Artwork generated to convey digital messages, and methods/apparatuses for generating such artwork
2D machine readable symbologies are stylized and made aesthetically-appealing, facilitating their use to convey plural-symbol data on product packaging and other articles. In some arrangements, a symbology is mimicked as a collage of excerpts from a style image (e.g., a photograph or other artwork). To human viewers, the stylized symbology resembles the style image, rather than the 2D symbology. A great variety of other features and arrangements are also detailed.
1. A method of creating a stylized counterpart to a target image, the target image comprising a 2D machine-readable code, the target image having an attribute of being decodable by a compliant reader apparatus to produce a plural-symbol payload earlier encoded therein, the method comprising the acts:
(a) identifying component blocks of the target image, each component block having a respective location within said target image, wherein an array of P of said component blocks collectively forms a pattern that spans the target image, in which P comprises a positive integer;
(b) receiving one or more style images;
(c) performing a similarity comparison between a first of said component blocks of the target image, and plural candidate excerpts within the one or more style images, yielding a set of similarity values—each corresponding to a respective candidate excerpt within a style image;
(d) selecting one of said candidate excerpts, based at least in part on its corresponding similarity value;
(e) copying the selected candidate excerpt into a location in an output frame that corresponds to the location of the first component block within the target image; and
(f) repeating acts (c) through (e) for second through Pth component blocks of the target image;
wherein the output frame is filled with selected excerpts from one or more style images, yielding a stylized counterpart image having said attribute of being decodable by said compliant reader apparatus to produce said plural-symbol payload.
2. The method of claim 1 that further includes computing at least one edge similarity metric between a candidate excerpt within the one or more style images, and a previously-selected excerpt, wherein the selecting act is based in part on said edge similarity metric.
3. The method of claim 1 in which the stylized counterpart image evokes, to a majority of human viewers, a visual impression of the style image rather than a visual impression of the target image.
4. The method of claim 1 in which the target image comprises an artwork in which a digital watermark has been inserted.
5. The method of claim 1 in which at least one of said selected candidate excerpts is larger than the component block of the target image to which it corresponds.
6. The method of claim 1 in which all of the candidate excerpts are subsets of a single style image.
7. The method of claim 6 in which the single style image comprises an image of a QR code.
8. The method of claim 6 in which:
each of the candidate excerpts is I×J pixels in size, in which I and J are both positive integers;
the single style image is U×V pixels in size, in which U and V are both positive integers; and
act (c) includes performing a similarity comparison that compares the first of said component blocks of the target image with every different I by J pixel excerpt within the style image, yielding (U−I+1)*(V−J+1) similarity values in said set.
9. The method of claim 1 in which two of said candidate excerpts are taken from a single style image, and overlap in the style image, so said two candidate excerpts have a patch of plural pixels in common.
10. The method of claim 1 in which two of said selected excerpts are taken from a single image, and overlap, so said two selected excerpts have a patch of plural pixels in common.
11. The method of claim 1 in which a selected excerpt appears at two spaced-apart locations within the output frame.
12. The method of claim 1 in which a component block has a side dimension that is between 1-3% of a side dimension of the target image.
13. The method of claim 1 in which a component block has a side dimension that is between 3-7% of a side dimension of the target image.
14. The method of claim 1 in which a component block has a side dimension that is between 7-20% of a side dimension of the target image.
15. The method of claim 1 in which the similarity comparison of act (c) comprises a correlation operation.
16. The method of claim 15 in which act (c) includes performing a cross-correlation operation between the first of said component blocks and a style image, yielding said set of correlation values.
17. The method of claim 1 in which the similarity comparison of act (c) comprises evaluating a Euclidean distance measurement.
18. The method of claim 1 in which the similarity comparison of act (c) comprises evaluating a Euclidean distance measurement between oct-axis filtered counterparts of the target and style images. 19 - 35 . canceled.
19. The method of claim 1 that includes printing the stylized counterpart image on a plastic food container using dry offset printing.
20. A non-transitory computer readable medium comprising instructions stored thereon that, when executed by one or more processors of a computer system, cause the computer system to perform operations comprising:
(a) identifying component blocks of a target image, the target image comprising a 2D machine-readable code, the target image having an attribute of being decodable by a compliant reader apparatus to produce a plural-symbol payload earlier encoded therein, each component block having a respective location within said target image, wherein an array of P of said component blocks collectively forms a pattern that spans the target image, in which P comprises a positive integer;
(b) receiving one or more style images;
(c) performing a similarity comparison between a first of said component blocks of the target image, and plural candidate excerpts within the one or more style images, yielding a set of similarity values—each corresponding to a respective candidate excerpt within a style image;
(d) selecting one of the candidate excerpts, based at least in part on its corresponding similarity value;
(e) copying the selected candidate excerpt into a location in an output frame that corresponds to the location of the first component block within the target image; and
(f) repeating operations (c) through (e) for second through Pth component blocks of the target image;
wherein the output frame is filled with selected excerpts from one or more style images, yielding a stylized counterpart image of the target image, the stylized counterpart image having the attribute of being decodable by the compliant reader apparatus to produce the plural-symbol payload.
21. The non-transitory computer readable medium of claim 20 further including instructions that, when executed by the one or more processors of the computer system, cause the computer system to perform an operation comprising:
computing at least one edge similarity metric between a candidate excerpt within the one or more style images, and a previously-selected excerpt, wherein the selecting operation is based in part on said edge similarity metric.
22. The non-transitory computer readable medium of claim 20 in which the stylized counterpart image evokes, to a majority of human viewers, a visual impression of the style image rather than a visual impression of the target image.
23. The non-transitory computer readable medium of claim 20 in which the target image comprises an artwork in which a digital watermark has been inserted.
24. The non-transitory computer readable medium of claim 20 in which at least one of said selected candidate excerpts is larger than the component block of the target image to which it corresponds.
25. The non-transitory computer readable medium of claim 20 in which all of the candidate excerpts are subsets of a single style image.
26. The non-transitory computer readable medium of claim 25 , in which the single style image comprises an image of a QR code.
27. The non-transitory computer readable medium of claim 25 , in which:
each of the candidate excerpts is I×J pixels in size, in which I and J are both positive integers;
the single style image is U×V pixels in size, in which U and V are both positive integers; and
operation (c) includes performing a similarity comparison that compares the first of said component blocks of the target image with every different I by J pixel excerpt within the style image, yielding (U−I+1) * (V−J+1) similarity values in said set.
28. The non-transitory computer readable medium of claim 20 , in which two of said candidate excerpts are taken from a single style image, and overlap in the style image, so said two candidate excerpts have a patch of plural pixels in common.
29. The non-transitory computer readable medium of claim 20 , in which two of said selected excerpts are taken from a single image, and overlap, so said two selected excerpts have a patch of plural pixels in common.
30. The non-transitory computer readable medium of claim 20 , in which a selected excerpt appears at two spaced-apart locations within the output frame.
31. The non-transitory computer readable medium of claim 20 , in which a component block has a side dimension that is between 1-3% of a side dimension of the target image.
32. The non-transitory computer readable medium of claim 20 , in which a component block has a side dimension that is between 3-7% of a side dimension of the target image.
33. The non-transitory computer readable medium of claim 20 , in which a component block has a side dimension that is between 7-20% of a side dimension of the target image.
34. The non-transitory computer readable medium of claim 20 , in which the similarity comparison of operation (c) comprises a correlation operation.
35. The non-transitory computer readable medium of claim 34 , in which operation (c) includes performing a cross-correlation operation between the first of said component blocks and a style image, yielding said set of correlation values.
36. The non-transitory computer readable medium of claim 20 , in which the similarity comparison of operation (c) comprises evaluating a Euclidean distance measurement.
37. The non-transitory computer readable medium of claim 20 , in which the similarity comparison of operation (c) comprises evaluating a Euclidean distance measurement between oct-axis filtered counterparts of the target and style images.
38. An apparatus comprising:
memory storing: i) a target image comprising a 2D machine-readable code, the target image having an attribute of being decodable by a compliant reader apparatus to produce a plural-symbol payload earlier encoded therein, each component block having a respective location within said target image, and ii) one or more style images;
(a) means for identifying component blocks of a target image, wherein an array of P of said component blocks collectively forms a pattern that spans the target image, in which P comprises a positive integer;
(b) means for performing a similarity comparison between a first of the component blocks of the target image, and plural candidate excerpts within the one or more style images, yielding a set of similarity values—each corresponding to a respective candidate excerpt within a style image;
(c) means for selecting one of the candidate excerpts, based at least in part on its corresponding similarity value;
(d) means for copying the selected candidate excerpt into a location in an output frame that corresponds to the location of the first component block within the target image; and
(e) means for re-executing the means of (b) through (d) for second through Pth component blocks of the target image;
means for filling the output frame with selected excerpts from one or more style images, yielding a stylized counterpart image of the target image, the stylized counterpart image having the attribute of being decodable by the compliant reader apparatus to produce the plural-symbol payload.
39. The apparatus of claim 38 further comprising: means for determining at least one edge similarity metric between a candidate excerpt within the one or more style images, and a previously-selected excerpt, wherein said means for selecting utilizes the at least one edge similarity metric.
40. The apparatus of claim 38 , in which:
each of the candidate excerpts is I×J pixels in size, in which I and J are both positive integers;
the single style image is U×V pixels in size, in which U and V are both positive integers; and
said means for performing a similarity comparison utilizes a similarity comparison that compares the first of the component blocks of the target image with every different I by J pixel excerpt within the style image, yielding (U−I+1) * (V−J+1) similarity values in the set of similarity values.
41. The apparatus of claim 38 , in which two of the candidate excerpts are taken from a single style image, and overlap in the style image, so said two candidate excerpts have a patch of plural pixels in common.
42. The apparatus of claim 38 , in which two of the selected excerpts are taken from a single image, and overlap, so the two selected excerpts have a patch of plural pixels in common.
43. The system of claim 38 in which means for performing a similarity comparison (b) executes a correlation operation.
44. The system of claim 43 in which means for performing a similarity comparison (b) executes a cross-correlation operation between the first of the component blocks and a style image, yielding the set of correlation values.
45. The system of claim 38 in which means for performing a similarity comparison (b) executes an evaluation a Euclidean distance measurement between oct-axis filtered counterparts of the target image and the style image.