METHOD AND APPARATUS FOR PARTITIONING AN OBJECT FROM AN IMAGE

• US 20080118149A1
• Filed: 11/12/2007
• Published: 05/22/2008
• Est. Priority Date: 11/17/2006
• Status: Active Grant

• Current Assignee: Alpine Electronics Incorporated (Alps Alpine Co., Ltd.)
• Original Assignee: Alpine Electronics Incorporated (Alps Alpine Co., Ltd.)
• Inventors: Su, Shanghai, Fu, Pengyu, Yuan, Huai
• Granted Patent: US 8,098,933 B2
• Time in Patent Office: Days
• US Class Current: 382/174
• 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

The invention relates to a method and apparatus for partitioning an object from an image, such that substantially the entire object is contained in the partitioned region. An embodiment of the present invention includes pre-setting a reference width for the object; extracting a shadow underneath the object from the image and determining a candidate region containing the object in the image based on the extracted shadow underneath an object in the image; acquiring an imaging width of the reference width at the location of the extracted shadow underneath the object in the image, and adjusting the candidate region based on the imaging width of the reference width such that the adjusted candidate region substantially completely contains the object; and partitioning the adjusted candidate region as a region containing the object from the image.

Claims

1. A method of partitioning an object from an image, the method comprising:

pre-setting a reference width for an object;

extracting a shadow underneath the object from an image;

determining a candidate region containing the object in the image based on the extracted shadow underneath the object;

acquiring an imaging width of the reference width at a location of the extracted shadow underneath the object in the image;

adjusting the candidate region based on the imaging width of the reference width such that the adjusted candidate region substantially completely contains the object; and

partitioning the adjusted candidate region as a region containing the object from the image.

2. The method of partitioning an object from an image according to claim 1, the method further comprising:

determining whether the width of the extracted shadow underneath the object is less than the imaging width of the reference width at the location of the extracted shadow underneath the object in the image,

wherein, if the width of the extracted shadow underneath the object is less than the imaging width of the reference width at the location of the extracted shadow underneath the object in the image, the candidate region is adjusted to substantially completely contain the object by extending the candidate region based on a difference between the width of the extracted shadow underneath the object and the imaging width of the reference width at the location of the extracted shadow underneath the object in the image.

3. The method of partitioning an object from an image according to claim 2, the method further comprising:

determining whether a left vertical edge of the object can be located near the left edge of the extracted shadow underneath the object in the image; and

determining whether a right vertical edge of the object can be located near the right edge of the extracted shadow underneath the object in the image.

4. The method of partitioning an object from an image according to claim 3, wherein, if both the left and right vertical edges of the object can be found, the candidate region is adjusted to substantially completely contain the object by extending the left and right edges of the candidate region by one half the difference between the width of the extracted shadow underneath the object and the imaging width of the reference width.

5. The method of partitioning an object from an image according to claim 3, wherein, if neither the left nor right vertical edges of the object can be found, the candidate region is adjusted to substantially completely contain the object by extending the left and right edges of the candidate region by one half the difference between the width of the extracted shadow underneath the object and the imaging width of the reference width.

6. The method of partitioning an object from an image according to claim 3, further comprising:

determining the difference between the width of the shadow underneath the object and the imaging width of the reference width;

wherein, if only the right vertical edge of the object is found, the candidate region is adjusted to substantially completely contain the object by extending the left side of the candidate region outward by the difference between the width of the extracted shadow underneath the object and the imaging width of the reference width.

7. The method of partitioning an object from an image according to claim 6, wherein, if only the right vertical edge of the object is found, the right side of the adjusted candidate region is extended outward by a predetermined distance.

8. The method of partitioning an object from an image according to claim 3, further comprising:

determining the difference between the width of the extracted shadow underneath the object and the imaging width of the reference width;

wherein, if only the left vertical edge of the object can be found near the left edge of the extracted shadow underneath the object in the image, the candidate region is adjusted to substantially completely contain the object by extending the right side of the candidate region outward by the difference between the width of the shadow underneath the object and the imaging width of the reference width.

9. The method of partitioning an object from an image according to claim 8, wherein if only the left vertical edge of the object is found, the left side of the adjusted candidate region is extended outward by a predetermined distance.

10. The method of partitioning an object from an image according to claim 2, the method further comprising:

determining whether a ratio of the width of the extracted shadow underneath the object to the imaging width of the reference width at the location of the extracted shadow underneath the object in the image is within a predetermined range, if the width of the shadow underneath the object is determined not to be less than the imaging width of the reference width at the location of the shadow underneath the object in the image; and

partitioning a first region substantially completely containing a first object if the ratio is within the predetermined range, wherein the first region is located rightward from the left edge of the shadow underneath the object in the determined candidate region, and the width of the first region is the width of the imaging width of the reference width; and

partitioning a second region substantially completely containing a second object if the ratio is within the predetermined range, wherein the second region is located leftward from the right edge of the shadow underneath the object in the candidate region, and the width of the second region is the width of the imaging width of the reference width.

11. The method according to claim 10, wherein the predetermined range is between about 1.2 and about 2.5.

12. The method of partitioning an object from an image according to claim 1, wherein the reference width is the maximum width of the object.

13. The method of partitioning an object from an image according to claim 1, wherein the object in the image includes a vehicle.

14. The method of partitioning an object from an image according to claim 1, the method further comprising, calculating the imaging width of the reference width at the location of the shadow underneath the object in the image.

15. The method of partitioning an object from an image according to claim 14, further comprising, calculating the imaging width of the reference width in the image using the pinhole imaging principle.

16. A method of partitioning an object from an image, the method comprising:

pre-setting a reference width for an object;

pre-calculating an imaging width of the reference width at the location of an extracted shadow underneath the object in an image;

extracting a shadow underneath the object from the image;

determining a candidate region containing the object in the image based on the extracted shadow underneath the object;

retrieving the pre-calculated imaging width of the reference width at the location of the extracted shadow underneath the object in the image;

adjusting the candidate region based on the imaging width of the reference width such that the adjusted candidate region substantially completely contains the object; and

partitioning the adjusted candidate region as a region containing the object from the image.

17. The method of partitioning an object from an image according to claim 16, wherein the reference width is the maximum width of the object.

18. The method of partitioning an object from an image according to claim 16, the method further comprising, calculating the imaging width of the reference width in the image using the pinhole imaging principle.

19. An apparatus for partitioning an object from an image, comprising:

a setting unit for pre-setting a reference width for an object;

an extracting and determining unit for extracting a shadow underneath the object from an image, and determining a candidate region containing the object in the image based on the extracted shadow underneath the object, wherein the extracting and determining unit is configured to communicate electronically with the setting unit;

an acquiring unit for acquiring an imaging width of the reference width at the location of the extracted shadow underneath the object in the image, wherein the acquiring unit is configured to communicate electronically with the setting unit and an adjusting and partitioning unit; and

an adjusting and partitioning unit for adjusting the candidate region according to the imaging width of the reference width such that the adjusted candidate region substantially completely contains the object, and partitioning the adjusted candidate region as a region containing the object from the image, wherein the adjusting and partitioning unit is configured to communicate electronically with the acquiring unit.