined starting point for brightness analysis or color intensity analysis.

3. The method according to claim 1 , characterized in that the center of the surface of the road sign is determined and used as a defined starting point.

4. The method according to claim 1 , characterized in that the first classification unit determines the inner surface of the road sign of a particular class and the brightness or color intensity analysis along the scanning beams is only performed within the inner surface of the road sign.

5. The method according to claim 4 , characterized in that a region of a bright-background road sign for which the potential contour points were determined as dark-to-bright transitions during the brightness analysis performed along the scanning beams from the starting point outward is extracted as an information-bearing part.

6. The method according to claim 4 , characterized in that a region of a dark-background road sign for which the potential contour points were determined as bright-to-dark transitions during brightness analysis performed along the scanning beams from the starting point outward is extracted as an information-bearing part.

7. The method according to claim 1 , characterized in that the region of the road sign which is extracted as an information-bearing part is determined as a smallest rectangle that includes all said potential contour points.

8. The method according to claim 1 , characterized in that the image data of the road sign is subjected to
a) brightness normalization and/or
b) noise suppression and/or
c) masking a surrounding region of the road sign that surrounds the surface,

prior to the brightness or color intensity analysis.

9. The method according to claim 1 , characterized in that the brightness or color intensity analysis of the surface of the road sign is performed along a particular number of the radially extending scanning beams, wherein the number of radially extending scanning beams depends on the determined number of image regions with potential road signs in a camera image or on the number of the road signs identified in a camera image by the first classification unit.

10. The method according to claim 1 , characterized in that
the brightness or color intensity analysis of the surface of the road sign is performed along a predetermined minimum number of the radially extending scanning beams,
the limited region extracted on the basis of brightness analysis or color intensity analysis is submitted to the second classification unit for semantic interpretation, and
if the second classification unit cannot classify the extracted limited region successfully, the number of scanning beams is increased before again performing the brightness or color intensity analysis and extraction of the region.

11. The method according to claim 10 , characterized in that the number of scanning beams is only increased up to a defined maximum number.

12. The method according to claim 1 , characterized in that all angles between adjacent scanning beams are the same.

13. The method according to claim 1 , characterized in that twenty-four scanning beams radially extending from the starting point outward are provided, wherein all angles between adjacent scanning beams ( 5 ) are each 15°.

14. A device for performing the method for road sign recognition according to claim 1 , said device comprising a camera for acquiring images of a region surrounding the vehicle and an image processing unit with a first classification unit that comprises means for determining the information-bearing part of the road sign from the image data of the camera, and for forwarding the extracted region of the road sign that comprises the information-bearing part to a second classification unit for semantic interpretation.

15. A method for road sign recognition, in which images of a region surrounding the vehicle are acquired by a camera,
image data of the camera are analyzed, wherein at least one image region that contains a potential road sign is determined,
the image region that contains the potential road sign is submitted to a first classification unit that identifies the potential road sign, on the basis of at least one class-specific feature, as a road sign belonging to a particular class having the class-specific feature,

characterized in that
a brightness or a color intensity of a surface of the road sign is analyzed along radially extending scanning beams starting from a defined starting point within the surface of the road sign,
potential contour points of an information-bearing part of the road sign are determined from results of the brightness or color intensity analysis,
a limited region of the surface of the road sign is extracted as an information-bearing part of the road sign that includes all determined potential contour points, and
the extracted limited region is submitted to a second classification unit for semantic interpretation,
wherein the brightness or color intensity analysis is performed from the starting point outward along the scanning beams, and
wherein the limited region that is extracted as the information bearing part is a region of a bright-background road sign for which the potential contour points were determined as dark-to-bright transitions, or is a region of a dark-background road sign for which the potential contour points were determined as bright-to-dark transitions, respectively in the brightness or color intensity analysis.

16. A method of automatically evaluating a road sign in a driver assistance system in a motor vehicle, comprising steps:
a) with a camera capturing image data of surroundings of the motor vehicle including a road sign;
b) in an image processing arrangement, analyzing the image data to thereby recognize that there is a road sign represented in the image data;
c) in the image processing arrangement, after optionally pre-processing the image data, evaluating a brightness or a color intensity of selected pixels of the image data representing at least a portion of the road sign, wherein the selected pixels are located only along scanning beams that exten

Assignees

• Zobel Matthias
• Conti Temic Microelectronic

Inventors

• Zobel Matthias

Classifications

• G06V20/582 Primary

  of traffic signs · CPC title

• G06V10/421

  by analysing segments intersecting the pattern · CPC title

• G06K9/50 Primary

  Physics · mapped topic

• G06K9/00818

  Physics · mapped topic