Face detecting camera and method

• US 20050264658A1
• Filed: 05/05/2005
• Published: 12/01/2005
• Est. Priority Date: 02/28/2000
• Status: Active Grant

• Current Assignee: Monument Peak Ventures, LLC (Dominion Harbor Enterprises, LLC)
• Original Assignee: Eastman Kodak Company
• Inventors: Ray, Lawrence A., Nicponski, Henry
• Granted Patent: US 7,583,294 B2
• Time in Patent Office: Days
• US Class Current: 348/239
• CPC Class Codes: G06V 40/161 Detection; Localisation; Normalisation
• CPC Class Codes: G06V 40/162 using pixel segmentation or colour matching
• CPC Class Codes: G06V 40/165 using facial parts and geometric relationships
• CPC Class Codes: H04N 2101/00 Still video cameras
• CPC Class Codes: H04N 23/611 where the recognised objects include parts of the human body
• CPC Class Codes: H04N 23/64 Computer-aided capture of images, e.g. transfer from script file into camera, check of taken image quality, advice or proposal for image composition or decision on when to take image
• CPC Class Codes: H04N 23/70 Circuitry for compensating brightness variation in the scene
• CPC Class Codes: H04N 5/772 the recording apparatus and the television camera being placed in the same enclosure

Abstract

A method for determining the presence of a face from image data utilizes at least first and second algorithms. The first algorithm prescreens the image data, by determining a plurality of face candidates utilizing a pattern matching technique that identifies image windows likely to contain faces based on color and shape information. The second algorithm processes the face candidates determined by the first algorithm, and uses a posterior probability function classifier to determine the presence of the face.

Claims

1. A method for determining the presence of a face from image data, said method comprising the steps of:

(a) prescreening the image data with a first algorithm, the first algorithm determining a plurality of face candidates utilizing a pattern matching technique that identifies image windows likely to contain faces based on color and shape information; and

(b) operating on the face candidates with a second algorithm, the second algorithm processing the face candidates using a posterior probability function classifier to determine the presence of the face.

2. The method as claimed in claim 1 wherein the first algorithm has a first rate of false positives, and the second algorithm has a second rate of false positives lower than the first rate of false positives.

3. The method as claimed in claim 1 wherein the method is implemented in a digital camera for capturing an image of a scene, said digital camera comprising:

a capture section for capturing an image and producing the image data;

an electronic processing section for processing the image data to determine the presence of one or more faces in the image;

face data means associated with the processing section for generating face data corresponding to at least one attribute of at least one of the faces in the image;

a storage medium for storing the image data; and

recording means associated with the processing section for recording the face data with the image data on the storage medium.

4. The method as claimed in claim 3 wherein the face data corresponds to at least one of the location, orientation, scale and pose of at least one of the faces in the image.

5. The method as claimed in claim 1 further comprising the step of providing an indication that one or more faces have been detected.

6. The method as claimed in claim 1 further comprising the step of recording the image data in a storage medium in one or more digital folders each dedicated to images with a particular number of faces associated therewith.

7. The method as claimed in claim 1 further comprising the steps of utilizing a face recognition algorithm and a data base of known faces to generate facial identities for respective detected faces, and labeling one or more images in a storage medium with the facial identities of known faces.

8. The method as claimed in claim 3 further comprising the step of controlling an exposure of the capture section responsive to the presence of one or more faces for optimally exposing the image for at least one of the faces in the scene.

9. The method as claimed in claim 3 further comprising the step of controlling an activation of a flash unit of the capture section in order to optimize exposure for at least one of the faces in the scene.

10. The method as claimed in claim 1 wherein the method is implemented in a digital camera for capturing an image of a scene, said digital camera comprising:

a capture section for capturing an image and producing the image data;

an electronic processing section for processing the image data to determine the presence of one or more faces in the scene;

face data means associated with the processing section for generating face data corresponding to at least one attribute of at least one of the faces in the image;

a composition algorithm associated with the processing section for processing the face data and generating composition suggestions for a user of the digital camera in response to the processed face data; and

a display device for displaying the composition suggestions to the user.

11. The method as claimed in claim 10 wherein the composition suggestions include at least one of (a) an indication that a main subject is too small in the image, (b) an indication that following the law of thirds will lead to a more pleasing composition, (c) an indication that one or faces have been cut off in the image, and (d) an indication that a horizontal alignment of subjects should be avoided in the image.

12. The method as claimed in claim 1 wherein the method is implemented in a digital camera for capturing an image of a scene, said digital camera comprising:

a capture section for capturing an image and producing the image data;

an electronic processing section for processing the image data to determine the presence of one or more faces in the scene and generating face data therefrom;

an orientation algorithm associated with the processing section for generating orientation data indicating orientation of the image based on the orientation of at least one of the faces in the image;

a storage medium for storing the image data; and

recording means associated with the processing section for recording the orientation data with the image data on the storage medium.

13. The method as claimed in claim 1 wherein the method is implemented in a digital camera for capturing an image of a scene, said digital camera comprising:

a capture section for capturing an image and producing the image data;

an electronic processing section for processing the image data to determine the presence of one or more faces in the scene;

a red eye detection algorithm associated with the electronic processing section for generating red eye signals indicating the presence of red eye in one or more of the faces; and

a display device responsive to the red eye signals for displaying a red eye warning to a user of the digital camera.

14. The method as claimed in claim 13 further comprising the step of performing red eye correction responsive to the red eye signals for correcting the red eye in said one or more faces.

15. The method as claimed in claim 1 further comprising the steps of:

processing the image data with a red eye detection algorithm for generating red eye signals indicating the presence of red eye in the image; and

corroborating the existence of red eye by verifying that the red eye signals correspond to the presence of one or more faces in the image.

16. The method as claimed in claim 15 further comprising the step of correcting the red eye in the image.

17. A digital camera configured to perform the first and second algorithms of the method as claimed in claim 1.

18. A method for determining the presence of a face from image data, said method comprising the steps of:

(a) prescreening the image data with a first algorithm, the first algorithm determining a plurality of face candidates; and

(b) operating on the face candidates with a second algorithm, the second algorithm processing the face candidates to determine the presence of the face;

wherein the first algorithm has a first rate of false positives, and the second algorithm has a second rate of false positives lower than the first rate of false positives.

19. The method as claimed in claim 18 further comprising the steps of:

processing the image data with a red eye detection algorithm for generating red eye signals indicating the presence of red eye in the image; and

corroborating the existence of red eye by verifying that the red eye signals correspond to the presence of one or more faces in the image.

20. A computer program product for determining the presence of a face, comprising:

a computer readable storage medium having a computer program stored thereon for performing the steps of;

(a) prescreening the image data with a first algorithm, the first algorithm determining a plurality of face candidates utilizing a pattern matching technique that identifies image windows likely to contain faces based on color and shape information; and

(b) operating on the face candidates with a second algorithm, the second algorithm processing the face candidates using a posterior probability function classifier to determine the presence of the face.