Method for constructing a composite image

• US 8,068,693 B2
• Filed: 07/18/2007
• Issued: 11/29/2011
• Est. Priority Date: 07/18/2007
• Status: Expired due to Fees

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Sorek, Noam, Bregman-Amitai, Orna
• Primary Examiner(s): Kassa; Yosef
• Application Number:
  US11/826,772
• Publication Number:
  US 20090022422A1
• Time in Patent Office: 1,595 Days
• Field of Search: 382/235, 382/243, 382/284, 382/282, 358/540, 358/450
• US Class Current: 382/284
• CPC Class Codes: G06T 3/4038 Image mosaicing, e.g. composing plane images from plane sub-images
• CPC Class Codes: G06T 3/4084 in the transform domain, e.g. fast Fourier transform [FFT] domain scaling
• CPC Class Codes: H04N 1/3876 Recombination of partial images to recreate the original image
• CPC Class Codes: H04N 23/698 for achieving an enlarged field of view, e.g. panoramic image capture

Abstract

A method for constructing a composite image comprises determining in an alignment between at least a first and a second image based on the content of the first and second image frames, and combining in a transform domain at least the first and second images with the determined alignment to form a composite image.

Claims

1. A computerized method for constructing a composite image using a processor, the method comprising:

determining in an alignment between at least a first and a second image based on the content of the first and second image frames; and

combining in a discrete cosine transform (DCT) domain at least the first and second images with the determined alignment to form a composite image using a processor.

2. The method according to claim 1 wherein the composite image is a panoramic image.

3. The method according to claim 2 wherein the panoramic image is a substantially seamless panoramic image.

4. The method according to claim 2 wherein the panoramic image is a mosaic panoramic image frame.

5. The method according to claim 1 wherein determining an alignment between at least a first and a second image is performed in the image domain.

6. The method according to claim 5 wherein the alignment is determined based on a DC portion of the DCT domain.

7. The method according to claim 5 comprising determining in the DCT domain, the intensity level of one of the first and second image frames.

8. The method according to claim 1 wherein the alignment includes a horizontal or vertical shift of the second image frame with respect to the first image frame.

9. The method according to claim 8 wherein the horizontal or vertical shift is divided into block shifts and residual shifts.

10. The method according to claim 9 wherein the residual shifts are performed in an image domain prior to transforming the second image to the DCT domain.

11. The method according to claim 9 wherein the block shifts are performed in the DCT domain.

12. The method according to claim 1 comprising:

detecting in the image domain a location of a common interesting feature in each of the first and second image frame; and

determining the alignment based on the location of the interesting feature in each of the first and second image frames.

13. The method according to claim 1 comprising:

performing registration between preview image frames corresponding to the first and second image frames; and

determining the alignment between the first and second image frames based on the registration.

14. The method according to claim 1 comprising determining alignment between the first and second image frames based on alignment with at least one image frame captured between the first and second image frames.

15. The method according to claim 14 wherein the at least one image frame is a preview image frame.

16. The method according to claim 1 comprising recording a size of the composite image based on the alignment.

17. The method according to claim 1 comprising performing geometrical transformation of the first and second image frame.

18. The method according to claim 14 wherein the geometrical transformation is a cylindrical transformation.

19. The method according to claim 1 wherein at least one of the first or second image frames is a full snapshot image frame.

20. The method according to claim 1 comprising saving the first and second image frame in a JPEG format.

21. The method according to claim 1 comprising:

defining an empty image frame in the DCT domain; and

positioning the first and second image frame onto the empty image frame based on the alignment.

22. The method according to claim 1 comprising:

defining a header file with memory allocation; and

adding a stream of image data blocks from the first and second image frame to the header file.

23. The method according to claim 1 comprising applying a mask in the DCT domain to combine the first and second image frame.

24. The method according to claim 23 wherein the mask is a binary mask.

25. The method according to claim 23 wherein the mask is a continuous mask.

26. The method according to claim 23 wherein the mask is a low pass version of a binary mask.

27. The method according to claim 23 wherein the mask is to define a stitching line to combine the first and second image frame in an overlapping region.

28. The method according to claim 27 wherein the stitching line is a curved stitching line.

29. The method according to claim 1 comprising:

detecting a region including a pre-defined pattern; and

defining a stitching line that circumvents the region.

30. The method according to claim 29 wherein the pre-defined pattern is a face.

31. The method according to claim 1 comprising applying a mask configured to correct image intensity differences between the first and second image frame.

32. The method according to claim 31 wherein the mask is configured to correct for vignettes.

33. The method according to claim 1 comprising combining in the DCT domain a decorative background with the composite image.

34. The method according to claim 1 comprising combining in the DCT domain text with the composite image.

35. The method according to claim 1 comprising combining in the DCT domain graphics with the composite image.

36. The method according to claim 1 comprising combining decorative frames with the composite image.

37. The method according to claim 1 comprising:

capturing a plurality of image frames; and

selecting in the image domain, image frames from the plurality of image frames to be included in the composite image.

38. The method according to claim 37 wherein the selecting is based on a detected intensity level of the image frames.

39. The method according to claim 37 wherein the selecting is based on a detected sharpness level of the image frames.

40. The method according to claim 37 wherein the selecting is based on the alignment between the image frames.

41. The method according to claim 1 wherein the combining is performed in a vertical direction.

42. The method according to claim 1 wherein determining an alignment between at least a first and a second image is performed in the DCT domain.