Method and apparatus for confirmation of relevant white inner circle in environment of circular traffic sign recognition

1. An apparatus for confirming circular traffic sign recognition, comprising:

an image acquisition unit configured to photograph a front of a vehicle using a camera to acquire a front image;

a sign recognition unit detecting circular shapes indicating circles of objects in the front image, classifying traffic sign circles indicating a traffic sign for the circles, and recognizing the circular traffic sign; and

a control unit confirming the circular traffic sign recognition using a white inner circle and informing a driver of contents of the confirmed circular traffic sign.

2. The apparatus of claim 1 , wherein the control unit sets a base reference area at a center of the classified traffic sign circle, sets one or a plurality of edge areas, and calculates a reference value for comparison from the base reference area.

3. The apparatus of claim 2 , wherein the control unit verifies a circle at which the white inner circle is accurately positioned when a maximum intensity of the area is brighter than the reference value for comparison or a ratio of pixels within the certain tolerance satisfies a minimum quality threshold value, with respect to the one or the plurality of edge areas.

4. The apparatus of claim 2 , wherein the control unit sets the base reference area at the center of the classified traffic sign circle and sets one or the plurality of edge areas within a radius from the center.

5. The apparatus of claim 1 , wherein the control unit confirms the circular traffic sign recognition as being correct by the white inner circle being accurately positioned when a ratio of pixels depending on sharpness of each area is brighter than a reference value for comparison or satisfies a minimum quality threshold value based on the white inner circle of a traffic sign partitioned in a red color or a clear traffic sign, with respect to the classified traffic sign circles.

6. A method for confirming circular traffic sign recognition, comprising:

acquiring a front image by photographing a front of a vehicle using a camera;

detecting circular shapes indicating circles of objects in the front image;

classifying traffic sign circles indicating a traffic sign for the circles;

recognizing a circular traffic sign for the classified traffic sign circles; and

confirming the circular traffic sign recognition using a white inner circle.

7. The method of claim 6 , further comprising setting a base reference area at a center of the classified traffic sign circle, setting one or a plurality of edge areas, and calculating a reference value for comparison from the base reference area.

8. The method of claim 7 , further comprising verifying a circle at which the white inner circle is accurately positioned when a maximum intensity of the area is brighter than the reference value for comparison or a ratio of pixels within the certain tolerance satisfies a minimum quality threshold value, with respect to the one or plurality of edge areas.

9. The method of claim 7 , further comprising setting the base reference area at the center of the classified traffic sign circle and setting one or the plurality of edge areas within a radius from the center.

10. The method of claim 6 , further comprising confirming the circular traffic sign recognition as being correct by the white inner circle being accurately positioned when a ratio of pixels depending on sharpness of each area is brighter than a reference value for comparison or satisfies a minimum quality threshold value based on the white inner circle of a traffic sign partitioned in a red color or a clear traffic sign, with respect to the classified traffic sign circles.