KR20180096503A - Method for indicating and recognizing graphics in textbook for education using dot pattern recognition technology - Google Patents

Abstract

The present invention relates to a graphic display and recognition method of an educational textbook using a dot pattern recognition technique. According to the present invention, the graphic display and recognition method of an educational textbook using a dot pattern recognition technique is a method of displaying and recognizing a graphic by printing, in a graphic region of an educational textbook, a specific pattern which is difficult to be recognized by naked eyes, recognizing the specific pattern by an optical recognition pen depending on contact of the optical recognition pen, and outputting a voice or a music corresponding to the corresponding graphic. A unit graphic region of the educational textbook comprises a content part and a header part. The content part is divided into a plurality of virtual sections, wherein at least one dot pattern having content related information is formed in each section. The header part is composed of one section or two sections having the same size as that of each unit section of the plurality of virtual sections of the content part. A framing dot pattern having information on a data frame structure and direction is formed in the one section or the two sections of the header part. The framing dot pattern is composed of two closest dot patterns in the entire frame structure comprising the content part and the header part, wherein the two closest dot patterns are found by using a framing dot pattern searching algorithm. The present invention is able to easily grasp a structure and the direction of a data frame.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0002] The present invention relates to a dot pattern recognition technique,

The present invention relates to a graphic display and recognition method of teaching materials, and more particularly, to a method and apparatus for enhancing recognition accuracy of a digital pattern by performing a high-speed image preprocessing process using a dot pattern recognition technique based on an image processing technique The present invention relates to a graphic display and recognition method of a teaching material using a dot pattern recognition technology.

Today, there are various language teaching (learning) facilities in the language learning machine field that enable students to learn the language electronically by the optical code identification pen. One of such lack of language learning (learning) is the pen learning machine for education. Educational pen learning machine 801 (refer to FIG. 12) is an educational aids called a voice pen, a reading pen, or a talented pen. It is not a general pen for writing letters or drawing pictures on paper, , And an optical lens is provided at an end of the pen, whereby an image printed on the paper can be optically read. Such a pen-learning machine for education is mainly used for a child's foreign language or native language education. Such a pen learning machine is designed such that corresponding voice or music flows according to a dot pattern inserted in the background of the contact surface when the lens portion of the small camera provided at the end portion of the pen is brought into contact with the ground surface of the teaching material. Such a pen learner is used in conjunction with a specially prepared textbook, and the text has a specific pattern printed on it that is hard to be identified by the naked eye. Therefore, the optical recognition lens installed in the pen learning device recognizes such a specific pattern, and corresponding sound source code and sound source data are searched for, thereby outputting voice or music.

With regard to the pen learning machine as described above, with the advancement of technology in the IT field and the printing field, a technique including a graphic indicator together with visible text information or graphic information has been applied to various fields (in particular, education or learning materials field) .

On the other hand, a "graphic indicator" is disclosed in Japanese Patent Application Laid-Open No. 10-2010-0040826 (Patent Document 1), and the graphic indicator 10 according to the present invention includes, as shown in FIG. 1, - units (16) are distributed and divided into a plurality of first state zones (18), one first micro-unit is distributed in each said first state zone, and each of said first states The zone is equally divided into a plurality of virtual sections, the first micro-unit being located in one of the virtual sections to form a plurality of distinct candidate states; And a plurality of second micro-units (16) are distributed and divided into a plurality of second state zones (18), and one second micro-unit is distributed in each of the second state zones, The at least one second micro-unit 16 'is vertically shifted to the connecting line of the center of two adjacent second state zones with respect to the center of the second state zone to provide header information, Each of which is a graphical indicator comprising a header portion 14 disposed at the center of the second state zone and two graphic indicators 10 with different content portions 12 recognize each of the different header portions 14 And are distinguished from each other.

In the case of Patent Document 1 as described above, a smaller number of dots (smaller dot distribution density) can be obtained while displaying the same amount of data as the conventional design, thereby obtaining a better visual effect, (9/16 = 0.56) is low and the overhead (header portion 14) is large, compared with the case where the overhead (header portion 14) is large, the actual frame (Lattice) and the direction and direction of a large number of operations are needed. That is, in order to understand that the overhead dots are located on a straight line, an algorithm is required to calculate straight lines connecting the centers of gravity of the respective dots and to calculate whether the slopes of the overhead dots are within the error ranges. In addition, there is a problem that the overhead is large, but the key dots 16 'indicating the directionality have only one dependency on the corresponding dots.

Published Patent Application No. 10-2010-0040826 (published on April 21, 2010)

The present invention has been made in order to overcome the problems of the related art as described above, and it is an object of the present invention to provide a digital image processing method and a digital image processing method which can realize a high speed image preprocessing process using a dot pattern recognition technology based on an image processing technique, The present invention provides a graphic display and recognition method of a teaching material using a dot pattern recognition technology capable of further improving recognition accuracy of a digital pattern.

According to another aspect of the present invention, there is provided a graphic display and recognition method of a teaching material using a dot pattern recognition technology,

The specific pattern is recognized by the optical recognition pen in accordance with the contact of the optical recognition pen, so that the voice or music corresponding to the graphic is outputted and the graphic is displayed And a method for recognizing,

Wherein the unit graphic area of the educational teaching material includes a content part and a header part, the content part is divided into a plurality of virtual sections, at least one dot pattern having content related information is formed in each section, A framing dot pattern having information about a data frame structure and a direction is formed in one or two sections of the header section, Is formed,

Wherein the framing dot pattern is composed of two dot patterns which are closest in the entire frame structure including the content portion and the header portion,

The nearest two dot patterns are found by using a framing dot pattern search algorithm.

Here, the framing dot pattern search algorithm includes:

a) arranging all n dots with respect to the x-coordinate when n dots are present in an arbitrary image;

b) dividing the image into left and right based on the n / 2th dot among all aligned dots;

c) obtaining a dot pair (pL, pR) having a minimum distance (dL, dR) in both regions;

d) defining an intermediate region M based on median (x-coordinate of n / 2-th dot) when? = min {dL, dR};

e) obtaining a dot pair min {dL, dR, dM} having a minimum distance dM in the region M; And

f) determining a minimum distance by min {dL, dR, dM} and determining the dot pair (one of pL, pR, pM) having this distance.

Preferably, in step b), the image is divided into left and right images, and the steps b) to f) are repeated until the divided image has one or two dots.

According to the present invention, by using a frame pattern for indicating the data frame structure and direction, it is possible to improve the ratio of effective dots in the content area by using a dot pair (framing dot pattern) having a minimum distance, The structure and direction of the data frame can be easily grasped.

In addition, by performing a high-speed image preprocessing process using a dot pattern recognition technology based on an image processing technique, the time required for digital pattern recognition is drastically shortened, and the recognition accuracy of the digital pattern is further improved .

1 is a diagram illustrating an example of a conventional graphic indicator.
FIG. 2 is a view showing a process of an image process as a preprocessing process.
3 is a diagram illustrating an example of a frame pattern formed in a graphic area for realizing a graphic display and recognition method of an educational material using the dot pattern recognition technology according to the present invention.
4 is a flowchart showing a process of finding a framing dot pattern by executing a framing dot pattern search algorithm employed in the method of the present invention.
5 is a diagram schematically illustrating an execution process of a framing dot pattern search algorithm employed in the method of the present invention.
6 is a diagram schematically illustrating a process of grasping an entire frame structure based on a frame dot pattern employed in the method of the present invention.
7 is a diagram illustrating a process of decoding information dots in an image frame.
8 is a diagram showing that "standP" inside the dot pattern is arranged in a square shape.
9 is a diagram schematically illustrating a process of calculating an arbitrary line point and then calculating another "standP" by rotating 90 degrees and -90 degrees with respect to the line point.
10 is a diagram schematically illustrating a process of obtaining an azimuth with respect to an entire frame in an arbitrary image.
FIG. 11 is a diagram illustrating a process of converting the orientation obtained in FIG. 10 into a code to obtain a code for the entire frame.
12 is a diagram showing an example of a general education pen learning machine.
FIG. 13 is a diagram showing an educational text material to which a dot pattern is printed and a dot pattern that is specially printed on a graphic area.

The terms and words used in the present specification and claims should not be construed as limited to ordinary or dictionary terms and the inventor can properly define the concept of the term to describe its invention in the best way Should be construed in accordance with the principles and meanings and concepts consistent with the technical idea of the present invention.

Throughout the specification, when an element is referred to as "comprising ", it means that it can include other elements as well, without excluding other elements unless specifically stated otherwise. Also, the terms " part, "" module, "and" device " Lt; / RTI >

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Before describing the embodiments of the present invention in full, the image processing technology that is the basis of the dot pattern recognition technology employed in the present invention will be described first.

The recognition accuracy of a pattern recognition method using a digital image is based on various nonuniformity and distortion factors (illumination unevenness, color difference, geometric difference (movement, rotation, size difference), optical or motion blur Etc.) can be greatly improved by a preprocessing process.

FIG. 2 is a view showing a process of an image process as a preprocessing process.

Referring to FIG. 2, in the capture portion, a CMOS output (original image) is received and stored in a buffer. In a load portion, image data stored in the buffer is fetched and stored in a data structure for image processing. In the image process part, image processing algorithm is applied to extract information on dots from an image. In a decoding part, a information bit in a frame is read to obtain a bit stream for the corresponding pattern .

Referring to FIG. 2, first, since the CMOS output image is not a color space used in an image processing algorithm, an input image for image processing is prepared through format conversion. In the binarization section, pre-processing such as illumination normalization, histogram stretching, and morphology operations are performed to remove distortion components in the input image, and to distinguish backgrounds from dots. Information on each dot in the image is obtained through a labeling part, and the framing dot pattern search algorithm shown in FIG. 4 is executed through a processing part.

3 is a view showing an example of a frame pattern formed in a graphic area for realizing a graphic display and recognition method of an educational material using a dot pattern recognition technology according to an embodiment of the present invention.

Referring to FIG. 3, a graphic display and recognition method of a teaching material using the dot pattern recognition technology according to the present invention is characterized in that, in the graphic area of the teaching material 901 shown in FIG. 13, Is printed by the optical recognition pen 903 (originally not visible to the naked eye, but shown as shown for the sake of explanation), and is printed by the optical recognition pen according to the contact of the optical recognition pen (education pen learning device 801 3, the unit graphic area 300 of the teaching material includes a content unit 310, a content area 310, and a content area 310. The content area 310 is a part of the unit graphic area 300, And a header unit 320. The content unit 310 is divided into a plurality of virtual sections and at least one dot pattern 311 having content related information is formed in each section , The header part 320 is composed of at least one section having the same size as each unit section of the plurality of virtual sections of the content part 310, and the header part includes a framing A framing dot pattern is formed.

The framing dot pattern is composed of two dot patterns 321a and 321b closest to the entire frame structure including the content unit 310 and the header unit 320. [

In the method of the present invention, the nearest two dot patterns 321a and 321b are searched using the framing dot pattern search algorithm.

4 is a flowchart showing a process of finding a framing dot pattern by executing a framing dot pattern search algorithm employed in the method of the present invention.

Referring to FIG. 4, when there are n dots in an arbitrary image, the framing dot pattern search algorithm (a kind of software program) aligns all n dots based on the x coordinate (step S401). Then, as shown in FIG. 5, the image is divided into left and right based on the n / 2th dot among all aligned dots (step S402). In Fig. 5, "median" represents the x coordinate of the n / 2 < th > dot, L represents the left region, and R represents the right region.

When the division is completed in this manner, the dot pairs pL and pR having the minimum distances dL and dR in both areas are obtained (step S403). Here, the minimum distances dL and dR of the pair of dots pL and pR mean the minimum distance between two dots irrespective of the vertical, horizontal and slant directions between the two dots. Here, the reason why the dot pair (pL, pR) having the minimum distance is obtained is to find the framing dot pattern in the entire frame structure.

Then, when? = Min (dL, dR), an intermediate region M is defined based on median (x coordinate of n / 2th dot) (step S404). Here, δ indicates a smaller value among dL and dR. Also, the region M is a rectangular region with x = [(median-delta), (median + delta)], as shown in Fig.

Here, the n / 2 < th > dot serving as a reference point for dividing the region L and R is not included when dL and dR are obtained. It should be examined because the dot pair with the minimum distance may contain reference points. In order to include a reference point in a pair of dots having the minimum distance, the distance M from the reference point to the nearest dot should be smaller than dL and dR, and therefore, the region M is defined with?.

Next, the dot pair min {dL, dR, dM} having the minimum distance dM in the area M is obtained (step S405). At this time, min {dL, dR, dM} has a minimum distance for all dots. The dot pair min {dL, dR, dM} having the minimum distance in all of these areas becomes the framing dot pattern.

Then, a minimum distance is defined by min {dL, dR, dM}, and a dot pair (one of pL, pR, pM) having this distance is determined (step S406).

Preferably, in the step S402, the steps S402 to S406 are repeated until one or two dots are left in one area in dividing the image into left and right images. In this method, since the time complexity is large when searching for the closest dot pair and distance in the entire image, the method of the present invention uses a divide and conquer algorithm to find a framing dot pattern existing in an image, By repeating the above steps S402 to S406 until one or two dots remain in one area.

As a result, if the framing dot pattern is found in the image, then the entire frame structure is grasped based on this information.

In the dot pattern of the present invention, the framing dot pattern and the crosspoint are in a straight line, and the interval between the intersections is a predetermined multiple (for example, two times) of the length of the framing dot pattern. Accordingly, if two straight lines passing through the framing dot pattern (inverse line) are obtained as shown in FIG. 6, the entire frame structure can be grasped.

Hereinafter, a process of decoding information dots in an image frame in association with the framing dot pattern will be described.

7 is a diagram illustrating a process of decoding information dots in an image frame.

Referring to FIG. 7, in the capture part, the CMOS output (original image) is received and stored in a buffer. In the load part, the image data stored in the buffer is fetched and stored in a data structure for image processing. In the image process part, image processing algorithm is applied to extract information on dots from an image. In a decoding part, a information bit in a frame is read to obtain a bit stream for the corresponding pattern .

In the above process, it is necessary to obtain the intersection (i.e., standP) since the information dot is located at one of the first to fourth quadrants with respect to the cross point as shown in FIG. Further, in order to obtain "standP" from the framing dot pattern, it is confirmed that "standP" is arranged in a square shape as shown in the dot pattern. The line point existing on the framing line can be obtained through the length and coordinates of the framing dot pattern.

After obtaining the line point, another "standP" can be obtained as shown in FIG. When the i + jth line point is rotated counterclockwise by 90 degrees, "standP" is calculated upward by j, and when "-90" is clockwise rotated by "j" Is calculated. At this time, a plurality of lines (for example, about 7 lines) including a frame line are obtained in order to secure sufficient data.

Then, an actual dot (realP) closest to the obtained "standPs " In other words, there are two straight lines passing through the lattice points, which can be expressed by the following equations.

xsinθ _frame + yconθ _frame = ρ _frame

xsinθ _inverse + yconθ _inverse = ρ _inverse

The dot existing within the tolerance in the two straight lines is the actual dot (realP).

In the method of the present invention, one algorithm (realP search algorithm: a kind of software program) is employed to find the actual dots (realP). We will briefly explain this.

According to the "realP search algorithm ", first calculate the length (minLen) between each realP and the nearest line point for all" realP ". Then, divide "minLen" by the line-to-line spacing (len) to calculate how many lines apart from the frame line, and then estimate the row to be searched. And the index of the nearest line point is designated by a column (vertical line).

Then, search standP [startIdx + row, col + 1], standP [startIdx - row, col + 1], find the index satisfying the condition, and bind the corresponding "standP" to "realP".

After obtaining all the "standP" and "realP" existing in the image, calculate the orientation of "realP" based on "standP" and code it. That is, when the values obtained by assigning coordinates (x _i , y _i ) of realP to two straight lines passing "standP" are called "rhoFrame" and "rhoInverse", the following orientation of "realP" .

| RhoFrame - ? _Frame |? MaxDiffRho & rhoInverse - ? _Inverse |? MaxDiffRho → quadrant (where maxDiffRho represents the error correction coefficient)

rhoFrame> ρ _frame &rhoInverse> ρ _inverse → 1 quadrant

rhoFrame> ρ _frame & rhoInverse < ρ _inverse → quadrant 2

rhoFrame < rho _frame & rhoInverse < rho _inverse → quadrant 3

rhoFrame < ρ _frame &rhoIInverse> ρ _inverse → quadrant 4

Based on this, the orientation of the entire frame as shown in FIG. 10 can be obtained in the image.

Thereafter, when the orientation obtained by the above process is converted into a code as shown in Fig. 11, a code for the entire frame can be obtained. Then, by reading the converted values (codes) as shown in Fig. 11 in a predetermined order, a code for the dot pattern can be obtained.

As described above, the graphic display and recognition method of teaching materials using the dot pattern recognition technology according to the present invention uses a frame pattern to represent a data frame structure and a direction, ), It is possible to improve the ratio of effective dots in the content area and to easily grasp the structure and direction of the data frame even with a small amount of calculation.

In addition, by performing a high-speed image preprocessing process using a dot pattern recognition technology based on an image processing technique, the time required for digital pattern recognition is drastically shortened, and the recognition accuracy of the digital pattern is further improved .

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but many variations and modifications may be made without departing from the spirit and scope of the invention. Be clear to the technician. Accordingly, the true scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of the same should be construed as being included in the scope of the present invention.

10: Graphic indicator 12: Content area
14: header section 16, 16 ': second micro-unit
18: first and second state zones 300: unit graphics area
310: contents section 311, 321a, 321b: dot pattern
320: header portion

Claims (12)

The specific pattern is recognized by the optical recognition pen in accordance with the contact of the optical recognition pen, so that the voice or music corresponding to the graphic is outputted and the graphic is displayed And a method for recognizing,
Wherein the unit graphic area of the educational teaching material includes a content part and a header part, the content part is divided into a plurality of virtual sections, at least one dot pattern having content related information is formed in each section, A framing dot pattern having information about a data frame structure and a direction is formed in one or two sections of the header section, ),
Wherein the framing dot pattern is composed of two dot patterns which are closest in the entire frame structure including the content portion and the header portion,
Wherein the nearest two dot patterns are found using a framing dot pattern search algorithm.
The method according to claim 1,
Wherein the framing dot pattern search algorithm comprises:
a) arranging all n dots with respect to the x-coordinate when n dots are present in an arbitrary image;
b) dividing the image into left and right based on the n / 2th dot among all aligned dots;
c) obtaining a dot pair (pL, pR) having a minimum distance (dL, dR) in both regions;
d) defining an intermediate area M based on median (x-coordinate of n / 2-th dot) when? = min {dL, dR};
e) obtaining a dot pair min {dL, dR, dM} having a minimum distance dM in the region M; And
(pL, pR, pM) having a minimum distance defined by f (m, f) min {dL, dR, dM}. / RTI &gt;
3. The method of claim 2,
Wherein the steps b) to f) are repeatedly performed until one or two dots are left in one area in dividing the image to the left and right in the step b) / RTI &gt;
The method of claim 3,
In repeating the steps b) to f), an algorithm for finding a framing dot pattern existing in an image is performed by using a divide and conquer algorithm until the one or two dots are left in one area (b) to (f) are repeatedly performed. The graphic display and recognition method of teaching materials using the dot pattern recognition technology.
3. The method of claim 2,
Further comprising the step of determining the framing dot pattern in the step e), and then grasping the entire frame structure based on the obtained framing dot pattern information. The graphic display and recognition method of the teaching material using the dot pattern recognition technique .
6. The method of claim 5,
The entire frame structure is grasped by obtaining two line segments (inverse line) passing through the framing dot pattern in grasping the whole frame structure. Recognition method.
The method according to claim 1,
Further comprising the step of finding the closest two dot patterns using the framing dot pattern search algorithm and then decoding the information dots in the image frame. Graphical display and recognition methods.
8. The method of claim 7,
Wherein decoding the information dots in the image frame comprises:
Receiving a CMOS output (original image) by a capture unit and storing the output in a buffer;
Fetching image data stored in the buffer by a load unit and storing the fetched image data in a data structure for image processing;
Extracting information on dots from an image by applying an image processing algorithm by an image process unit; And
And reading the information dots in the frame by a decode unit to obtain a bit stream for the pattern. [3] The method of claim 1,
9. The method of claim 8,
Quot; standP "as an intersection from a framing dot pattern for reading the information dot is obtained.
10. The method of claim 9,
The method of claim 1, further comprising: finding the realP that is closest to the obtained "standP " after finding the" standP "
11. The method of claim 10,
Wherein the step of searching for the actual dot is performed by performing a realP search algorithm.
12. The method of claim 11,
The realP search algorithm includes:
For each "realP &quot;, finding a length (minLen) between each" realP "
Dividing the obtained "minLen" by the line-point interval len to calculate a number of lines apart from the frame line, and estimating a row to be searched;
Designating an index of a nearest line point corresponding to the estimated row as a column; And
, and binding the corresponding "standP &quot;to" realP "by searching for standP [startIdx + row, col +/- 1], standP [startIdx-row, col + Graphical Display and Recognition Method of Educational Textbook Using Dot Pattern Recognition Technology.

Images