Image processing device and image processing method, and imaging device

• US 7,324,669 B2
• Filed: 01/27/2004
• Issued: 01/29/2008
• Est. Priority Date: 01/31/2003
• Status: Active Grant

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Takano, Hiroaki, Nakanishi, Toshiaki, Sugizaki, Kyoko
• Primary Examiner(s): Ahmed; Samir
• Assistant Examiner(s): BLOOM, NATHAN J
• Application Number:
  US10/543,163
• Publication Number:
  US 20060133654A1
• Time in Patent Office: 1,463 Days
• Field of Search: 382/100, 382103-104, 382/118, 382/224
• US Class Current: 382/118
• CPC Class Codes: G03B 17/53 for automatically delivering a finished picture after a signal causing exposure has been given, e.g. by pushing a button, by inserting a coin
• CPC Class Codes: G06T 1/0007 Image acquisition
• CPC Class Codes: G06V 10/50 by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
• CPC Class Codes: G06V 40/165 using facial parts and geometric relationships

Abstract

Disclosed is an image processing apparatus in which the cheek of a person captured in an image of a photo may be corrected to a finish with the cheek looking slimmer than in a two-dimensional photo. A face correction unit (400) calculates, in a calculation unit (411), a face length L1 and a cheek width L2 of a face of a person as an object and L3 corrected from the face length L1. A face classifying unit (412) compares L2 and L3 to each other to classify the face shape into a ‘round shape’, ‘a long shape’ and a ‘square shape’. Depending on the results of classification, the image trimming is carried out by a first contour correction unit (413), a second contour correction unit (414), and a third contour correction unit (415) so that the cheek will look slim.

Claims

1. An image processing apparatus comprising

face area extraction means for extracting a face area from an image of a person;

detection means for detecting the contour of the face of said person from the face area extracted by said face area extraction means;

face classification means for classifying the shape of the face of said person, based on the contour of the face of said person as detected by said detection means; and

image correcting means for correcting the contour of said face of said person based on the type of the face shape of said person classified by said face classification means.

2. The image processing apparatus according to claim 1 wherein said detection means detects the positions of the head top, eyes, mouth and the jaw, and the cheek;

and wherein

said face classification means classifies the shape of the face of said person into at least three types of L2=L3, L2<

L3 and L2>

L3, wherein L1 is the length from the head top to the jaw of said person as detected by said detection means, L2 is the width of the cheek of said person at the position of his/her mouth as detected by said detection means, and L3 is said length L1 multiplied by a preset coefficient α

.

3. The image processing apparatus according to claim 2 wherein said image correcting means effects image correction for reducing the width-wise direction of said face area when the face shape of the person as classified by said face classification means is L2=L3.

4. The image processing apparatus according to claim 2 wherein said image correcting means effects image correction for shading the contour portion of the face of said person in a range of said face area from a preset location between the eyes and the mouth up to the jaw of said person when the face shape of the person as classified by said face classification means is L2=L3.

5. The image processing apparatus according to claim 2 wherein said image correcting means effects image correction for reducing the width-wise direction of said face area in a range of said face area from a preset location between the eyes and the mouth up to the jaw of said person when the face shape of the person as classified by said face classification means is L2<

L3.

6. The image processing apparatus according to claim 2 wherein said image correcting means effects image correction for shading the contour portion of the face of said person in a range of said face area from a preset location between the eyes and the mouth up to the jaw of said person when the face shape of the person as classified by said face classification means is L2<

L3.

7. The image processing apparatus according to claim 2 wherein said image correcting means effects image correction for reducing the width-wise direction of said face area and image correction for reducing the width-wise direction of said face area in a range of said face area from a preset location between the eyes and the mouth up to the jaw of said person when the face shape of the person as classified by said face classification means is L2>

L3.

8. The image processing apparatus according to claim 2 wherein said image correcting means effects image correction for reducing the width-wise direction of said face area and image correction for shading the contour portion of the face of said person in a range of said face area from a preset location between the eyes and the mouth up to the jaw of said person when the face shape of the person as classified by said face classification means is L2>

L3.

9. An image processing method comprising

a face area extraction step for extracting a face area from an image of a person;

a detection step for detecting the contour of the face of said person from the face area extracted by said face area extraction step;

a face classification step for classifying the shape of the face of said person, based on the contour of the face of said person as detected by said detection step; and

an image correcting step for correcting the contour of said face of said person based on the type of the face shape of said person.

10. The image processing method according to claim 9 wherein said detection step detects the positions of the head top, eyes, mouth and the jaw, and the cheek;

and wherein

said face classification step classifies the shape of the face of said person into at least three types of L2=L3, L2<

L3 and L2>

L3, wherein L1 is the length from the head top to the jaw of said person, L2 is the width of the cheek of said person at the position of his/her mouth, and L3 is said length L1 multiplied by a preset coefficient α

.

11. The image processing method according to claim 10 wherein said image correcting step effects image correction for reducing the width-wise direction of said face area when the face shape of the person is L2=L3.

12. The image processing method according to claim 10 wherein said image correcting step effects image correction for shading the contour portion of the face of said person in a range of said face area from a preset location between the eyes and the mouth up to the jaw of said person when the face shape of the person as classified by said face classification step is L2=L3.

13. The image processing method according to claim 10 wherein said image correcting step effects image correction for reducing the width-wise direction of said face area in a range of said face area from a preset location between the eyes and the mouth up to the jaw of said person when the face shape of the person as classified by said face classification step is L2<

L3.

14. The image processing method according to claim 10 wherein said image correcting step effects image correction for shading the contour portion of the face of said person in a range of said face area from a preset location between the eyes and the mouth up to the jaw of said person when the face shape of the person as classified by said face classification step is L2<

L3.

15. The image processing method according to claim 10 wherein said image correcting step effects image correction for reducing the width-wise direction of said face area and image correction for reducing the width-wise direction of said face area in a range of said face area from a preset location between the eyes and the mouth up to the jaw of said person when the face shape of the person as classified by said face classification step is L2>

L3.

16. The image processing method according to claim 10 wherein said image correcting step effects image correction for reducing the width-wise direction of said face area and image correction for shading the contour portion of the face of said person in a range of said face area from a preset location between the eyes and the mouth up to the jaw of said person when the face shape of the person as classified by said face classification step is L2>

L3.

17. An imaging apparatus comprising

imaging means for imaging a person;

face area extraction means for extracting a face area from an image of a person imaged by said imaging means;

detection means for detecting the contour of the face of said person from the face area extracted by said face area extraction means;

face classification means for classifying the shape of the face of said person based on the contour of the face of said person detected by said detection means; and

image correction means for correcting the contour of said face of said person based on the type of the face of said face classified by said face classification means.

18. The imaging apparatus according to claim 17 wherein said detection means detects the positions of the head top, eyes, mouth and the jaw, and the cheek;

and wherein

said face classification means classifies the shape of the face of said person into at least three types of L2=L3, L2<

L3 and L2>

L3, wherein L1 is the length from the head top to the jaw of said person as detected by said detection means, L2 is the width of the cheek of said person at the position of his/her mouth as detected by said detection means, and L3 is said length L1 multiplied by a preset coefficient α

.

19. The imaging apparatus according to claim 18 wherein said image correcting means effects image correction for reducing the width-wise direction of said face area when the face shape of the person as classified by said face classification means is L2=L3.

20. The imaging apparatus according to claim 18 wherein said image correcting means effects image correction for shading the contour portion of the face of said person in a range of said face area from a preset location between the eyes and the mouth up to the jaw of said person when the face shape of the person as classified by said face classification means is L2=L3.

21. The imaging apparatus according to claim 18 wherein said image correcting means effects image correction for reducing the width-wise direction of said face area in a range of said face area from a preset location between the eyes and the mouth up to the jaw of said person when the face shape of the person as classified by said face classification means is L2<

L3.

22. The imaging apparatus according to claim 18 wherein said image correcting means effects image correction for shading the contour portion of the face of said person in a range of said face area from a preset location between the eyes and the mouth up to the jaw of said person when the face shape of the person as classified by said face classification means is L2<

L3.

23. The imaging apparatus according to claim 18 wherein said image correcting means effects image correction for reducing the width-wise direction of said face area and image correction for reducing the width-wise direction of said face area in a range of said face area from a preset location between the eyes and the mouth up to the jaw of said person when the face shape of the person as classified by said face classification means is L2>

L3.

24. The imaging apparatus according to claim 18 wherein said image correcting means effects image correction for reducing the width-wise direction of said face area and image correction for shading the contour portion of the face of said person in a range of said face area from a preset location between the eyes and the mouth up to the jaw of said person when the face shape of the person as classified by said face classification means is L2>

L3.

25. The imaging apparatus according to claim 17 further comprising

printing means for printing an image of a person trimmed by said image correcting means.

26. The imaging apparatus according to claim 17 further comprising

illuminating means for illuminating said person.