Detecting orientation of digital images using face detection information

• US 8,081,844 B2
• Filed: 11/18/2010
• Issued: 12/20/2011
• Est. Priority Date: 06/26/2003
• Status: Expired due to Term

• Current Assignee: FotoNation Limited (Adeia Inc.)
• Original Assignee: DigitalOptics Corporation Europe Limited (Adeia Inc.)
• Inventors: Steinberg, Eran, Prilutsky, Yury, Corcoran, Peter, Bigioi, Petronel, Blonk, Leo, Gângea, Mihnea, Vertan, Constantin
• Primary Examiner(s): COUSO, YON JUNG
• Application Number:
  US12/949,751
• Publication Number:
  US 20110064329A1
• Time in Patent Office: 397 Days
• Field of Search: 382/100, 382/159, 382/170, 382/224, 382/228, 382/289, 382/291, 382/293, 382/296, 382/297, 345648-659, 348/583
• US Class Current: 382/296
• CPC Class Codes: G06V 10/242 by image rotation, e.g. by 90 degrees
• CPC Class Codes: G06V 40/161 Detection; Localisation; Normalisation

Abstract

A method of automatically establishing the correct orientation of an image using facial information. This method is based on the exploitation of the inherent property of image recognition algorithms in general and face detection in particular, where the recognition is based on criteria that is highly orientation sensitive. By applying a detection algorithm to images in various orientations, or alternatively by rotating the classifiers, and comparing the number of successful faces that are detected in each orientation, one may conclude as to the most likely correct orientation. Such method can be implemented as an automated method or a semi automatic method to guide users in viewing, capturing or printing of images.

Claims

1. A method of detecting an orientation of a main digital image using statistical classifier techniques, comprising:

using a processor;

generating in-camera, capturing or otherwise obtaining in-camera a collection of low resolution images including said face;

tracking said face within said collection of low resolution images;

acquiring a main digital image;

applying a set of face detection classifiers to at least one image of the collection of low resolution images in a first orientation;

determining a first level of match between a same or subsequent image of the collection of low resolution images at said first orientation and said classifiers;

rotating said same or subsequent image to a second orientation to obtain a rotated image, applying the classifiers to said rotated image at said second orientation;

determining a second level of match between said rotated image at said second orientation and said classifiers;

comparing said first and second levels of match between said classifiers and said same or subsequent image and between said classifiers and said rotated image, respectively; and

determining which of the first orientation and the second orientations has a greater probability of being a correct orientation based on which of the first and second levels of match, respectively, comprises a higher level of match.

2. The method of claim 1, further comprising, prior to applying said classifiers to said same or subsequent image, applying color conversion, edge enhancement, blurring, sharpening, tone reproduction correction, exposure correction, gray scale transformation, region segmentation, or cropping, or combinations thereof.

3. The method of claim 1, wherein said classifiers comprise elliptical classifiers.

4. The method of claim 3, wherein said elliptical classifiers are oriented at known orientations.

5. The method of claim 1, wherein said classifiers correspond to regions of a detected face.

6. The method of claim 5, wherein said regions include an eye, two eyes, a nose, a mouth, or an entire face, or combinations thereof.

7. A method of detecting an orientation of a main digital image using statistical classifier techniques comprising:

using a processor;

generating in-camera, capturing or otherwise obtaining in-camera a collection of low resolution images including said face;

tracking said face within said collection of low resolution images;

acquiring a main digital image;

applying a set of face detection classifiers to at least one image of the collection of low resolution images in a first orientation;

determining a first level of match between a same or subsequent image of the collection of low resolution images at said first orientation and said classifiers;

rotating said set of classifiers a first predetermined amount, applying the classifiers rotated said first amount to said same or subsequent image at said first orientation, and

determining a second level of match between said same or subsequent image at said first orientation and said classifiers rotated said first amount;

comparing said first and second levels of match between said classifiers and said digital image and between said rotated classifiers and said same or subsequent image, respectively; and

determining which of the first and second levels of match, respectively, comprises a higher level of match in order to determine whether said first orientation is a correct orientation of said main digital image.

8. The method of claim 7, further comprising, prior to applying said classifiers to said same or subsequent image, applying color conversion, edge enhancement, blurring, sharpening, tone reproduction correction, exposure correction, gray scale transformation, region segmentation, or cropping, or combinations thereof.

9. The method of claim 7, wherein said classifiers comprise elliptical classifiers.

10. The method of claim 9, wherein said elliptical classifiers are initially oriented at known orientations and, when rotated by said first and second amounts, are rotated to different known orientations.

11. The method of claim 7, wherein said classifiers correspond to regions of a detected face.

12. The method of claim 11, wherein said regions include an eye, two eyes, a nose, a mouth, or an entire face, or combinations thereof.

13. One or more non-transitory computer readable storage devices having processor readable code embodied thereon, said processor readable code for programming one or more processors to perform a method of detecting an orientation of a main digital image using statistical classifier techniques, the method comprising:

using a processor;

generating in-camera, capturing or otherwise obtaining in-camera a collection of low resolution images including said face;

tracking said face within said collection of low resolution images;

acquiring a main digital image;

applying a set of face detection classifiers to at least one image of the collection of low resolution images in a first orientation;

determining a first level of match between a same or subsequent image of the collection of low resolution images at said first orientation and said classifiers;

rotating said same or subsequent image to a second orientation to obtain a rotated image, applying the classifiers to said rotated image at said second orientation, and determining a second level of match between said rotated image at said second orientation and said classifiers;

comparing said first and second levels of match between said classifiers and said same or subsequent image and between said classifiers and said rotated image, respectively; and

determining which of the first orientation and the second orientations has a greater probability of being a correct orientation based on which of the first and second levels of match, respectively, comprises a higher level of match.

14. The one or more storage devices of claim 13, wherein the method further comprises, prior to applying said classifiers to said same or subsequent image, applying color conversion, edge enhancement, blurring, sharpening, tone reproduction correction, exposure correction, gray scale transformation, region segmentation, or cropping, or combinations thereof.

15. The one or more storage devices of claim 13, wherein said classifiers comprise elliptical classifiers.

16. The one or more storage devices of claim 15, wherein said elliptical classifiers are oriented at known orientations.

17. The one or more storage devices of claim 13, wherein said classifiers correspond to regions of a detected face.

18. The one or more storage devices of claim 17, wherein said regions include an eye, two eyes, a nose, a mouth, or an entire face, or combinations thereof.

19. One or more non-transitory computer readable storage devices having processor readable code embodied thereon, said processor readable code for programming one or more processors to perform a method of detecting an orientation of a main digital image using statistical classifier techniques, the method comprising:

using a processor;

generating in-camera, capturing or otherwise obtaining in-camera a collection of low resolution images including said face;

tracking said face within said collection of low resolution images;

acquiring a main digital image;

applying a set of face detection classifiers to at least one image of the collection of low resolution images in a first orientation;

determining a first level of match between a same or subsequent image of the collection of low resolution images at said first orientation and said classifiers;

rotating said set of classifiers a first predetermined amount, applying the classifiers rotated said first amount to said same or subsequent image at said first orientation, and

determining a second level of match between said same or subsequent image at said first orientation and said classifiers rotated said first amount;

comparing said first and second levels of match between said classifiers and said digital image and between said rotated classifiers and said same or subsequent image, respectively; and

determining which of the first and second levels of match, respectively, comprises a higher level of match in order to determine whether said first orientation is a correct orientation of said main digital image.

20. The one or more storage devices of claim 19, wherein the method further comprises, prior to applying said classifiers to said same or subsequent image, applying color conversion, edge enhancement, blurring, sharpening, tone reproduction correction, exposure correction, gray scale transformation, region segmentation, or cropping, or combinations thereof.

21. The one or more storage devices of claim 19, wherein said classifiers comprise elliptical classifiers.

22. The one or more storage devices of claim 21, wherein said elliptical classifiers are initially oriented at known orientations and, when rotated by said first and second amounts, are rotated to different known orientations.

23. The one or more storage devices of claim 19, wherein said classifiers correspond to regions of a detected face.

24. The one or more storage devices of claim 23, wherein said regions include an eye, two eyes, a nose, a mouth, or an entire face, or combinations thereof.

25. A portable digital camera, comprising:

one or more optics and a sensor for acquiring a digital image,

a processor, and

one or more processor readable storage devices having processor readable code embodied thereon for programming the processor to perform a method of detecting an orientation of a main digital image using statistical classifier techniques, wherein the method comprises;

using a processor;

generating in-camera, capturing or otherwise obtaining in-camera a collection of low resolution images including said face;

tracking said face within said collection of low resolution images;

acquiring a main digital image;

applying a set of face detection classifiers to at least one image of the collection of low resolution images in a first orientation;

determining a first level of match between a same or subsequent image of the collection of low resolution images at said first orientation and said classifiers;

rotating said same or subsequent image to a second orientation to obtain a rotated image, applying the classifiers to said rotated image at said second orientation, and determining a second level of match between said rotated image at said second orientation and said classifiers;

comparing said first and second levels of match between said classifiers and said same or subsequent image and between said classifiers and said rotated image, respectively; and

determining which of the first orientation and the second orientations has a greater probability of being a correct orientation based on which of the first and second levels of match, respectively, comprises a higher level of match.

26. The camera of claim 25, wherein the method further comprises, prior to applying said classifiers to said same or subsequent image, applying color conversion, edge enhancement, blurring, sharpening, tone reproduction correction, exposure correction, gray scale transformation, region segmentation, or cropping, or combinations thereof.

27. The camera of claim 25, wherein said classifiers comprise elliptical classifiers.

28. The camera of claim 27, wherein said elliptical classifiers are oriented at known orientations.

29. The camera of claim 25, wherein said classifiers correspond to regions of a detected face.

30. The camera of claim 29, wherein said regions include an eye, two eyes, a nose, a mouth, or an entire face, or combinations thereof.

31. A portable digital camera, comprising:

one or more optics and a sensor for acquiring a digital image,

a processor, and

one or more processor readable storage devices having processor readable code embodied thereon for programming the processor to perform a method of detecting an orientation of a main digital image using statistical classifier techniques, wherein the method comprises;

using a processor;

generating in-camera, capturing or otherwise obtaining in-camera a collection of low resolution images including said face;

tracking said face within said collection of low resolution images;

acquiring a main digital image;

applying a set of face detection classifiers to at least one image of the collection of low resolution images in a first orientation;

determining a first level of match between a same or subsequent image of the collection of low resolution images at said first orientation and said classifiers;

rotating said set of classifiers a first predetermined amount, applying the classifiers rotated said first amount to said same or subsequent image at said first orientation, and

determining a second level of match between said same or subsequent image at said first orientation and said classifiers rotated said first amount;

comparing said first and second levels of match between said classifiers and said digital image and between said rotated classifiers and said same or subsequent image, respectively; and

determining which of the first and second levels of match, respectively, comprises a higher level of match in order to determine whether said first orientation is a correct orientation of said main digital image.

32. The camera of claim 31, wherein the method further comprises, prior to applying said classifiers to said same or subsequent image, applying color conversion, edge enhancement, blurring, sharpening, tone reproduction correction, exposure correction, gray scale transformation, region segmentation, or cropping, or combinations thereof.

33. The camera of claim 31, wherein said classifiers comprise elliptical classifiers.

34. The camera of claim 33, wherein said elliptical classifiers are initially oriented at known orientations and, when rotated by said first and second amounts, are rotated to different known orientations.

35. The camera of claim 31, wherein said classifiers correspond to regions of a detected face.

36. The camera of claim 35, wherein said regions include an eye, two eyes, a nose, a mouth, or an entire face, or combinations thereof.