Vehicle area detecting apparatus and vehicle area determining method

• US 6,842,531 B2
• Filed: 03/05/2003
• Issued: 01/11/2005
• Est. Priority Date: 07/22/1998
• Status: Expired due to Fees

• Current Assignee: Honda Elesys Co., Ltd. (Nidec Corporation)
• Original Assignee: Honda Elesys Co., Ltd. (Nidec Corporation)
• Inventors: Kyo, Sholin, Ohtsuka, Kazuyoshi
• Primary Examiner(s): Mehta, Bhavesh M.
• Assistant Examiner(s): Choobin, Barry
• Application Number:
  US10/379,871
• Publication Number:
  US 20030156737A1
• Time in Patent Office: 678 Days
• Field of Search: 382/104, 382/103, 382/170, 382/108, 382/199, 340/907, 340/436, 116/63, 348/113, 348/118, 348/169, 348/149, 701/1, 701/23, 701/301, 359/834
• US Class Current: 382/104
• CPC Class Codes: B60R 16/0231 Circuits relating to the driving or the functioning of the vehicle monitoring tyres B60C23/00; indicating overspeed B60K31/00; for dash boards B60K37/00, B60Q3/10; for indicating emergencies B60Q1/52; brake control systems B60T; registering or indicating the working of vehicles G07C5/00; measuring distance G01C, e.g. combinations of speed and distance G01C23/00; engine indicators G01L; measuring speed or acceleration G01P
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
• CPC Class Codes: G06V 20/58 Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
• CPC Class Codes: G06V 2201/08 Detecting or categorising vehicles

Abstract

A vehicle area detector correctly detects a vehicle area. The vehicle area detector detects a vehicle area in a vehicle candidate area having a high probability of existence of a vehicle in an image produced by imaging means mounted on a vehicle. For the correct detection of the vehicle area, the detector includes a horizontal symmetry evaluator to evaluate a degree of horizontal symmetry with respect to a degree of gradation in the image of the area, a variance evaluator to evaluate a degree of variance in the degree of gradation in the image of the area, a lower-area mean gradation evaluator to evaluate a difference in the degree of gradation of an image between a lower area of a vehicle candidate area and an area in the proximity of a lower edge of the candidate area exceeds a predetermined threshold value, and an area determining unit to determine a vehicle candidate area as the vehicle area in accordance with results of evaluation from the symmetry evaluator, the variance evaluator, and the lower-are mean gradation evaluator.

Claims

1. An apparatus for detecting objects in a vehicle area, said apparatus detects said vehicle area from a vehicle candidate area having a high probability of existence of a vehicle in an image produced by an imager mounted on a vehicle, comprising:

horizontal symmetry evaluator for evaluating a degree of horizontal symmetry with respect to a degree of gradation in the image of the area,

variance evaluator for evaluating a degree of variance in the degree of gradation in the image of the area,

mean gradation evaluator for evaluating a difference in the degree of gradation of an image between a lower area of a vehicle candidate area and an area in the proximity of a lower edge of the candidate area;

area determinator for determining a vehicle candidate area as the vehicle area when either one of the results of evaluation from the symmetry evaluator, the variance evaluator, and the mean gradation evaluator exceeds a predetermined threshold value; and

area evaluator for detecting a vehicle candidate area at a position substantially equal to that of a vehicle candidate area as the vehicle area when an event in which either one of the results of evaluation from the symmetry evaluator, the variance evaluator, and the mean gradation evaluator exceeds a predetermined threshold value is detected up to this point a predetermined number of times for the vehicle candidate area by the area determinator.

2. An apparatus for detecting objects in a vehicle area, said, said apparatus detects said vehicle area from a vehicle candidate area having a high probability of existence of a vehicle in an image produced by an imager mounted on a vehicle, comprising:

horizontal symmetry evaluator for evaluating a degree of horizontal symmetry with respect to a degree of gradation in the image of the area,

variance evaluator for evaluating a degree of variance in the degree of gradation in the image of the area,

mean gradation evaluator for evaluating a difference in the degree of gradation of an image between a lower area of a vehicle candidate area and an area in the proximity of a lower edge of the candidate area;

area determinator for determining a vehicle candidate area as the vehicle area when two of the results of evaluation from the symmetry evaluator, the variance evaluator, and the mean gradation evaluator exceed respective predetermined threshold values; and

area determinator for detecting a vehicle candidate area at a position substantially equal to that of a vehicle candidate area as the vehicle area when an event in which two of the results of evaluation from the symmetry evaluator, the variance evaluator, and the mean gradation evaluator exceed respective predetermined threshold values is detected up to this point a predetermined number of times for the vehicle candidate area by the area determinator.

3. An apparatus for detecting objects in a vehicle area, said apparatus detects said vehicle area from a vehicle candidate area having a high probability of existence of a vehicle in an image produced by an imager mounted on a vehicle, comprising:

horizontal symmetry evaluator for evaluating a degree of horizontal symmetry with respect to a degree of gradation in the image of the area,

variance evaluator for evaluating a degree of variance in the degree of gradation in the image of the area,

mean gradation evaluator for evaluating a difference in the degree of gradation of an image between a lower area of a vehicle candidate area and an area in the proximity of a lower edge of the candidate area;

area determinator for determining a vehicle candidate area as the vehicle area when the results of evaluation from the symmetry evaluator, the variance evaluator, and the mean gradation evaluator exceed respective predetermined threshold values; and

area determinator for detecting a vehicle candidate area at a position substantially equal to that of a vehicle candidate area as the vehicle area when an event in which the results of evaluation from the symmetry evaluator, the variance evaluator, and the mean gradation evaluator exceed respective predetermined threshold values is detected up to this point a predetermined number of times for the vehicle candidate area by the area determinator.

4. A method for detecting objects in an area surrounding a vehicle, comprising the steps of:

imaging an image of an object; and

detecting a vehicle area including a vehicle in the image produced by the imaging step, wherein the vehicle area detecting step includes the step of detecting the vehicle area when the vehicle area includes an image for which two of the conditions of

a degree of horizontal symmetry with respect to a degree of gradation in the image of the area,

a degree of variance in the degree of gradation in the image of the area, and

a difference in the degree of gradation of an image between a lower area of a vehicle candidate area and an area in the proximity of a lower edge of the candidate area exceed respective predetermined threshold values.

5. A method for detecting objects in an area surrounding a vehicle, comprising the steps of:

imaging an image of an object; and

detecting a vehicle area including a vehicle in the image produced by the imaging step, wherein the vehicle area detecting step includes the step of detecting the vehicle area when the vehicle area includes an image for which

a degree of horizontal symmetry with respect to a degree of gradation in the image of the area is larger than a predetermined threshold value,

a degree of variance in the degree of gradation in the image of the area is larger than a predetermined threshold value, and

a difference in the degree of gradation of an image between a lower area of a vehicle candidate area and an area in the proximity of a lower edge of the candidate area is larger than a predetermined threshold value.

6. A method for detecting objects in an area surrounding a vehicle, comprising the steps of:

imaging an image of an object; and

detecting a vehicle area including a vehicle in the image produced by the imaging step, wherein the vehicle area detecting step includes the step of detecting the vehicle area as an area at a position substantially equal to that of the vehicle area when the area satisfies up to this point a predetermined number of times two of the conditions of

a degree of horizontal symmetry with respect to a degree of gradation in the image of the area,

a degree of variance in the degree of gradation in the image of the area, and

a difference in the degree of gradation of an image between a lower area of a vehicle candidate area and an area in the proximity of a lower edge of the candidate area.

7. A method for detecting objects in an area surrounding a vehicle, comprising the steps of:

imaging an image of an object; and

detecting a vehicle area including a vehicle in the image produced in said imaging step, wherein the vehicle area detecting step includes the step of detecting the vehicle area as an area at a position substantially equal to that of the vehicle area when the area satisfies up to this point a predetermined number of times the conditions for

a degree of horizontal symmetry with respect to a degree of gradation in the image of the area,

a degree of variance in the degree of gradation in the image of the area, and

a difference in the degree of gradation of an image between a lower area of a vehicle candidate area and an area in the proximity of a lower edge of the candidate area.

8. A method for detecting a vehicle area in a vehicle candidate area that has a high probability of the existence of a vehicle in an image produced by an imager mounted on a vehicle, comprising:

a horizontal symmetry evaluating step of evaluating a degree of horizontal symmetry with respect to a degree of gradation in the image of the area,

a variance evaluating step of evaluating a degree of variance in the degree of gradation in the image of the area,

a mean gradation evaluating step of evaluating a difference in the degree of gradation of an image between a lower area of a vehicle candidate area and an area in the proximity of a lower edge of the candidate area; and

an area determining step of determining a vehicle candidate area as the vehicle area when either one of the results of evaluation from the symmetry evaluating step, the variance evaluating step, and the mean gradation evaluating step exceeds a predetermined threshold value.

9. A method for detecting a vehicle area in a vehicle candidate area that has a high probability of the existence of a vehicle in an image produced by an imager mounted on a vehicle, comprising:

a horizontal symmetry evaluating step of evaluating a degree of horizontal symmetry with respect to a degree of gradation in the image of the area,

a variance evaluating step of evaluating a degree of variance in the degree of gradation in the image of the area,

a mean gradation evaluating step of evaluating a difference in the degree of gradation of an image between a lower area of a vehicle candidate area and an area in the proximity of a lower edge of the candidate area; and

an area determining step of determining a vehicle candidate area as the vehicle area when two of the results of evaluation from the symmetry evaluating step, the variance evaluating step, and the mean gradation evaluating step exceed respective predetermined threshold values.

10. A method for detecting a vehicle area in a vehicle candidate area that has a high probability of the existence of a vehicle in an image produced by an imager mounted on a vehicle, comprising:

a horizontal symmetry evaluating step of evaluating a degree of horizontal symmetry with respect to a degree of gradation in the image of the area,

a variance evaluating step of evaluating a degree of variance in the degree of gradation in the image of the area,

a mean gradation evaluating step of evaluating a difference in the degree of gradation of an image between a lower area of a vehicle candidate area and an area in the proximity of a lower edge of the candidate area; and

an area determining step of determining a vehicle candidate area as the vehicle area when the results of evaluation from the symmetry evaluating step, the variance evaluating step, and the mean gradation evaluating step exceed respective predetermined threshold values.

11. A method for detecting a vehicle area in a vehicle candidate area that has a high probability of the existence of a vehicle in an image produced by an imager mounted on a vehicle, comprising:

a horizontal symmetry evaluating step of evaluating a degree of horizontal symmetry with respect to a degree of gradation in the image of the area,

a variance evaluating step of evaluating a degree of variance in the degree of gradation in the image of the area,

a mean gradation evaluating step of evaluating a difference in the degree of gradation of an image between a lower area of a vehicle candidate area and an area in the proximity of a lower edge of the candidate area;

an area determining step of determining a vehicle candidate area as the vehicle area when either one of the results of evaluation from the symmetry evaluating step, the variance evaluating step, and the mean gradation evaluating step exceeds a predetermined threshold value; and

an area evaluating step for detecting a vehicle candidate area at a position substantially equal to that of a vehicle candidate area as the vehicle area when an event in which either one of the results of evaluation from the symmetry evaluating step, the variance evaluating step, and the mean gradation evaluating step exceeds a predetermined threshold value is detected up to this point a predetermined number of times for the vehicle candidate area by the area determining step.

12. A method for detecting a vehicle area in a vehicle candidate area that has a high probability of the existence of a vehicle in an image produced by an imager mounted on a vehicle, comprising:

a horizontal symmetry evaluating step of evaluating a degree of horizontal symmetry with respect to a degree of gradation in the image of the area,

a variance evaluating step of evaluating a degree of variance in the degree of gradation in the image of the area,

a mean gradation evaluating step of evaluating a difference in the degree of gradation of an image between a lower area of a vehicle candidate area and an area in the proximity of a lower edge of the candidate area;

an area determining step of determining a vehicle candidate area as the vehicle area when two of the results of evaluation from the symmetry evaluating step, the variance evaluating step, and the mean gradation evaluating step exceed respective predetermined threshold values; and

an area evaluating step for detecting a vehicle candidate area at a position substantially equal to that of a vehicle candidate area as the vehicle area when an event in which two of the results of evaluation from the symmetry evaluating step, the variance evaluating step, and the mean gradation evaluating step exceed respective predetermined threshold values is detected up to this point a predetermined number of times for the vehicle candidate area by the area determining step.

13. A method for detecting a vehicle area in a vehicle candidate area that has a high probability of the existence of a vehicle in an image produced by an imager mounted on a vehicle, comprising:

a horizontal symmetry evaluating step of evaluating a degree of horizontal symmetry with respect to a degree of gradation in the image of the area,

a variance evaluating step of evaluating a degree of variance in the degree of gradation in the image of the area,

a mean gradation evaluating step of evaluating a difference in the degree of gradation of an image between a lower area of a vehicle candidate area and an area in the proximity of a lower edge of the candidate area;

an area determining step of determining a vehicle candidate area as the vehicle area when the results of evaluation from the symmetry evaluating step, the variance evaluating step, and the mean gradation evaluating step exceed respective predetermined threshold values; and

an area evaluating step for detecting a vehicle candidate area at a position substantially equal to that of a vehicle candidate area as the vehicle area when an event in which the results of evaluation from the symmetry evaluating step, the variance evaluating step, and the mean gradation evaluating step exceed respective predetermined threshold values is detected up to this point a predetermined number of times for the vehicle candidate area by the area determining step.