Machine-controlled image cropping with default

• US 20080019574A1
• Filed: 07/20/2006
• Published: 01/24/2008
• Est. Priority Date: 07/20/2006
• Status: Abandoned Application

• Current Assignee: Eastman Kodak Company
• Original Assignee: Eastman Kodak Company
• Inventors: Scalise, Anthony, Delong, Kevin R., Poetker, Robert B.
• Application Number:
  US11/489,965
• Publication Number:
  US 20080019574A1
• Time in Patent Office: Days
• US Class Current: 382/118
• CPC Class Codes: G06T 2207/20132 Image cropping
• CPC Class Codes: G06T 2207/30201 Face
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/194 involving foreground-background segmentation
• CPC Class Codes: G06T 7/90 Determination of colour characteristics
• CPC Class Codes: G06V 40/161 Detection; Localisation; Normalisation

Abstract

In a machine-controlled imaging method and system, an original digital image is displayed. During the displaying, a user input is accepted designating machine-controlled cropping of the original digital image. The machine-controlled cropping of the original digital image is performed without further user intervention. In that cropping, a subject portion and a background portion of the digital image are found. A retention area of the digital image is set. The retention area is fully inclusive of the subject portion and minimally inclusive of the background portion. When the retention area is larger than a preset percentage of the original digital image, the retention area is reset to a default crop. The digital image is reduced to the retention area as set or reset, respectively.

Claims

1. A machine-controlled imaging method comprising the steps of:

displaying an original digital image;

during said displaying, accepting a user input designating machine-controlled cropping of the original digital image; and

performing the machine-controlled cropping of the original digital image without further user intervention;

wherein said cropping further comprises;

finding a subject portion and a background portion of said digital image;

setting a rectangular retention area of said digital image, said retention area being fully inclusive of said subject portion and minimally inclusive of said background portion;

when said retention area is larger than a preset percentage of the original digital image, resetting said retention area to a default crop; and

reducing said digital image to the respective said retention area.

2. The method of claim 1 wherein said finding further comprises:

segmenting said digital image into a plurality of regions;

computing a subject map having a plurality of subject values, each of said subject values being associated with a respective one of said regions, each of said subject values indicating a likelihood that a respective said region is a subject of said digital image; and

determining said subject and background portions from said subject map.

3. The method of claim 2 wherein said resetting is independent of said map.

4. The method of claim 2 wherein said determining further comprises:

quantizing said regions of said subject map into high, medium, and low subject value regions; and

designating as subject portions said high subject value regions and ones of said medium subject value regions contiguous with one or more of said high subject value regions.

5. The method of claim 2 wherein said resetting is independent of said map.

6. The method of claim 2 wherein said computing further comprises locating face zones in said digital image, said face zones being included in said subject portion;

and said setting further comprises;

determining one or more face windows, each of said face windows being at least as large as and inclusive of a respective one or more of said face zones; and

sizing said retention area to be fully inclusive of said subject portion and all of said face windows and minimally inclusive of said background portion.

7. The method of claim 6 wherein said determining further comprises filtering said face zones.

8. The method of claim 1 wherein said resetting further comprises changes said retention area to a preset percentage of said original digital image.

9. The method of claim 8 wherein said preset percentage is 80 percent of said original digital image.

10. The method of claim 1 allowing manual cropping of said digital image independent of said machine-controlled cropping.

11. The method of claim 1 wherein said setting further comprises adjusting aspect ratio of said retention area to best fit said subject portion and said background portion.

12. A computer program product for machine-controlled cropping, the computer program product comprising computer readable storage medium having a computer program stored thereon for performing the steps of claim 1.

13. A machine-controlled imaging system comprising:

a circuit configured to display an original digital image;

a circuit configured to accept a user input during said displaying, said input designating machine-controlled cropping of the original digital image; and

a circuit configured to perform the machine-controlled cropping of the original digital image without further user intervention;

wherein said cropping includes;

finding a subject portion and a background portion of said digital image;

setting a retention area of said digital image, said retention area being fully inclusive of said subject portion and minimally inclusive of said background portion;

when said retention area includes the entirety of said digital image, resetting said retention area to exclude part of said subject portion; and

reducing said digital image to the respective said retention area.

14. The apparatus of claim 13 wherein said retention area is rectangular.

15. The apparatus of claim 14 wherein said setting includes adjusting aspect ratio of said retention area to best fit said subject portion and said background portion.

22. A computer program product for machine-controlled cropping, the computer program product comprising computer readable storage medium having a computer program stored thereon for performing the steps of claim 15.

16. A method for cropping a digital image, said digital image having an a plurality of pixels defining an initial size, said method comprising the steps of:

segmenting the digital image into a plurality of regions;

generating a main subject map having a plurality of main subject values, each of said main subject values being associated with a respective one of said regions, each of said main subject values indicating a likelihood that a respective one of said regions is a main subject of the digital image;

locating main subject regions and remaining regions of said map;

determining a rectangular area of said digital image fully inclusive of said regions of said main subject and minimally inclusive of said remaining regions;

when said rectangular area is said initial size, setting cropping boundaries to a default crop smaller than said initial size;

when said rectangular area is smaller than said default crop, setting the cropping boundaries to delimit said rectangular area; and

reducing the digital image to respective said cropping boundaries to provide a cropped image.

17. The method of claim 16 wherein, when said rectangular area is between said initial size and said default crop, setting said cropping boundaries to said default crop.

18. The method of claim 16 wherein said locating further comprises:

grouping said main subject values into high, medium, and low levels;

identifying a main subject inclusive of ones of said regions having said high levels and ones of said regions having said medium levels and being contiguous with one or more of said regions having said high levels, and a background having remaining said regions.

19. The method of claim 16 wherein said default crop is a preset percentage of said original digital image.

20. The method of claim 19 wherein said preset percentage is greater than 80 percent and less than 100 percent.

21. The method of claim 16 further comprising, prior to said segmenting:

displaying an original digital image;

during said displaying, accepting a user input designating machine-controlled cropping of the original digital image; and

responsive to said input, displaying said cropped image.

23. A system for cropping a digital image, said digital image having an a plurality of pixels defining an initial size, said method comprising:

a circuit configured to segment the digital image into a plurality of regions;

a circuit configured to generate a main subject map having a plurality of main subject values, each of said main subject values being associated with a respective one of said regions, each of said main subject values indicating a likelihood that a respective one of said regions is a main subject of the digital image;

a circuit configured to identify a set of main subject regions and a set of remaining regions of said digital image, said main subject regions each having a respective main subject value beyond a predetermined threshold, said remaining regions each having a respective main subject value within a predetermined threshold;

a circuit configured to determine a rectangular area of said digital image fully inclusive of said main subject regions and minimally inclusive of said remaining regions;

a circuit configured to set cropping boundaries to said rectangular area, when said rectangular area is smaller than a predetermined default crop size;

a circuit configured to set the cropping boundaries to said default crop size, when said rectangular area is between said default crop size and said initial size of the digital image; and

a circuit configured to reduce the digital image to respective said cropping boundaries to provide a cropped image.