KR20100105865A

KR20100105865A - Information encoding method for two-dimensional bar code subjected to wavelet transformation

Info

Publication number: KR20100105865A
Application number: KR1020107016930A
Authority: KR
Inventors: 마사키 무라나카
Original assignee: 자크 가부시키가이샤
Priority date: 2008-01-09
Filing date: 2008-01-09
Publication date: 2010-09-30
Also published as: TW200941353A; WO2009087764A1; JPWO2009087764A1

Abstract

By embedding data on the lattice dots, an information coding method of a two-dimensional bar code is provided without expanding the arrangement shape of the lattice dots. In the Harr-converted space, the embedded data, a marker indicating the position of the data, and noise for masking the data are placed (steps 10 to 40), and the Harr-converted space in which the data and noise are arranged is inversely Harr-converted (step) 50) The obtained numerical data is embedded on the lattice of blocks as binary image data (step 70).

Description

Information encoding method of wavelet transformed 2D barcode {INFORMATION ENCODING METHOD FOR TWO-DIMENSIONAL BAR CODE SUBJECTED TO WAVELET TRANSFORMATION}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information encoding method of a two-dimensional barcode in which data is embedded on a lattice dot of a block of a medium face. Specifically, the information of a two-dimensional barcode in which binary image data to be embedded is created by wavelet transformation. It relates to a coding method.

There has been proposed a method of describing information for reading and converting barcodes printed on printed matter and obtaining information. This printed barcode method has a problem in that it is difficult to arrange a plurality of barcodes in terms of layout because barcodes that are obstacles to printed matter are large and occupy a part of the ground.

In order to solve these problems, there has been proposed a method of describing information in which a dot pattern in which fine dots are arranged based on a predetermined law is printed and then read out as digital image data by a camera to obtain information. The printed fine dot pattern is a method of correcting the dot pattern photographed by imaging and stretching of print media due to lens distortion and tilting, curvature of the surface, and distortion during printing. There was a problem that technology was needed.

In order to solve these problems, Patent Document 1 defines a large amount of data in a dot pattern by assigning different functions to each dot of the dot pattern, and recognizes the directionality at the time of informatization from the dot pattern to quickly informatize the information. The information input / output method using the dot pattern which can simultaneously check the error of the arrangement | positioning of a dot, and can improve security is described.

Patent Document 1: Japanese Patent No. 3706385

Specifically, a block in which a plurality of lattice dots are arranged in a rectangular shape on a medium surface such as a printed matter is continuously arranged in the vertical direction and / or the left and right directions, and at least one lattice dot at a predetermined position in the block is originally stored. The distance from the virtual point to the end point represented by the vector, with the key dot representing the direction of the block, shifted in a predetermined direction rather than the position of the grid dot, as a virtual point, the center of which is surrounded by four grid dots as the starting point. A dot pattern is generated by arranging a plurality of information dots defining information in the and directions, and a block constituting the dot pattern is read as image data by a camera, and information and a program are outputted from numerical values obtained by digitizing it.

According to this, at least one lattice dot at a predetermined position in the block is moved in a predetermined direction rather than the position of the original lattice dot and used as a key dot indicating the direction of the block. As a result, since the key dots are arranged in the lattice of the original lattice dots, there is a problem that printing requires a higher resolution than the resolution necessary for the arrangement of the original lattice dots. In addition, there is a problem that the arrangement shape of the original lattice dots becomes large when the maximum resolution is assigned and printed with the resolution of the key dots.

SUMMARY OF THE INVENTION The present invention has been made to solve such a problem, and an object thereof is to provide a method for encoding information of a two-dimensional barcode in which the arrangement shape of the grid dots is not expanded by embedding data on the grid dots.

The information encoding method of the wavelet-converted two-dimensional barcode of the present invention is the information encoding method of the wavelet-converted two-dimensional barcode embedded in lattice dots of a medium plane in which blocks in which lattice dots are arranged in a rectangular shape are continuously arranged. In the Harr transformed space, the embedded data and the noise for masking the data are arranged, Harr inverse transform is performed on the Harr transformed space in which the data and the noise are disposed, and the obtained numerical data is obtained as binary image data. It is characterized in that it is embedded in the phase.

The information encoding method of the wavelet-converted two-dimensional bar code according to the present invention is characterized by arranging a marker indicating the presence of data to be embedded in the Harr transformed space.

The information encoding method of the wavelet transformed two-dimensional bar code according to the present invention is characterized in that the embedded data and the marker are L components, and the noise is H components.

The information encoding method of the wavelet transformed two-dimensional barcode according to the present invention is characterized in that the H component of the encoding is determined so as to be decoded as binary data at the time of decoding the binary image data.

According to the present invention, it is possible to embed binary image data obtained by arranging data in a Harr transformed space and further performing Harr inverse transformation on a lattice dot in units of blocks. Therefore, it is possible to create a block without increasing the arrangement shape of the grid dots, and to provide a two-dimensional barcode information coding method that can expand information by printing the created block in a lattice pattern on the medium surface. can do.

1 is a flowchart showing a procedure of an encoding process according to the present invention.
Fig. 2 is a bit block diagram showing the arrangement of information bit strings according to the present invention.
3 is an L and H component block diagram showing the configuration of the L component and the H component according to the present invention.
Fig. 4 is a marker arrangement diagram showing the arrangement of the markers of the L and H components according to the present invention.
5 is a numerical data diagram in which Harr inverse transformation is performed on four bits of data of the first row of the present invention.
Fig. 6 is a printing pattern showing a state in which one block is printed on the medium according to the present invention.
7 is a flowchart showing the procedure of the decoding process according to the present invention.
8 is a decoding processing diagram showing steps of a decoding process according to the present invention.

EMBODIMENT OF THE INVENTION Embodiment of the information encoding method of the 2D barcode by this invention is described with reference to drawings. Although the number of bits of embedded data can be arbitrarily set, the bit length of the data and the size of the matrix after encoding must be determined in advance so as not to cause contradiction in the encoding and decoding processes. In addition, taking into account the possibility that rotation transformation is performed from encoding to decoding, the lengths of the rows and columns after encoding are set equal. For this reason, when the data of the information to be embedded is 32 bits, the bit string to be embedded is 40 bits, and since the data of the information to be embedded is 32 bits, the excess 8 bits are used as checksum bits for error detection.

The encoding and decoding thereof in this case will be described below. 1 is a flowchart showing a procedure of an encoding process according to the present invention. In Fig. 1, an 8-bit checksum is first added to the 32-bit bit length of data (step 10). Next, these 40 bits are decomposed into bit strings of 4 bits each (step 20). Fig. 2 is a bit block diagram showing a bit string of information according to the present invention. The bit string structure in step 10, and the initial steps of decomposing 4 bits in step 20 to form a bitmap of 10 rows and 10 columns are shown. have.

Next, since each of these 4-bit data is embedded in a Harr transformed space, which is a one-dimensional wavelet transform, it is arranged as an L component which is a low frequency component whose numerical value after the Harr inverse transform is 2 or 0, and an H component which is a high frequency component (step 30). . The L component contains a bit string and a checksum which are information components of data. The H component is a high frequency component, that is, noise, for adding difficulty toxicity or irregularity to the information component. In addition, a marker M for detecting the head of the bit string is arranged in one bit LM in the L component and one bit HM in the H component.

3 is an L and H component block diagram showing the configuration of the L component and the H component of the present invention. In Fig. 3A, when the bit of the corresponding data is 1, the L component is predetermined and arranged so that the H component is 0. In addition, in FIG. 3B, when the bit of the corresponding data is 0, any one of (1, 1) and (1, -1), which are pairs of L and H components, is determined and arranged to be set randomly. In this case, (1, -1) is selected for 0 of the 2nd bit, and (1, 1) is selected for 0 of the 4th bit. In FIG. 3C, as a result of each bit passing through these processes, the L component of the data and the markers (1, 0, 1, 0, M) becomes (2, 1, 2, 1, LM), and the H component is ( 0, -1, 0, 1, HM).

Next, if the number of 4-bit rows of data is prime, each marker is determined in advance so that LM = 2 and HM = 0, and if not, LM = 0 and HM = 0, and the numerical value of each marker Is placed (step 40). 4 is a marker arrangement diagram showing arrangement of markers of L and H components. For 4 bits of data in the first row, since 1 is not a prime number, the marker LM of the L component and the marker HM of the H component are both zero. Since four bits of data of the second row are decimal rows, the marker LM of the L component is 2, and the marker HM of the H component is 0. Thus, if up to 10 rows are decimal, (2, 0), otherwise (0, 0) is arranged at each (LM, HM).

After the bitmaps of the 10 rows and 10 columns created in this manner are embedded in the Harr transformed space, Harr inverse transformation is performed one row at a time, and when the inverse transformation to the tenth row is completed, the following steps are executed (steps 50 and 60). 5 is a numerical data diagram in which Harr inverse transformation is performed on four bits of data of a first row. In FIG. 5, numerical data (2, 1, 2, 1, 0, 0, -1, 0, -1, 0) in which the raw data (1, 0, 1, 0) and the marker are Harr transformed are Harr inverse transformed. , Numerical data (1, 1, 0, 1, 1, 1, 1, 0, 0, 0) is shown.

Next, the numerical data obtained by the Harr inverse transformation is embedded on a lattice of 10 x 10 dots on the medium surface as binary image data in a two-dimensional array (step 70). Using this 10 x 10 dot grid as one block, bits of predetermined information are embedded, and the blocks are arranged in a rectangular shape to expand the embedding of information. 6 is a printing pattern showing a state in which one block is printed on the medium. The first first row shows a pattern in which binary image data of original data (1, 0, 1, 0) is printed on the medium side. The print pattern of arbitrary raw data is shown as an image below the 2nd row, and the state printed on the grid of 10x10 dots as a whole is shown.

7 is a flowchart showing the procedure of the decoding process according to the present invention. In Fig. 7, the printed pattern detected in the area of the block from the medium surface is repeatedly subjected to Harr transformation while performing rotation processing or rotation to detect the pattern of the marker (steps 110 to 130). When the pattern of the marker is detected, the bits of the H component and the marker bits, which are noise, and the marker bits in the L component are discarded, and only the information bits of the L component are extracted (step 140). On the extracted information bits, bit inverse transformation is performed by applying a predetermined conversion rule to embed the original data in the Harr transformed space, and the original data is reproduced (step 150).

8 is a decoding processing diagram showing a step of decoding processing. 8A shows a print pattern of binary image data of original data (1, 0, 1, 0). Fig. 8B shows a bit pattern in which image-processed print patterns are image processed and quantized to reproduce Harr inverse transformed numerical data (1, 1, 0, 1, 1, 1, 1, 0, 0, 0). . Fig. 8C shows the bit pattern (2, 1, 2, 1, 0, 0, -1, 0, 1, 0) of Harr transformed original data when the position of the block is determined by marker detection. FIG. 8D shows the original data (1, 0, 1, 0) reproduced by applying a predetermined transformation rule and performing bit inverse transformation to embed the original data in the Harr transformed space.

As described above, according to the present invention, the binary image data obtained by arranging data having a position of data, hard readability and irregularity in a Harr transformed space, and performing inverse Harr conversion is further determined in units of blocks. It becomes possible to embed it on the lattice dot to make. Therefore, it is possible to create a block without increasing the arrangement shape of the lattice dots, and a two-dimensional barcode information coding method capable of expanding information by printing the created block in a lattice form on the medium surface. Can provide.

M marker
LM: Marker of L component
HM: Marker of H component

Claims

A information encoding method of wavelet-converted two-dimensional barcodes embedded in a lattice dot of a medium plane in which blocks in which lattice dots are arranged in a rectangular shape are continuously arranged,
In the Harr transformed space, the embedded data and noise to mask the data are placed.
Harr inverse transforms the Harr transformed space in which the data and the noise are arranged,
The information encoding method of the wavelet transformed two-dimensional barcode which embeds the obtained numerical data as binary image data on the said grating | lattice of the said block.

The method according to claim 1,
And a marker indicative of the presence of the embedded data in the Harr transformed space.

The method according to claim 2,
The information encoding method of the wavelet transformed two-dimensional barcode which makes the said embedded data and the said marker the L component, and the said noise the H component.

The method according to claim 3,
And the H component of the encoding is determined such that the binary image data is decoded as binary data.