Information insertion method, information extraction method, and information extraction apparatus using dot-based information robust to geometric distortion

• US 9,710,872 B2
• Filed: 06/03/2015
• Issued: 07/18/2017
• Est. Priority Date: 01/28/2015
• Status: Active Grant

• Current Assignee: Electronics and Telecommunications Research Institute
• Original Assignee: Electronics and Telecommunications Research Institute
• Inventors: Ryu, Seung Jin, Kim, Min Sik, Yoon, Han Jun, Lee, Do Hoon
• Primary Examiner(s): Yentrapati, Avinash
• Application Number:
  US14/729,228
• Publication Number:
  US 20160217546A1
• Time in Patent Office: 776 Days
• CPC Class Codes: G06T 1/0064 Geometric transfor invariant watermarking, e.g. affine transform invariant
• CPC Class Codes: G06T 2201/0051 Embedding of the watermark in the spatial domain
• CPC Class Codes: H04N 1/0087 with hidden additional data, e.g. data invisible to the human eye
• CPC Class Codes: H04N 1/32203 Spatial or amplitude domain methods H04N1/32309 takes precedence

Abstract

An information insertion method, an information extraction method, and an information extraction apparatus using dot-based information are provided herein. The information extraction apparatus includes an identification unit and an extraction unit. The identification unit identifies a set of at least one dot output to a target. The extraction unit extracts information from the set of at least one dot.

Claims

1. An information extraction method, comprising:

identifying a set of at least one dot output to a target;

extracting information from the set of at least one dot; and

performing correction on geometric distortion of the set of at least one dot, wherein;

performing correction on the geometric distortion comprises performing correction on projection and rotation distortion of the set of at least one dot;

performing correction on the projection and rotation distortion comprises detecting four reference dots in a region of the set of at least one dot and performing correction on the projection and rotation distortion using the four reference dots; and

performing correction on the projection and rotation distortion comprises generating a set of horizontal lines and a set of vertical lines passing through the at least one dot, extracting two horizontal lines from the set of horizontal lines, and extracting two vertical lines from the set of vertical lines;

the four reference dots are intersections of the two horizontal lines and the two vertical lines;

the two horizontal lines comprise a horizontal line closest to an origin and a horizontal line farthest from the origin; and

the two vertical lines comprise a vertical line closest to the origin and a vertical line farthest from the origin.

2. The information extraction method of claim 1, wherein identifying the set of at least one dot comprises detecting the set of at least one dot within the target by analyzing data of the target.

3. The information extraction method of claim 1, wherein identifying the set of at least one dot comprises:

capturing an image of the target; and

detecting the set of at least one dot in the captured image.

4. The information extraction method of claim 3, further comprising emitting a predetermined type of light, which makes a substance, output to the target, perceptible, to the target.

5. The information extraction method of claim 1, wherein:

performing correction on the projection and rotation distortion comprises performing Hough transform on at least one central coordinate of the at least one dot; and

horizontal lines and vertical lines passing through the at least one dot are horizontal lines and vertical lines passing through the at least one central coordinate to which the Hough transform has been applied.

6. The information extraction method of claim 1, wherein generating the set of horizontal lines and the set of vertical lines comprises:

detecting at least one rectilinear line passing through the at least one dot;

detecting two angle ranges in which the rectilinear lines are maximally present; and

including the rectilinear lines, which correspond to one of the two angle ranges, in the set of horizontal lines or the set of vertical lines.

7. The information extraction method of claim 1, further comprising performing correction on size distortion of the set of at least one dot;

wherein performing the correction on the size distortion comprises;

generating a cumulative value in a horizontal and a cumulative value in a vertical direction by accumulating values of the central coordinates of the at least one dot in the horizontal direction and the vertical direction;

detecting a horizontal period and a vertical period, in which the at least one dot appears, by analyzing the cumulative value in the horizontal direction and the cumulative value in the vertical direction; and

correcting an image of the set of at least one dot so that the horizontal period and the vertical period match an original period.

8. The information extraction method of claim 7, wherein the horizontal period and the vertical period are detected by frequency domain analysis of the cumulative value in the horizontal direction and the cumulative value in the vertical direction.

9. The information extraction method of claim 1, wherein extracting the information comprises:

for each dot of the set of at least one dot, determining a dot pattern that corresponds to the each dot among a plurality of different dot patterns; and

using a value of a bit string of the determined dot pattern as at least part of the information.

10. An information extraction apparatus, comprising:

a hardware processor configured to identify a set of at least one dot output to a target,

to extract information from the set of at least one dot, and to perform correction on size distortion of the set of at least one dot by;

generating a cumulative value in a horizontal and a cumulative value in a vertical direction by accumulating values of the central coordinates of the at least one dot in the horizontal direction and the vertical direction;

detecting a horizontal period and a vertical period, in which the at least one dot appears, by analyzing the cumulative value in the horizontal direction and the cumulative value in the vertical direction; and

correcting an image of the set of at least one dot so that the horizontal period and the vertical period match an original period.