Image processing apparatus, image processing method, and storage medium

• US 8,923,553 B2
• Filed: 07/18/2011
• Issued: 12/30/2014
• Est. Priority Date: 07/21/2010
• Status: Active Grant

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Yashiro, Satoshi
• Primary Examiner(s): BEKELE, MEKONEN T
• Application Number:
  US13/185,388
• Publication Number:
  US 20120020519A1
• Time in Patent Office: 1,261 Days
• Field of Search: 382/103, 382/104, 382/229, 382/254, 382/132, 382/107, 375/240.15, 375/240.16, 375/240.23, 375/240.27, 348/234, 348/E09.053, 348/239, 348/207.99
• US Class Current: 382/103
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
• CPC Class Codes: G06V 40/103 Static body considered as a whole, e.g. static pedestrian or occupant recognition

Abstract

An image processing apparatus includes a region setting unit configured to set a specific region where a reflection may occur in an image, a size setting unit configured to set a size of an object to be detected in association with a position in the image, and a changed region detection unit configured to detect a changed region by comparing a background model and an input image, wherein the changed region detection unit outputs the changed region in the specific region based on the size of the object associated with a position of the changed region, in a case where the changed region extends beyond a boundary of the specific region.

Claims

1. An image processing apparatus comprising:

a region setting unit configured to set a specific region in an image in advance of detecting a changed region;

a size setting unit configured to set a size of an object to be detected in association with a position in the image;

a changed region detection unit configured to detect a changed region from an input image other than the specific region, by comparing a background model and the input image;

an object candidate region setting unit configured to set an object region in the specific region based on the size of the object associated with a position of the changed region, in a case where the changed region contacts with a boundary of the specific region; and

an extension unit configured to extend the changed region by combining the detected changed region with the object candidate region included in the specific region.

2. The image processing apparatus according to claim 1, further comprising an object detection unit configured to detect the object from the changed region based on the size of the object.

3. The image processing apparatus according to claim 2, wherein the object detection unit comprises,

an object candidate selection unit configured to select an object candidate from a region clipped from the input image, and

a determination unit configured to determine whether the object candidate is the object.

4. The image processing apparatus according to claim 3, wherein the object candidate selection unit selects the object candidate from the clipped region based on the size of the object.

5. The image processing apparatus according to claim 3, wherein the object candidate selection unit selects the object candidate based on a ratio of the changed region in the clipped region.

6. The image processing apparatus according to claim 1, wherein the changed region detection unit detects a changed pixel based on a difference between the background model and the input image, and combines a plurality of changed pixels adjacent to each other outside the specific region as a changed region block.

7. An image processing apparatus comprising:

a region setting unit configured to set a specific region where a reflection occurs in an image;

a size setting unit configured to set a size of an object to be detected in association with a position in the image; and

a changed region detection unit configured to detect a changed region by comparing a background model and an input image,

wherein the changed region detection unit outputs the changed region in the specific region based on the size of the object associated with a position of the changed region, in a case where the changed region extends beyond a boundary of the specific region,

wherein the changed region detection unit detects a changed pixel based on a difference between the background model and the input image, and combines a plurality of changed pixels adjacent to each other outside the specific region as a changed region block, and

wherein, in a case where the changed region block is in contact with the specific region, the changed region detection unit sets an object candidate region to the changed region block, and combines a region included in the object candidate region and the specific region, with the changed region block.

8. A method for processing an image, the method comprising:

setting a specific region in the image in advance of detecting a changed region;

setting a size of an object to be detected in association with a position in the image;

detecting a changed region from an input image other than the specific region, by comparing a background model and the input image;

setting an object region in the specific region based on the size of the object associated with a position of the changed region, in a case where the changed region contacts with a boundary of the specific region; and

extending the changed region by combining the detected changed region with the object candidate region included in the specific region.

9. A non-transitory storage medium storing a program that causes a computer to execute a method according to claim 8.

10. An image processing method comprising:

setting a specific region where a reflection occurs in an image;

setting a size of an object to be detected in association with a position in the image;

detecting a changed region by comparing a background model and an input image; and

outputting the changed region in the specific region based on the size of the object associated with a position of the changed region, in a case where the changed region extends beyond a boundary of the specific region,

wherein in detecting the changed region, a changed pixel is detected based on a difference between the background model and the input image, and a plurality of changed pixels adjacent to each other outside the specific region are combined into a changed region block, and

wherein, in a case where the changed region block is in contact with the specific region, setting an object candidate region to the changed region block, and combining a region included in the object candidate region and the specific region, with the changed region block.

11. A non-transitory storage medium storing a program that causes a computer to execute a method according to claim 10.