TWI286706B - Two-dimensional barcode card and the decoding method - Google Patents

Abstract

The present invention discloses a two-dimensional barcode card, which uses the positioning hole as the identification mark, and can be scanning accompanying with decoding, and with large information capacity, and simple decoding algorithm. The card includes the carrier and the code map of the two-dimensional barcode diagram attached on the carrier using matrix cells to store the information. The code map includes the upper boundary, the left and right boundaries having positioning holes, and the bit stream area. Moreover, the present invention discloses the decoding method for the code map, which is first to calculate the position and gradient of the upper boundary; then, estimating and calculating the central coordinates of the first pair of positioning holes based on the above calculation to obtain the central coordinates of the matrix cells; reading the bit stream information from the image; estimating based on the central coordinates of the previous pair of positioning holes and calculating the central coordinates of the next pair of positioning holes; reading one bit from the bit stream information for each matrix cell to pass the central coordinates for the upper and lower two matrix cells and calculate the coordinates of all matrix cells between two columns, and simultaneously reading the corresponding bit stream information.

Description

1286706 V. INSTRUCTION DESCRIPTION (1) <Technical Field> The present invention relates to a method for storing two codes of codes that can be attached to various carriers. ', Maca' and the re-continuation of the two-dimensional strip &lt;Prior&gt; The two-dimensional barcode is a type that can be scanned and can be used to store various encoded 2 S 3 ί-shaped symbols on the carrier such as paper. And other information - two, r news, can be attached to the carrier with low cost, easy to read and other J-style comparison, two-dimensional bar code commonly used two-dimensional bar code encoding method PDF4 °. The QR code generated by the QR code method (the j code mode and the QR code side arrangement of the black and white matrix cells store the information cardamom code, using three "back" type symbols on the corner of the matrix, (such as the first 4 inches) The sign of distribution is small in the probability of pattern recognition, and the figure is used for storage, as shown in the figure. The application of QR matrix code error to obtain image by digital camera is two = shell, large area, suitable for 2D barcode recognition algorithm ^琢琢^ _ However, the application of this encoding method to the image must be recognized in the whole way by scanning. [After the image is scanned, the PDF417 encoding method and the QR encoding code can be used. The adjacent black bars and white bars are different in type. PDF417 is called bar-space ratio code. It and the scale value store information. The black and white interval lines with fixed space ratio are characterized by two graphics. The second side of the PDF417 encoding method is shown in Figure 2. The rate, feature recognition algorithm is simple, the right has the smallest misrecognition, and there are a large number of mismatches. (Theoretically, 5 pages, 5, invention Description (2), can tolerate 5 0% image breakage), ^What distortion has the greatest tolerance, can be graphical linear and non-linear applications, can be in the scanned image = the quality of the image obtained in various images (print resolution) * The decoding device performs the decoding; the recognition device is costly. The storage capacity is relatively::, <Description of the Invention> The object of the present invention is to provide a storage-small--and a large amount of recognition and a large amount of recognition operation while scanning. Another object of the invention is to provide a method for information in a card. The method for interpreting the information in the card is to provide a method for storing the information on the two-dimensional bar code ί, using a matrix lattice to store information. And attached to: the code map includes an upper boundary, a left code code map, characterized by a jblt stream) region; an upper boundary is a boundary, and a matrix row of the bit stream =, the left and right boundaries are sub-picture edges The paper feed direction is the front=, the matrix column of the normal positioning hole, and is located at the outermost and upper boundaries on both sides of the code image. &amp; is the two-dimensional barcode card of the present invention located at the left and right boundaries, and the present invention can adopt a S$ U匕The following characteristics are included: · 乍 is the upper political circle, and then three rows of rectangular holes, one row of black matrix I, two sides of the matrix of the two columns are one ^ two left boundary and the combination of the right black matrix lattice and the white matrix 二袼, and the middle column is the side length of the matrix lattice R2/R1*L &gt;= 4 Busy is selected according to the following formula: 1286706 V. Invention Description (3) where R 1 is when the code image is printed on the carrier Resolution: R 2 is the resolution of the image sensor when reading the code picture; L is the design side length of the matrix. The bit stream area of the present invention may include a plurality of information blocks, each of which includes multiple Matrix rows and error correction codes. In the bit stream area, the information can be discharged in a manner that is dominated by behavior, from top to bottom; or it can be column-based, from left to right or from right to left.

There is a blank area to the left of the left border of the code map and to the right of the right border. The invention also discloses a two-dimensional barcode decoding method, which recovers a two-dimensional barcode code image stored on a carrier and uses the matrix lattice to store information into binary information, which comprises the following steps: A. The upper boundary of the sounding code map, Calculate and save the slope of the upper boundary B · the left and right boundaries of the probing pattern, and estimate the area of the first pair of positioning holes according to the inclination of the upper boundary; C · count the number of white pixels in the estimated area, Calculating a center coordinate of the first pair of positioning holes;

D · Calculate the central coordinate of the matrix lattice between the first pair of positioning holes, which is spaced apart by a matrix on one line, and read each matrix 1 bit from the image according to the central coordinate (bit Bit stream information; E · Estimate the area of the next pair of positioning holes according to the center coordinates of the previous pair of positioning holes, and the side length of the matrix, and calculate the center coordinates of the positioning holes from the detection, if the positioning holes Center coordinates

Page 7 1286706 V. Description of invention (4) If the detection is unsuccessful, the code map is interpreted, otherwise it will enter the next step; F · Calculate according to the center coordinates of the pair of positioning holes, and save the center coordinates of the matrix between them, and Reading the bit stream information of the corresponding position according to the central coordinate; G · calculating the coordinates of all the matrix cells between the two rows according to the central coordinates of the saved two rows of matrix cells, and reading the bit stream of the corresponding matrix lattice News, Η · Go back to step E. The two-dimensional barcode decoding method of the present invention may further include the following features: In step C, when the number of white pixels is greater than (Ν - 1 ) * (Μ - 1 ), less than (Ν + 1 ) * ( Μ + 1 ) When it is considered, there is an effective positioning hole in the area, where Ν and Μ are the lengths of the two sides of the matrix. In step D, the total line number information in the code map is read and the decoding is ended based on the total number of lines. Each time (S + Ε ) of the byte information is read, an error correction operation is performed, and the correct information of the S bits of the error correction success is output. The two-dimensional barcode card of the present invention uses a simple matrix code to store various binary information. Compared with the two-dimensional barcode generated by the QR encoding method, the recognition algorithm is simple; the decoding can be performed while scanning. The simple matrix code uses the positioning hole as the identification mark, and the area of the information stored under the same area is larger than that of the PDF41 7, and the storage capacity of the print carrier of the same area is improved. The decoding algorithm of the simple matrix code of the present invention can be decoded at the same time of scanning, and is made by positioning holes. The positioning mark of the graphic recognition has a fast recognition speed and a simple recognition algorithm.

1286706 V. DESCRIPTION OF THE INVENTION (5) &lt;Embodiment&gt; The scanner using CIS (Contact Image Sensor) has uniform illumination, and the acquired image has less nonlinear distortion than the original image, and the acquired image is obtained. The image quality is good, so a simple matrix code with a structure that is minimally simplistic can be designed on the two-digit barcode card, thereby increasing the storage capacity of the print carrier of the same area. The two-dimensional barcode card of the present invention comprises a carrier and a code map generated by a simple matrix code encoding method attached to the carrier (as shown in FIG. 3), the boundary contour of the code image is a rectangle, and the arrow in the figure indicates scanning. When reading the two-dimensional barcode card, the paper feed direction is set, and the matrix unit parallel to the paper feed direction constitutes a code map column, and the matrix unit perpendicular to the paper feed direction constitutes a code map row. The entire code map is composed of equal-sized matrix lattices, including an upper boundary 10, a left boundary 12, a right boundary 14, a positioning hole 20, and a bit stream region 16. The upper boundary 10 is a row of black matrix lattices. The left boundary 1 2 and the right boundary 1 4 are both composed of three columns of matrix lattices, and the two columns of the matrix lattice are respectively arranged in a row of black matrix grids, and the middle column is arranged by black and white matrix grids, wherein the white matrix grid is the positioning hole. 20. Below the upper boundary 10, between the left boundary 12 and the right boundary 14, is a bit stream area 16. The bit stream area 16 is used for storing valid information, and the boundary lines and columns are used for positioning when the image is read. . The matrix column with the positioning holes 20 may be arranged by a black matrix grid and a white matrix grid (as shown in FIG. 3), or may be arranged by a plurality of black matrix grids and a white matrix grid. It can be arranged by the black matrix grid and the white matrix grid at other combinations, which are all disclosed and protected by the present invention.

1286706 V. Description of invention (6) The matrix lattice of the bit stream area 16 is a storage unit of information, and each matrix lattice stores information of one bit (bit), for example, a black matrix lattice stores π 0 π, and the white matrix lattice stores " 1 " . The information in the bit stream area 16 can be discharged by means of behavior, from top to bottom, or from left to right or from right to left. The code picture includes an integer number of matrix lattices in the length direction and the width direction. The matrix lattice may be a square having the same number of pixels in the direction of the row and the direction of the column, or may be in the direction of the row and the direction of the column. For a rectangular grid with different numbers of pixels, the number of matrix cells can be determined according to the size of the attached carrier, and the length and width of the bit stream region are thus grafted. The error correction code can be added in the bit stream area as needed, and the error correction code can be designed according to the prior art. As described above, the nonlinear distortion of the image acquired by the CIS is small, that is, the relative position of the internal structure of the graphic is substantially fixed, so that accurately positioning the matrix image from the image obtained by scanning becomes the only important task of the reading matrix. The left and right positioning holes 20 located on one horizontal line on both sides of the code pattern are designed for this purpose. When a pair of positioning holes (one on each side) are recognized, the center position of all other matrix cells of the row can be accurately calculated and saved, so that the bit stream information can be directly obtained; The position of the hole can estimate and calculate the position of the pair of positioning holes below it, and the center position of all the matrix cells of the row can be accurately calculated, so that the position of the matrix lattice between the two rows and The bit stream information can be obtained by simple calculation according to the side length of the matrix. Therefore, the positioning hole is used as the positioning mark for pattern recognition, the recognition speed is fast, the recognition algorithm is simple, and the scanning and recognition can be processed in parallel.

1286706 V. INSTRUCTIONS (7) -- There may be distortion in the image acquired by the image sensor, for example, the matrix of n*n pixels is distorted into (N + 1 ) * ( N + 1 ) pixels, or For a matrix of straight (Nl)*(N-1) pixels, it can be seen that the smaller N is relative to the error ^$ When N&lt;= 3, the relative error will easily reach 30% or even 40%. The empirical table ′ shows that when the side length of the matrix lattice is greater than or equal to 4 scanning pixels, f has the effect of identifying soil. Let the resolution of the printing be R 1, the design side of the matrix 7 7 is longer than L pixels, and the resolution of the image sensor is R2, which should satisfy $2°/R1 ^ ^ = 4 (the average matrix has 4* "16 pixels" to reliably recognize graphics. Each positioning hole is a white matrix with a circle of black matrix of the same size around it. Therefore, it is included in a large enough area and includes only one small white. Hole. Count the number of white pixels in the area, and simultaneously add their abscissa and sit-in "if respectively" if the number of white pixels is at (|^_ 1 ) * (M-1 ) and (N+l ) * (M+1), it is considered that there is a valid positioning hole in the area, N, Μ is the length of the two sides of the matrix, and the unit is pixel. The center coordinate of the positioning hole is obtained by dividing the accumulated value of the above coordinate by the number of white pixels, respectively. According to the above principle, we have designed a practical application value with the contact image sensor SV252A8. The scheme is used to store a binary computer file in the graphic, and then recover the original binary file by scanning and reading. The resolution of the SV252A8 is 2〇〇DPI, and there are 448 image sensing units' which are arranged in a line at a dot pitch of 2 0 0 D P I . In order to adapt to the hardware characteristics and to most effectively increase the storage capacity on the same carrier area, we

1286706 ------------------ V. Description of the invention (8) The arrangement of the code maps used (as shown in Figure 4). Each matrix has a side length of 4 pixels, and 8 pixels are left as redundant pixels at both ends of the sensor, so that the image sensor can also be sensed when the pattern is skewed. The positioning holes on both sides and the black matrix on both sides of the positioning hole use 12 pixels in total, and the remaining 408 pixels constitute the bit stream area 16 for storing binary information, each line has 408/4 = 1〇 Two matrix grids can store binary information of 1 0 2 bits. In order to simplify the graphical identification item, the information stored in the bit stream area 16 can be organized in a _ format. For example, the bit stream area is divided into blocks by every twenty lines, and the information storage capacity of each block is l〇2bits*20/8 = 255 by tes. In order to correct possible errors, 32 2 bytes are used to store the error code generated by the reed - s ο 1 〇m error correction algorithm (a high-performance error correction algorithm) in every 255 bytes. Therefore, the actual capacity of each bit stream area is 2 2 3 bytes, and these error codes can correct any 16-bit byte errors in the 2 5 5 bytes, and the error correction ratio is 6 · 2 7 %. The bit stream area is an integer block (greater than or equal to 1). When the number of information bytes is not enough to fill the last block bit stream area, it can be filled with a predefined byte, after completing reed - s ο 1 〇m The padding byte is discarded after the error correction operation. A certain number of bytes are occupied at the beginning of the bit stream area 16 for storing specific information of the one-dimensional bar code card itself, such as the version number, the bit length of the information, and the like. The role of the version number is to make the software distinguish the different types of information structure / the length of the information byte is to let the software know the total length of the graphic after reading the first piece of information, in order to decide when to stop scanning the image. The two-dimensional barcode card of the invention can be used as a behavioral list by means of graphic scanning

Page 12 1286706 V. Invention Description (9) Position reading, combined with the decoding software of the simple matrix code can recover the binary information, the decoding method of the simple matrix code is shown in Figure 5, including the following steps: 1. Probing the code map Upper boundary 1 0, calculate and save the slope of the upper boundary 10; 2. Detect the left boundary 1 2 and the right boundary 1 4; 3. According to the position and inclination of the upper boundary, the left boundary 1 2 and the right boundary 1 4 The position estimates the area of the first pair of positioning holes 20; 4. Counts the number of white pixels in the estimated area of the positioning hole 20, when the number of white primitives is greater than (N-1)*(N-1), If it is less than (N + 1)*(N+1), it is considered that there is a valid positioning hole in the area, where N is the side length of the matrix lattice; the central coordinates of the left positioning hole and the right positioning hole are respectively calculated, 5 Since the center coordinates of the same row matrix are on a straight line and the spacing is equal, the center coordinates of all the matrix cells of the first row are calculated and saved according to the inclinations of the center coordinates and the upper boundary of the left and right positioning holes; Read the corresponding bit from the image according to the center coordinates of all the matrix cells in the first row The bit stream information (for example, the pixel is π black π indicates binary information '' 0 '', the pixel is π white π indicates binary information "1π); 7. According to the center coordinates of the previous pair of positioning holes and the matrix The side length estimates the area of the pair of positioning holes, and the center coordinate of the positioning hole is calculated from the detection; if the center coordinate of the positioning hole is successfully detected, the following calculation is performed; otherwise, the pattern recognition is ended; 8. Calculate and save the line The center coordinates of all matrix lattices (same as the calculation method in step 5); 9. Read the corresponding position from the image according to the center coordinates of each matrix lattice

1286706 V. Invention Description (ίο) Bit Stream Gongxun, 1 0. Calculate the coordinates of all matrix lattices between two rows according to the central coordinates of the saved two-row matrix lattice (to find the average of the center coordinates of the upper and lower matrix lattices) ), simultaneously read the corresponding bit stream information; 1 1 . Back to step 7. Since the image obtained by the scan cannot be exactly the same as the original image, the distortion in the image where the brightness changes drastically is severe. The square positioning hole in the original image may become an irregular shape if the square or square is searched. The small hole algorithm will be more complicated, and the recognition algorithm of the simple matrix code is simplified by adopting the method of directly obtaining the central coordinates of the matrix lattice. If the bit stream area 16 contains an error correction code, and each E bit group is added with E byte error correction codes, error correction is performed after each (S + E) byte information is read. Operation, if the error correction succeeds in outputting the correct information of the S bytes, otherwise the recognition process is exited, and the identification can be continued.

1286706 Brief description of the diagram Figure 1 is a 2D barcode code diagram generated by QR code. Figure 2 is a 2D barcode code diagram generated by the PDF41 7 encoding method. Figure 3 is a two-dimensional bar code code diagram generated by the simple matrix code encoding method of the present invention. Figure 4 is a matrix row element arrangement format of an embodiment of a two-dimensional barcode card to which the present invention is applied. Fig. 5 is a diagram showing a decoding method of the simple matrix code of the present invention. Figure number is a brief description of 10.. Upper boundary 12....Left boundary 14....Right boundary 16...Bit flow area 2 0... .... positioning hole

Claims (1)

1286706 VI. Patent application scope 1 type 2D barcode card, information boundary, left 2 4 matrix grid storage map including upper stream) area The upper boundary is the matrix column of the left boundary and the hole. The hole is a left border and a side of the array and a white matrix lattice as in the patent application, wherein the position includes a plurality of matrices as in the patent application, wherein the position is arranged vertically from left to right. For example, in the patent application, the moment R2/R1*L &gt; 2, the R 1 is the first white moment in the left range at the right edge of the code, and the right boundary is a combined row range turbulent zone row and a modified range element The flow zone, or the range of the 4th frame, includes a carrier and is attached to the carrier, and utilizes a two-dimensional bar code code map, which is characterized by: the code boundary, the right boundary, and the bit stream (bit map along the paper feed direction) The foremost matrix row; is located at the outermost edge of the code graph, and has the right column of the first row of the positioning column, respectively. The 1 item includes the error code 1 item is between the boundary and the upper boundary. a two-dimensional barcode card, wherein: the upper boundary is a row of black matrix lattices; comprising three columns of matrix lattices, and three columns of matrix matrices, the middle column is a black matrix grid or the two-dimensional barcode card of item 2 , multiple information blocks, and the information block contains the two-dimensional bar code card described in item 2, and the emission method of the message can be the main action, from the main column, from left to right, or from right to one or The two-dimensional barcode card described in item 2, the length of the side is according to the following formula Select: the resolution when printing the code map; Page 16 1286706 VI. Patent Application Range R 2 is the resolution of the image sensor when reading the code map; L is the design side length of the matrix grid. 6. The two-dimensional barcode card of claim 1 or 2, wherein the left side of the left border of the code pattern on the carrier and the right side of the right border have a blank area. 7 · A two-dimensional barcode decoding method, which is attached to the carrier and restored to the binary information by using the matrix to store the information of the two-dimensional barcode code, which includes the following steps: A. The upper boundary of the detection code map is calculated and saved. The slope of the upper boundary is 9 B · The left and right boundaries of the code map are estimated, and the area of the first pair of positioning holes is estimated according to the inclination of the upper boundary; C · The number of white pixels is counted in the estimated area, and the calculation is performed. The center coordinate of the first pair of positioning holes; D · Calculate the center coordinates of the matrix lattice between the first pair of positioning holes, which are spaced apart by a matrix on one line, and read from the image according to the center coordinates Bit stream information of 1 bit (bit) of each matrix; E · Estimating the area of the next pair of positioning holes according to the central coordinates of the previous pair of positioning holes and the side length of the matrix, and calculating the positioning from the detection The center coordinate of the hole, if the center coordinate of the positioning hole is unsuccessful, the code map is interpreted, otherwise it will enter the next step; F · Calculate according to the center coordinate of the pair of positioning holes, and save between them Central coordinate grid matrix, and according to the center coordinate reading 1286706 6. The bit stream information of the corresponding position of the patent application scope; G · Calculate the coordinates of all the matrix cells between the two rows according to the central coordinates of the saved two-row matrix lattice, and read the bit stream information of the corresponding matrix lattice; Η · Go back to step E. 8. The two-dimensional barcode decoding method according to claim 7, wherein in step C, when the number of white pixels is greater than (Ν-1)*(Μ-ΐ), less than (Ν + 1) * ( Μ + 1 ), it is considered that there is a valid positioning hole in the area, where Ν and Μ are the lengths of the two sides of the matrix. 9. The two-dimensional barcode decoding method according to claim 7, wherein in step D, the total line number information in the code map is read, and the decoding is ended according to the total number of lines. 1 0. The two-dimensional barcode decoding method according to claim 7, wherein each time (S + Ε ) of the byte information is read, an error correction operation is performed, and the S-bits for successful error correction are output. Married information.