Apparatus and method for detecting object from image, and program

• US 8,942,511 B2
• Filed: 08/18/2011
• Issued: 01/27/2015
• Est. Priority Date: 08/26/2010
• Status: Expired due to Fees

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Yashiro, Satoshi
• Primary Examiner(s): Patel, Nirav G
• Assistant Examiner(s): SAINI, AMANDEEP SINGH
• Application Number:
  US13/212,546
• Publication Number:
  US 20120051592A1
• Time in Patent Office: 1,258 Days
• Field of Search: None
• US Class Current: 382/286
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
• CPC Class Codes: G06V 10/34 Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
• CPC Class Codes: G06V 10/473 using gradient analysis

Abstract

An image processing apparatus includes an input unit configured to input an image, a determining unit configured to determine a foreground area and a background area in the image input by the input unit, an expansion unit configured to expand the foreground area determined by the determining unit, a calculating unit configured to calculate a feature amount of the foreground area expanded by the expansion unit, and a detecting unit configured to detect an object from the image using the feature amount.

Claims

1. An image processing apparatus, comprising:

a processor; and

a memory storing instructions that, when executed by the processor, cause the processor to perform operations comprising;

inputting an image;

determining a foreground area and a background area in the input image;

reducing the input image at a plurality of reduction rates;

expanding the foreground area with respect to each of a plurality of reduced images obtained by reducing the input image at the plurality of reduction rates, by adding, to the foreground area, a surrounding area which is required for obtaining a feature amount of a boundary between the background area and the foreground area;

calculating a feature amount of the expanded foreground area; and

detecting an object from the foreground area using the feature amount of the expanded foreground area.

2. The image processing apparatus according to claim 1, wherein calculating comprises calculating the feature amount by a convolution operation of the image.

3. The image processing apparatus according to claim 1, wherein calculating comprises calculating the feature amount representing a gradient magnitude by a differential operation of the image.

4. The image processing apparatus according to claim 3, wherein calculating comprises setting the feature amount of areas other than the expanded foreground area to a predetermined value.

5. The image processing apparatus according to claim 1, wherein expanding comprises expanding the foreground area with respect to each of the plurality of reduced images according to respective reduction rates.

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

inputting an image;

determining a foreground area and a background area in the image;

reducing the input image at a plurality of reduction rates;

expanding the determined foreground area with respect to each of a plurality of reduced images obtained by reducing the input image at the plurality of reduction rates, by adding, to the foreground area, a surrounding area which is required for obtaining a feature amount of a boundary between the background area and the foreground area;

calculating a feature amount of the expanded foreground area; and

detecting an object from the foreground area using the feature amount of the expanded foreground area.

7. A non-transitory storage medium storing a program for causing a computer to execute each step in a method according to claim 6.