Extracting Data From Images

• US 20090129625A1
• Filed: 11/21/2007
• Published: 05/21/2009
• Est. Priority Date: 11/21/2007
• Status: Active Grant

• Current Assignee: Seiko Epson Corporation (Seiko Group)
• Original Assignee: Seiko Epson Corporation (Seiko Group)
• Inventors: Zandifar, Ali, Liu, Che-Bin, Xiao, Jing
• Granted Patent: US 8,031,905 B2
• Time in Patent Office: Days
• US Class Current: 382/100
• CPC Class Codes: G06T 1/0028 Adaptive watermarking, e.g. Human Visual System [HVS]-based watermarking
• CPC Class Codes: G06T 1/005 Robust watermarking, e.g. average attack or collusion attack resistant
• CPC Class Codes: G06T 2201/0051 Embedding of the watermark in the spatial domain
• CPC Class Codes: G06T 2201/0065 Extraction of an embedded watermark; Reliable detection

Abstract

Disclosed are embodiments of systems and methods for embedding and/or extracting data from images. In embodiments, an image may be segmented into regions, and characters or other image groups within a segmented region may be determined to be embedding sites. A data vector may be embedded into a set of corresponding ordered embedding sites by representing each data element as different intensity values assigned to the pixels within one portion of an embedding site relative to the pixels in another portion of the embedding site. In embodiments, embedded data may be extracted from an image by extracting and decoding a set of bit values from a set of identified and ordered embedding sites.

Claims

1. A method for extracting data from an image comprising pixels, the method comprising:

identifying a set of ordered candidate embedding sites in the image wherein a candidate embedding site comprises a plurality of pixels;

extracting a set of data values from at least some of the set of ordered candidate embedding sites wherein a data value is extracted from an ordered candidate embedding site by examining a difference between a first intensity value of the pixels in a first portion of the ordered candidate embedding site and a second intensity value of the pixels in a second portion of the ordered candidate embedding site;

extracting at least one data vector from the set of data values by segmenting the set of data values; and

decoding the at least one data vector.

2. The method of claim 1 wherein the step of identifying a set of ordered candidate embedding sites in the image comprises:

segmenting the image into regions;

dilating pixels from one of the regions;

identifying a plurality of connected components from a set of pixels selected from the region with dilated pixels;

selecting a set of candidate embedding sites wherein a candidate embedding site comprises at least one connected component is selected from the plurality of connected components; and

assigning an order to the candidate embedding sites within the set of candidate embedding sites.

3. The method of claim 2 wherein selecting a set of candidate embedding sites wherein a candidate embedding site comprises at least one connected component is selected from the plurality of connected components further comprises:

selecting a candidate embedding site based upon at least one characteristic of the candidate embedding site.

4. The method of claim 3 wherein selecting a candidate embedding site based upon at least one characteristic of the candidate embedding site comprises:

responsive to a bounding box that bounds the candidate embedding site having a width that exceeds a first size threshold, selecting the candidate embedding site.

5. The method of claim 3 wherein selecting a candidate embedding site based upon at least one characteristic of the candidate embedding site comprises:

responsive to a bounding box that bounds the candidate embedding site having a height that exceeds a second size threshold, selecting the candidate embedding site.

6. The method of claim 2 wherein the step of assigning an order to the set of candidate embedding sites comprises using a position of a candidate embedding site within the image to assign an order to the candidate embedding site.

7. The method of claim 6 wherein using a position of a candidate embedding site within the image to assign an order to the candidate embedding site comprises:

performing a first direction sweep on the image to order a plurality of groups of candidate embedding sites; and

performing a second direction sweep on at least one group of candidate embedding sites within the plurality of groups to order a set of candidate embedding sites within the group.

8. The method of claim 1 wherein the step of extracting a set of data values from at least some of the set of ordered candidate embedding sites wherein a data value is extracted from an ordered candidate embedding site by examining a difference between a first intensity value of the pixels in a first portion of the ordered candidate embedding site and a second intensity value of the pixels in a second portion of the ordered candidate embedding site comprises:

responsive to a difference between the first intensity value and the second intensity value being greater than a threshold, extract an embed pattern by determining which intensity value is greater than the other intensity value; and

correlating the embed pattern with a data value.

9. The method of claim 1 wherein the step of extracting a set of data values from at least some of the set of ordered candidate embedding sites comprises:

identifying a first set of ordered candidate embedding sites in the image;

extracting a first set of data values from at least some of the first set of ordered candidate embedding sites;

rotating the image by a first angle of rotation;

identifying a second set of ordered candidate embedding sites in the rotated image;

extracting a second set of data values from at least some of the second set of ordered candidate embedding sites; and

determining which of the first and the second set of data values has a greater number of data values and selecting that set of data values as the set of data values.

10. The method of claim 1 wherein the at least one data vector is a bit vector, and decoding the at least one data vector comprises:

removing an indicator prefix from the bit vector; and

extracting the data by decoding the bit vector.

11. The method of claim 10 further comprising the step of applying an error-correcting code decoder to the bit vector.

12. A computer readable medium having instructions for performing the method of claim 1.

13. A system for extracting data from an image comprising pixels, the system comprising:

a message extractor, coupled to receive the image, that extracts at least one data vector from a set of ordered candidate embedding sites within the image responsive to a candidate embedding site containing a data value wherein a data value represents a difference between a first intensity value of the pixels in a first portion of the candidate embedding site and a second intensity value of the pixels in a second portion of the candidate embedding site; and

a decoder, coupled to receive the at least one data vector, that decodes the at least one data vector to extract the data.

14. The system of claim 13 wherein the at least one data vector is a bit vector, and decoding the at least one data vector comprises:

removing an indicator prefix from the bit vector; and

extracting the data by decoding the bit vector.

15. The system of claim 14 further comprising:

applying an error-correcting code decoder to the bit vector.

16. A system for extracting a data vector from a set of embedded data values within an image comprising pixels, the system comprising:

an embedding site selector, coupled to receive the image, that identifies a set of candidate embedding sites in the image wherein a candidate embedding site comprises a plurality of pixels;

an embedding site ordering processor, coupled to receive the set of candidate embedding sites, that orders the candidate embedding sites from the set of candidate embedding sites to form a set of ordered candidate embedding sites;

a bitstream extractor, coupled to receive the set of ordered candidate embedding sites, that extracts a set of data values from at least some of the set of ordered candidate embedding sites wherein a data value is extracted from an ordered candidate embedding site by examining a difference between a first intensity value of the pixels in a first portion of the ordered candidate embedding site and a second intensity value of the pixels in a second portion of the ordered candidate embedding site; and

a bitstream segmenter, coupled to receive the set of data values, that segments the set of data values to obtain at least one data vector.

17. The system of claim 16 wherein the embedding site selector identifies a set of ordered candidate embedding sites in the image by performing the steps comprises:

segmenting the image into regions;

dilating pixels from one of the regions;

identifying a plurality of connected components from a set of pixels selected from the region with dilated pixels;

selecting a set of candidate embedding sites wherein a candidate embedding site comprises at least one connected component is selected from the plurality of connected components; and

assigning an order to the candidate embedding sites within the set of candidate embedding sites.

18. The system of claim 17 wherein selecting a candidate embedding site from within the plurality of connected components further comprises:

selecting a candidate embedding site based upon at least one characteristic of the candidate embedding site.

19. The system of claim 18 wherein selecting a candidate embedding site based upon at least one characteristic of the candidate embedding site comprises:

responsive to a bounding box that bounds the candidate embedding site having a width that exceeds a first size threshold, selecting the candidate embedding site; and

responsive to a bounding box that bounds the candidate embedding site having a height that exceeds a second size threshold, selecting the candidate embedding site.

20. The system of claim 17 wherein assigning an order to the set of candidate embedding sites comprises using a position of a candidate embedding site within the image to assign an order to the candidate embedding site.

21. The system of claim 16 wherein extracting a set of data values from at least some of the set of ordered candidate embedding sites wherein a data value is extracted from an ordered candidate embedding site by examining a difference between a first intensity value of the pixels in a first portion of the ordered candidate embedding site and a second intensity value of the pixels in a second portion of the ordered candidate embedding site by performing the steps comprises:

responsive to a difference between the first intensity value and the second intensity value being greater than a threshold, extract an embed pattern by determining which intensity value is greater than the other intensity value; and

correlating the embed pattern with a data value.

22. The system of claim 21 wherein extracting a set of data values from at least some of the set of ordered candidate embedding sites by performing the steps comprises:

identifying a first set of ordered candidate embedding sites in the image;

extracting a first set of data values from at least some of the first set of ordered candidate embedding sites;

rotating the image by a first angle of rotation;

identifying a second set of ordered candidate embedding sites in the rotated image;

extracting a second set of data values from at least some of the second set of ordered candidate embedding sites; and

determining which of the first and the second set of data values has a greater number of data values and selecting that set of data values as the set of data values.