Image processing apparatus, image processing method, and program

• US 8,494,233 B2
• Filed: 12/15/2011
• Issued: 07/23/2013
• Est. Priority Date: 01/30/2007
• Status: Active Grant

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Kaneda, Yuji, Matsugu, Masakazu, Mori, Katsuhiko
• Primary Examiner(s): STREGE, JOHN B
• Application Number:
  US13/326,710
• Publication Number:
  US 20120082384A1
• Time in Patent Office: 586 Days
• Field of Search: 382/118, 382/190
• US Class Current: 382/118
• CPC Class Codes: G06V 40/165 using facial parts and geometric relationships
• CPC Class Codes: G06V 40/171 Local features and components; Facial parts eye characteristics G06V40/18; Occluding parts, e.g. glasses; Geometrical relationships
• CPC Class Codes: G06V 40/175 Static expression

Abstract

A feature point detection unit (153) and feature amount extraction unit (154) extract a plurality of features of an object from input image data. When there are unextracted features of the plurality of features, a weight setting unit (155) sets weights for the extracted features. A facial expression determination unit (156) executes recognition processing of the object based on the features weighted by the weight setting unit (155).

Claims

1. An image processing apparatus comprising:

a feature extraction unit adapted to extract a plurality of predetermined features in an object from input image data;

a determination unit adapted to determine whether all of the plurality of predetermined features have been extracted;

a first weight setting unit adapted to set weights preset in advance for the plurality of predetermined features when it is determined that all of the plurality of predetermined features have been extracted;

a second weight setting unit adapted to set respective weights for the features extracted by said feature extraction unit when it is determined that there are features which have not been extracted by said feature extraction unit, wherein the respective weights for the extracted features are determined in accordance with which features have not been extracted from among the plurality of predetermined features; and

a recognition processing unit adapted to execute recognition processing of the object based on the features weighted by said first or second weight setting unit,

wherein a processor is included in said apparatus.

2. The apparatus according to claim 1, wherein the object is a person, and said apparatus further comprises a face detection unit adapted to detect a face of the object from the image data,

wherein said feature extraction unit extracts, from the image data, a plurality of predetermined features on the face of the object detected by said face detection unit.

3. The apparatus according to claim 2, further comprising a normalization unit adapted to normalize a size and direction of the face of the object detected by said face detection unit,

wherein said feature extraction unit extracts, from the image data, a plurality of predetermined features on the face of the object normalized by said normalization unit.

4. The apparatus according to claim 1, further comprising a region setting unit adapted to set, for the image data, regions associated with the plurality of features to be extracted by said feature extraction unit,

wherein said feature extraction unit extracts the plurality of features from the regions set by said region setting unit.

5. The apparatus according to claim 1, wherein said feature extraction unit calculates variations associated with the features as the features on the object.

6. The apparatus according to claim 1, wherein the object is a person, and said recognition processing unit executes as the recognition processing one of recognition processing of a facial expression of the object and personal verification of the object.

7. The apparatus according to claim 1, further comprising a storage unit adapted to store the image data based on a result of the recognition processing by said recognition processing unit.

8. An image processing method performed by a processor, the method comprising:

a feature extraction step of extracting a plurality of predetermined features in an object from input image data;

a determination step of determining whether all of the plurality of predetermined features have been extracted;

a first weight setting step of setting weights preset in advance for the plurality of predetermined features when it is determined that all of the plurality of predetermined features have been extracted;

a second weight setting step of setting respective weights for the features extracted in the feature extraction step when it is determined that there are features which have not been extracted in the feature extraction step, wherein the respective weights for the extracted features are determined in accordance with which features have not been extracted from among the plurality of predetermined features; and

a recognition processing step of executing recognition processing of the object based on the features weighted in the first or second weight setting step.

9. A program stored in a non-transitory computer-readable storage medium for making a computer execute a method comprising:

a feature extraction step of extracting a plurality of predetermined features in an object from input image data;

a determination step of determining whether all of the plurality of predetermined features have been extracted;

a first weight setting step of setting weights preset in advance for the plurality of predetermined features when it is determined that all of the plurality of predetermined features have been extracted;

a second weight setting step of setting respective weights for the features extracted in the feature extraction step when it is determined that there are features which have not been extracted in the feature extraction step, wherein the respective weights for the extracted features are determined in accordance with which features have not been extracted from among the plurality of predetermined features; and

a recognition processing step of executing recognition processing of the object based on the features weighted in the first or second weight setting step.