Method for automatically determining the acceptability of a digital image

• US 20060126093A1
• Filed: 12/09/2004
• Published: 06/15/2006
• Est. Priority Date: 12/09/2004
• Status: Active Grant

• Current Assignee: Monument Peak Ventures, LLC (Dominion Harbor Enterprises, LLC)
• Original Assignee: Eastman Kodak Company
• Inventors: Niederbaumer, James R., Fedorovskaya, Elena A., Claver, Lorrie
• Granted Patent: US 7,809,197 B2
• Time in Patent Office: Days
• US Class Current: 358/1.140
• CPC Class Codes: G06T 2207/30168 Image quality inspection
• CPC Class Codes: G06T 7/0002 Inspection of images, e.g. flaw detection
• CPC Class Codes: G06T 7/44 using image operators, e.g. filters, edge density metrics or local histograms
• CPC Class Codes: G06V 10/98 Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns
• CPC Class Codes: G06V 10/993 Evaluation of the quality of the acquired pattern
• CPC Class Codes: G06V 40/166 using acquisition arrangements

Abstract

A method for automatically determining the acceptability of an input image for a predetermined output operation, such as printing, includes the steps of: (a) identifying one or more important areas in the input image; (b) calculating a salient acceptability feature for each of the important areas; (c) determining the acceptability of the input image from the salient acceptability features of the important areas; and (d) implementing the predetermined output operation based on the acceptability of the input image determined in step (c).

Claims

1. A method for automatically determining the acceptability of an input image for a predetermined output operation, said method comprising the steps of:

(a) identifying one or more important areas in the input image;

(b) calculating a salient acceptability feature for each of the important areas;

(c) determining the acceptability of the input image from the salient acceptability features of the important areas; and

(d) implementing the predetermined output operation based on the acceptability of the input image determined in step (c).

2. The method as claimed in claim 1 wherein the one or more important areas are identified in step (a) by using a measure of edge occurrence to identify edge pixels in the image.

3. The method as claimed in claim 2 wherein the one or more important areas are identified in step (a) by identifying the areas with the greatest number of edge pixels.

4. The method as claimed in claim 2 wherein the one or more important areas that are identified in step (a) includes at least an area having an edge pixel with a maximum edge gradient.

5. The method as claimed in claim 2 wherein the one or more important areas that are identified in step (a) includes at least a central area of the input image.

6. The method as claimed in claim 1 wherein the one or more important areas are identified in step (a) by detecting one or more areas in the input image where the main subject is located.

7. The method as claimed in claim 1 wherein the one or more important areas are identified in step (a) by selecting one or areas in the input image based on coordinates provided by a human operator.

8. The method as claimed in claim 1 wherein the one or more important areas are identified in step (a) by detecting one or more areas in the input image that have the greatest number of eye fixations.

9. The method as claimed in claim 1 wherein the one or more important areas are identified in step (a) by detecting one or more areas in the input image that have a human figure.

10. The method as claimed in claim 1 wherein the one or more important areas identified in step (a) comprise at least two areas of the input image.

11. The method as claimed in claim 1 wherein the salient acceptability feature is a measure of edge occurrence.

12. The method as claimed in claim 11 wherein the measure of edge occurrence is an edge density.

13. The method as claimed in claim 1 wherein the salient acceptability measure is selected from the group comprising edge occurrences, gradient values, Laplacian values, edge height values, edge width values, noise values, contrast values, color rendering values, overexposure values, and underexposure values derived from the input image.

14. The method as claimed in claim 1 wherein the step (c) of determining the acceptability of the input image from the salient acceptability features of the important areas comprises the step of using a statistical measure of the salient acceptability features over all the important areas.

15. The method as claimed in claim 14 wherein the salient acceptability feature is an edge density for each of the important areas and the statistical measure of the salient acceptability features is the mean of the edge densities over all the important areas.

16. The method as claimed in claim 1 wherein the output operation implemented in step (d) is a printing operation, whereby an input image found to be unacceptable is not printed.

17. The method as claimed in claim 1 wherein the input image is obtained from a digital camera, steps (a) - (d) are performed in the digital camera, and the output operation implemented in step (d) is a deletion operation, whereby an input image found to be unacceptable is deleted.

18. The method as claimed in claim 17 additionally comprising the step of suggesting to a camera user that the image be deleted and obtaining the user'"'"'s approval for the deletion operation before deleting the image.

19. The method as claimed in claim 1 wherein the input image is obtained from a digital camera and the output operation implemented in step (d) is to instruct a camera operator to recapture the image.

20. The method as claimed in claim 1 wherein the output operation implemented in step (d) is an image processing operation, whereby an input image found to be unacceptable is submitted to the image processing operation.

21. The method as claimed in claim 1 wherein step (a) comprises the steps of:

detecting edge pixels that represent edges observed in the input image;

obtaining an edge map of the edge pixels;

dividing the edge map into a plurality of blocks, where each block may contain edge pixels from the edge map;

identifying two or more blocks having the largest number of edge pixels; and

selecting two or more important areas in the input image that correspond to the two or more blocks.

22. The method as claimed in claim 21 wherein step (b) comprises the step of calculating a salient acceptability feature based upon edge occurrence in each of the important areas.

23. The method as claimed in claim 22 wherein step (c) comprises the steps of calculating a statistical measure of edge occurrence from the salient acceptability features and using the statistical measure of edge occurrence to determine the acceptability of the digital input image for printing.

24. A method for automatically determining the acceptability of an input digital image for a printing operation, said method comprising the steps of:

(a) forming a low resolution analysis image from the input digital image;

(b) detecting edge pixels representing edges in the low resolution analysis image;

(c) obtaining an edge map of the edge pixels in the low resolution analysis image;

(d) dividing the edge map into a plurality of blocks, where each block may contain edge pixels from the edge map;

(e) identifying two or more blocks having the largest number of edge pixels;

(f) obtaining from the input image two or more important areas in the input image that correspond to the two or more blocks identified in step (e);

(g) calculating a salient acceptability feature based upon edge occurrence in each of the important areas;

(h) calculating a statistical measure of edge occurrence from the salient acceptability features calculated in step (g); and

(i) using the statistical measure of edge occurrence to determine the acceptability of the digital input image for printing.

25. The method as claimed in claim 24 wherein the step (b) of detecting edge pixels representing edges in the low resolution analysis image comprises the step employing an edge detection algorithm to detect edges in the low resolution image by using one or more edge detection threshold to identify the edge pixels.

26. The method as claimed in claim 25 wherein the edge detection algorithm is the Canny edge detection algorithm and employs two edge detection thresholds for strong edges and weaker edges associated with the strong edges.

27. The method as claimed in claim 24 wherein step (g) of calculating a salient acceptability feature based upon edge occurrence in each of the important areas comprises the step of determining the edge density in each of the important areas.

28. The method as claimed in claim 24 wherein step (h) of calculating a statistical measure of edge occurrence from the salient acceptability features calculated in step (g) comprises the step of determining a mean of the edge densities for the important areas.

29. A system for automatically determining the acceptability of an input image for a predetermined output operation, said system comprising:

an input stage for identifying one or more important areas in the input image;

a processing stage for (a) calculating a salient acceptability feature for each of the important areas and (b) determining the acceptability of the input image from the salient acceptability features of the important areas; and

an output stage for implementing the predetermined output operation based on the acceptability of the input image determined in the processing stage.

30. The system as claimed in claim 29 wherein the input stage includes algorithms for (a) detecting edge pixels that represent edges observed in the input image;

(b) obtaining an edge map of the edge pixels;

(c) dividing the edge map into a plurality of blocks;

(d) identifying two or more blocks having the largest number of edge pixels; and

(e) selecting two or more important areas in the input image that correspond to the two or more blocks.

31. The system as claimed in claim 29 wherein the processing stage includes algorithms for (a) calculating a salient acceptability feature based upon edge occurrence in each of the important areas and (b) calculating a statistical measure of edge occurrence from the salient acceptability features and using the statistical measure of edge occurrence to determine the acceptability of the digital input image for printing.