KR102089724B1 - Method and apparatus for mutual learning based on image using image meta index - Google Patents

Abstract

The present invention relates to a mutual learning method based on a video using a video meta index and a device thereof. The mutual learning method based on a video using a video meta index comprises the steps of: allowing a video transmission user device to upload a video to a mutual learning server; allowing the mutual learning server to transmit the video to a video reception user device in accordance with selection of the video reception user device; and allowing the video reception user device to transmit evaluation information with regard to the video to the mutual learning server.

Description

Method and apparatus for mutual learning based on image using image meta index}

The present invention relates to an image-based mutual learning method and apparatus using an image meta index. More specifically, the present invention relates to an image-based mutual learning method and apparatus using an image meta index for enhancing a learning effect by performing a judgment on mutual learning based on an image based on a measurement value.

A lot of effort and cost is invested in developing e-Learning contents at home and abroad. In the online education market, education can be performed without time and space limitations, and learning required for each learner can be flexibly and repeatedly performed. In addition, when online interaction is possible, a synergistic effect based on various mutual cooperation may exist.

While offline training is limited and unidirectional, online training can be conducted freely in both directions. In addition, learners can form their own knowledge structures and collaborate with instructors and other learners.

In this online education, a mutual learning process of learning and teaching each other can be introduced. As one of the mutual learning methods, one of the traditional Jewish educational methods is Habruta education. It is a traditional Jewish teaching method of debating and questioning each other in pairs, such as in Habruta education, where you can gain knowledge and gain knowledge through discussion.

If such a mutual learning method is introduced online, education using online can be conducted more effectively. Therefore, research on an online education system for such mutual learning is needed.

The present invention aims to solve all the above-mentioned problems.

In addition, the present invention aims to accurately measure the performance of online mutual education when performing mutual education based on images online.

In addition, another object of the present invention is to perform accurate judgment on a user who uploads an image through objective judgment on uploading of the image.

The representative configuration of the present invention for achieving the above object is as follows.

According to an aspect of the present invention, an image-based mutual learning method using an image meta index comprises: uploading an image to a mutual learning server by an image transmitting user device, and the mutual learning server receiving the image according to selection of an image receiving user device. It may include the step of transmitting to the image receiving user device and the step of transmitting the evaluation information for the image by the image receiving user device to the mutual learning server.

According to another aspect of the present invention, an image-based mutual learning system using an image meta index is an image transmitting user device implemented to upload an image to a mutual learning server, and the mutual learning server receives the image according to the selection of the image receiving user device. The image receiving user device may be configured to transmit the mutual learning server implemented to transmit to the image receiving user device and the evaluation information for the image to the mutual learning server.

According to the present invention, it is possible to accurately measure the performance of online-based mutual education when performing online-based mutual education.

Through an objective judgment on the upload of the image, an accurate judgment on the user who uploads the image can be performed.

1 is a conceptual diagram illustrating an image-based mutual learning system according to an embodiment of the present invention.
2 is a conceptual diagram illustrating a method of determining an image meta-cognitive index according to an embodiment of the present invention.
3 is a conceptual diagram showing an image meta-cognitive index according to an embodiment of the present invention.
4 is a conceptual diagram illustrating a method of calculating a meta-learning readiness index according to an embodiment of the present invention.
5 is a conceptual diagram illustrating a method of calculating a meta learning habit index according to an embodiment of the present invention.
6 is a conceptual diagram illustrating a method of calculating a meta-learning understanding index according to an embodiment of the present invention.
7 is a conceptual diagram illustrating an image analysis method according to an embodiment of the present invention.
8 is a conceptual diagram illustrating an automatic handwriting note generation function for each image according to an embodiment of the present invention.
9 is a conceptual diagram illustrating a method of recommending a learning image based on an image receiver device according to an embodiment of the present invention.

For a detailed description of the present invention, which will be described later, reference is made to the accompanying drawings that illustrate, by way of example, specific embodiments in which the invention may be practiced. These embodiments are described in detail enough to enable those skilled in the art to practice the present invention. It should be understood that the various embodiments of the present invention are different, but need not be mutually exclusive. For example, specific shapes, structures, and characteristics described in this specification may be implemented by changing from one embodiment to another without departing from the spirit and scope of the present invention. In addition, it should be understood that the position or arrangement of individual components within each embodiment may be changed without departing from the spirit and scope of the present invention. Therefore, the detailed description to be described later is not intended to be done in a limiting sense, and the scope of the present invention should be taken to cover the scope claimed by the claims of the claims and all equivalents thereto. In the drawings, similar reference numerals denote the same or similar components throughout several aspects.

Hereinafter, various preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings in order to enable those skilled in the art to easily implement the present invention.

In an image-based mutual learning method and apparatus using an image meta index according to an embodiment of the present invention, a method of performing learning between learners based on an uploaded image is disclosed.

Among the learning methods, there is a method of teaching others and organizing the learning contents and improving the learning effect again. For example, it is a way to improve learning effectiveness by teaching a particular concept to another student.

In an image-based mutual learning method using an image meta index according to an embodiment of the present invention, a method for increasing a learning effect through an image-based mutual education method is disclosed.

1 is a conceptual diagram illustrating an image-based mutual learning system according to an embodiment of the present invention.

In FIG. 1, a mutual learning system for performing mutual learning based on an image is disclosed.

Referring to FIG. 1, the mutual learning system may include an image transmitting user device 100, a mutual learning server 120, and an image receiving user device 140.

The video transmitting user of the video transmitting user device 100 and the video receiving user of the video receiving user device 140 may be students of similar age.

The video transmission user device 100 may be implemented to transmit a video related to learning. The video transmission user device 100 may be implemented to upload an image to be taught to others to the mutual learning server 120.

For example, the first user of the video transmission user device 100 may upload a learning image for progressing learning on the first concept. Based on the learning video uploaded by the video transmitting user device 100, a determination on the learning state of the video transmitting user may be performed.

The mutual learning server 120 may determine the learning status of the video sender based on the learning video uploaded by the video transmission user device 100. The mutual learning server 120 may determine the video meta-cognition index of the video transmitting user based on the uploaded learning video information and the input information change of the video receiving user.

Specifically, the image meta-awareness index may be determined based on uploaded image evaluation information of the image transmitting user who uploaded the image, image upload count information, and reaction information of the image receiving user. Through the calculation of the objective image meta-cognition index, the learning status of the image transmission user can be determined.

The image receiving user device 140 may receive the training image and perform evaluation on the received training image. The learning video may be viewed, and evaluation information on the watched learning video may be transmitted to the mutual learning server 120.

2 is a conceptual diagram illustrating a method of determining an image meta-cognitive index according to an embodiment of the present invention.

In FIG. 2, a method of determining an image meta-cognitive index based on the uploaded image is disclosed.

Referring to FIG. 2, a plurality of learning images may be uploaded by the image transmission user device. The plurality of learning images may be image information for an image transmitting user to indicate his or her learning status, such as an image for an image transmitting user to perform learning and to explain the learned results to others.

The plurality of learning images may be delivered to the image receiving user device, and the image receiving user device may generate evaluation information on the learning image and transmit the evaluation information to the mutual learning server.

As described above, the mutual learning server may determine the image meta-cognition index of the video transmitting user based on the uploaded learning video information and the input information change of the video receiving user.

The image meta-cognition index may be determined based on the meta learning preparation index 210, the meta learning habit index 220, and the meta learning understanding index 230.

The meta learning preparation index 210 is an index value for indicating the learning readiness of the video transmission user, and the meta learning habit index 220 is an index value for indicating the learning habit of the video transmission user, and the meta learning understanding index ( 230) may be an index value related to the evaluation of the image receiving user for the learning image uploaded by the image transmitting user.

In FIG. 3, the meta learning preparation index 210, the meta learning habit index 220, and the meta learning understanding index 230 are specifically disclosed.

3 is a conceptual diagram showing an image meta-cognitive index according to an embodiment of the present invention.

Referring to FIG. 3, as described above, the image meta-cognition index may include a meta learning preparation index 310, a meta learning habit index 320, and a meta learning understanding index 330.

The meta learning readiness index 310 is an index of learning readiness and planning of the video sender with the length of the video uploaded by the video sender.

The meta learning habit index 320 indexes the learning habits of the video sender based on the upload frequency of the video uploaded by the video sender.

The meta-learning understanding index 330 indexes whether the ability to understand the other party of the video sender is improved based on the response of the video receiver.

A determination of a learner's learning state may be performed through an image metacognitive index determined based on the meta learning preparation index 310, the meta learning habit index 320, and the meta learning understanding index 330.

4 is a conceptual diagram illustrating a method of calculating a meta-learning readiness index according to an embodiment of the present invention.

4, a method for calculating a meta-learning readiness index is disclosed.

Referring to FIG. 4, an image length constant value for determining the meta-learning readiness index may be determined as shown in Table 1 according to the length of the image.

<Table 1>

As described above, an image length constant value 410 according to the image length 400 may be set.

The meta preparation index 420 for each image may be determined to calculate the meta learning preparation index 430. The meta preparation index 420 may be determined by considering the image length constant value 410 and the image upload order. The meta-learning readiness index 430 may be determined based on the meta-readiness index 420 of successively uploaded learning images.

Equation 1 below is an equation for calculating the meta preparation index 420.

<Equation 1>

Meta preparation index = image length constant value (L) x (n / 10),

Here, n is an image upload order.

For example, the image sender is the first learning image (n = 1) 3 minutes long, the second training image (n = 2) 4 minutes long, the third training image (n = 3) 5 minutes long, and the 7 minutes long. Uploading of the fourth learning image (n = 4), the fifth learning image (n = 5) 6 minutes long, and the sixth learning image (n = 6) 4 minutes 55 seconds long may be performed.

In this case, the meta preparation index 420 for the six learning images may be determined as shown in Equation 2 below.

<Equation 2>

Meta preparation index 1 (first learning video) = 10x (1/10) = 1

Meta Preparation Index 2 (second learning video) = 10x (2/10) = 2

Meta Preparation Index 3 (3rd learning video) = 10x (3/10) = 3

Meta Preparation Index 4 (4th learning video) = 7x (4/10) = 2.8

Meta Preparation Index 5 (Fifth Learning Video) = 7x (5/10) = 3.5

Meta preparation index 6 (sixth learning video) = 10x (6/10) = 6

The meta-learning readiness index 430 may be determined based on the meta-readiness index 420 of each learning image calculated as described above. Equation 3 below discloses a method for calculating the meta-learning readiness index 430.

<Equation 3>

Case 1) When the image length constant values of the first meta preparation index and the second meta preparation index are different:

Meta Learning Readiness Index = (First Meta Readiness Index + Second Meta Readiness Index) / 2

Case 2) When the first meta preparation index and the second meta preparation index have the same image length constant:

Meta-learning readiness index = (first meta-ready index x second meta-ready index) / 2

Equation 4 below is an example of calculating the meta-learning readiness index 430 based on Equation 3.

<Equation 4>

Meta Learning Preparation Index 12 = (Meta Preparation Index 1 x Meta Preparation Index 2) / 2 = (1x2) / 2 = 1

Meta Learning Preparation Index 23 = (Meta Preparation Index 2 x Meta Preparation Index 3) / 2 = (1x3) /2=1.5

Meta Learning Preparation Index 34 = (Meta Preparation Index 3 + Meta Preparation Index 4) / 2 = (1.5 + 2.8) /2=2.15

Meta Learning Readiness Index 45 = (Meta Readiness Index 4 x Meta Readiness Index5) / 2 = (2.15x3.5) /2=3.76

Meta Learning Preparation Index 56 = (Meta Preparation Index 5 + Meta Preparation Index 6) / 2 = (3.76 + 6) /2=4.88

The meta-learning readiness index 12, the meta-learning readiness index 23, and the meta-learning readiness index 45 have the same image input value, and thus the meta-learning readiness index 430 may be calculated through the product of the meta-readiness index 420. The meta-learning readiness index 34 and the meta-learning readiness index 56 have different image input values, and thus the meta-learning readiness index 430 may be calculated through the sum of the meta-readiness index 420.

5 is a conceptual diagram illustrating a method of calculating a meta learning habit index according to an embodiment of the present invention.

5, a method for calculating a meta-learning habit index is disclosed.

Referring to FIG. 5, the meta learning habit index may include a first lower meta learning habit index 510, a second lower meta learning habit index 520, and a third lower meta learning habit index 530.

The first lower meta learning habit index 510 may be determined based on whether uploading of learning images occurs at a certain time. When the upload time of the learning video of the video transmitting user is relatively constant, a high score is given to the first lower meta learning habit index 510 as the meta learning habit index, and the upload time of the learning video of the video transmitting user is relatively In case it is not constant, a low score may be assigned to the first lower meta-learning habit index 510 as a meta-learning habit index.

The second lower meta learning habit index 520 may be determined based on the upload cycle of the learning image. Specifically, the second lower meta learning habit index 520 may be given a relatively high score if the upload period of the learning image is relatively short, and a relatively low score if the upload time of the learning image is relatively long.

The third lower meta learning habit index 530 may be determined based on the upload amount of the learning image. Specifically, the third lower meta learning habit index 530 may be assigned a relatively high score as the upload amount of the learning video is relatively high, and a relatively low score as the upload amount of the learning video is relatively small.

Specifically, a method of calculating the first lower meta learning habit index 510 is disclosed.

In order to determine the first lower meta learning habit index 510, a value obtained by subtracting the upload time of the previous learning video from the upload time of the last learning video is calculated in minutes, and this is 1440 minutes (total minutes corresponding to 24 hours) Divide by) to find the N value. That is, the same N value may be set if they have the same value in one unit.

Equation 5 below is an expression for calculating the first lower meta learning habit index 510.

<Equation 5>

First sub meta learning habit index = (variable determined based on whether the previous first sub meta learning habit index + N matches) / 2

Here, the first lower meta-learning habit index 510 may be calculated based on the current target learning image and the first previous learning image that is the immediately preceding learning image of the current target learning image, and the previous first lower meta-learning habit index ( 510) may be calculated based on the first previous learning image and the second previous learning image, which is the immediately preceding learning image of the first previous learning image.

The variable determined based on whether or not N matches may be 25 when the values of consecutive Ns coincide, and may be 10 when mismatched.

Table 2 below is an example of calculating the first lower meta learning habit index 510.

<Table 2>

Table 2-1 below is a table showing whether the current upload time matches the previous upload time.

<Table 2-1>

Based on the above table, the first lower meta learning habit index 510 is calculated as in Equation 6 below.

<Equation 6>

First lower meta learning habit index 12 = (0 + 25) /2=12.5

First lower meta learning habit index 23 = (12.5 + 10) /2=11.25

First lower meta learning habit index 34 = (11.25 + 25) /2=18.12

First lower meta learning habit index 45 = (18.12 + 10) / 2 = 14,06

The final value of the first lower meta-learning habit index 45 may be the first lower meta-learning habit index. A meta learning habit index may be determined based on the first lower meta learning habit index 510.

As described above, the second lower meta learning habit index 520 may be determined based on the upload period of the learning image.

N, the difference between the upload time of the current learning video and the previous learning video, is the difference between the upload time of the current learning video and the previous learning video. If it is within 15 or 25 days, it may be 5 after 10 or 25 days.

If the same measurement value is continuous, it is possible to increase the exponential value by adding 1 to the constant value, starting at the initial constant value of 10 each time it is repeated once.

Equation 7 below is a formula for calculating the second lower meta learning habit index 520.

<Equation 7>

Second sub meta learning habit index = (formerly second sub meta learning habit index + current second sub meta learning habit index based variable) / 2

Table 3 below is a table showing the second lower meta learning habit index 520.

<Table 3>

Table 4 below is an example of uploading a learning image for calculating the second lower meta learning habit index 520.

<Table 4>

Based on Table 4 above, the second lower meta-learning habit index 520 may be determined as shown in Equation 8 below.

<Equation 8>

2nd lower meta learning habit index 12 = (0 + 30) / 2 = 15

Second lower meta learning habit index 23 = (15 + 30) /2=22.5

Second lower meta learning habit index 34 = (22.5 + 25) /2=23.75

Second lower meta learning habit index 45 = (23.75 + 20) /2=21.87

The final value of the second lower meta learning habit index 45 may be the second lower meta learning habit index. A meta learning habit index may be determined based on the second lower meta learning habit index 520.

As described above, the third lower meta learning habit index 530 may be determined based on the number of uploaded learning images.

10 if 1 uploaded learning video, 12 if 2 uploaded learning videos, 14 if 3 uploaded learning videos, 16 if 4 uploaded learning videos, 18 if 5 uploaded learning videos, ..., if there are 30 uploaded learning images, it can be set to a value that increases by 2, such as 68.

The third lower meta learning habit index 530 may be determined by dividing the total number of uploaded learning images by the total sum.

Equation 9 below is a formula for determining the third lower meta learning habit index 530.

<Equation 9>

3rd lower meta learning habit index = (formerly based on the 3rd lower meta learning habit index + current 3rd lower meta learning habit index) (Un)) / 2

Un is 10 + n, where 'n' may be an upload order of learning videos. For example, n = 1 for the first uploaded image and n = 2 for the second uploaded image.

Equation 10 below is a formula for calculating the third lower meta learning habit index 530.

<Equation 10>

3rd lower meta learning habit index 12 = (10 + 1) /2=5.5

3rd lower meta learning habit index 23 = (5.5 + 12) /2=8.75,

3rd lower meta learning habit index 34 = (8.75 + 13) /2=10.87

3rd lower meta learning habit index 45 = (10.87 + 14) /2=12.43

The final value of the third lower meta-learning habit index 45 may be the third lower meta-learning habit index. A meta learning habit index may be determined based on the third lower meta learning habit index 530.

Equation 11 below is an equation for determining a meta learning habit index based on the first sub meta learning habit index 510, the second sub meta learning habit index 520, and the third sub meta learning habit index 530. to be.

<Equation 11>

Meta Learning Habit Index = (First Sub Meta Learning Habit Index + Second Sub Meta Learning Habit Index + Third Sub Meta Learning Habit Index) / 3

Equation 11 above is an example of meta learning by using various formulas in the first sub meta learning habit index 510, the second sub meta learning habit index 520, and the third sub meta learning habit index 530. Habit index can be determined.

6 is a conceptual diagram illustrating a method of calculating a meta-learning understanding index according to an embodiment of the present invention.

6, a method for calculating a meta-learning understanding index is disclosed. As described above, the meta-learning understanding index is an index of whether the ability of the video sender to understand the other party is improved based on the response of the video receiver.

Referring to FIG. 6, the first lower meta-learning understanding index 610 may be determined by comparing the number of clicks of the previous learning image of the same type with the number of clicks of the subsequent learning image. The first lower meta-learning understanding index 610 may be calculated as a cumulative average value by obtaining a measurement value at a period of a threshold time (eg, 24 hours) based on a time point when broadcast for a learning image is activated.

Also, the second lower meta-learning understanding index 620 may be determined by comparing the number of clicks of different types of previous learning images with the number of clicks of subsequent learning images.

The meta-learning understanding index may be finally determined based on the average of the first lower meta-learning understanding index 610 and the second lower meta-learning understanding index 620.

The following is a method of finally determining the final meta-learning understanding index 630 based on the first lower meta-learning understanding index 610 and the second lower meta-learning understanding index 620.

Equation 12 below is a formula for calculating the first lower meta-learning understanding index 610.

<Equation 12>

First lower meta learning comprehension index =

(Last Learning Video Views / (Last Learning Video Views + Previous Learning Video Views Average) x100) / 2

The first number of learning video views may be 32, the second number of learning video views is 21, the third number of learning video views may be 56, the fourth number of learning video views is 83, and the fifth number of learning video views may be 102.

In this case, the first lower meta-learning understanding index 610 may be calculated as shown in Equation 13 below.

<Equation 13>

First lower meta learning comprehension index 12 = (Last Learning Video Views / (Last Learning Video Views + Previous Learning Video Views Average) x100) / 2 = (21 / (21 + 32) x100) /2=19.81

First lower meta-learning understanding index 23 = (56 / (56 + 19.81) x100) /2=36.93

First sub meta learning understanding index 34 = (83 / (83 + 36.93) x100) /2=34.60

First lower meta learning understanding index 45 = (102 / (102 + 34.60) x100) /2=37.33

The final value of the first lower meta-learning understanding index 45 may be the first lower meta-learning understanding index.

Equation 14 below is a formula for calculating the second lower meta-learning understanding index based on different types of learning images.

<Equation 14>

Second sub meta learning understanding index =

(Last Learning Video Views / (Last Learning Video Views + Previous Learning Video Views Average) x100) / 2

The first number of learning video views is 17, the second number of learning video views is 44, the third number of learning video views is 76, the fourth number of learning video views is 100, and the fifth number of learning video views is 207.

In this case, the second lower meta-learning understanding index 620 may be calculated as in Equation 15 below.

<Equation 15>

Second lower meta learning understanding index 12 = (44 / (44 + 17) x100) /2=36.06

Second sub meta learning understanding index 23 = (76 / (76 + 36.06) x100) /2=33.91

Second sub meta learning understanding index 34 = (100 / (100 + 33.91) x100) /2=37.33

Second sub meta learning understanding index 45 = (207 / (207 + 37.22) x100) /2=84.72

The final value of the second lower meta-learning understanding index 45 may be the second lower meta-learning understanding index 620.

The final lower meta learning understanding index 630 may be calculated based on Equation 16 below.

<Equation 16>

Final sub meta learning comprehension index = (first sub meta learning comprehension index + second sub meta learning comprehension index) / 2

As described above, when the first sub meta learning understanding index 610 is 37.33 and the second sub meta learning understanding index 620 is 84.72, the final sub meta learning understanding index 630 may be 61.02.

When the meta-learning readiness index, meta-learning habit index, and meta-learning comprehension index are calculated as described above, the image meta-cognition index of meta-learning readiness index, meta-learning habit index, and meta-learning comprehension index as shown in Equation 17 below. It can be calculated as sum.

<Equation 17>

Video meta-cognition index = meta learning readiness index + meta learning habit index + meta learning understanding index

7 is a conceptual diagram illustrating an image analysis method according to an embodiment of the present invention.

In FIG. 7, a method in which a mutual learning server performs automatic editing on a learning image is disclosed.

Referring to FIG. 7, the mutual learning server may provide a learning image based on automatic editing of the learning image.

The mutual learning server may classify the type for the learning image.

For example, in the case of a learning image, there may be various types of images for learning, such as an image having an image sender (or teacher) and a blackboard, an image having a voice and handwritten notes, and an image having only an image sender. have.

The mutual learning server may perform image editing in consideration of the learning effect through image analysis.

In order to increase the concentration of the image, editing is performed centering on the teaching object 700 in the image area where the voice of the teaching object 700 exists, and focusing on the writing object 720 on the writing area of the object. Can be.

First, through the judgment on the teaching object 700, an image area in which only a voice is present without writing can be edited on the image so that the teaching object 700 is the center. Editing of such an image may be performed only when an area (eg, head to shoulder) of the set teaching object 700 is included.

For a non-existing image of the shape of the entire teaching object, an image may be output without editing the teaching object 700.

Next, through the determination of the handwriting object 720, editing based on the current position of the handwriting object 720 may be performed on the image area where the handwriting is present. Editing may be performed to output an image on a threshold area set based on the current handwriting position. The threshold area may be set such that the previous writing area has more weight in consideration of the writing direction based on the writing area. The position of the critical region may be continuously changed in consideration of the existing handwriting region based on the determination of the existing handwriting image. In the beginning of handwriting, the current handwriting position may be centered, but as the handwriting continues, the position of the critical area may be changed such that the existing handwriting area is visible together with the current handwriting area by setting weights in consideration of the existing handwriting area. .

Depending on the handwriting object, the threshold area may be set differently. When the handwriting object is a notebook, the first threshold area (for example, a square area of 10 cm), and when the handwriting object is a blackboard, the second threshold area (for example, , 1 m square area).

Through this video editing, the video receiver can more effectively watch the video, learn based on the video, and evaluate the video.

8 is a conceptual diagram illustrating an automatic handwriting note generation function for each image according to an embodiment of the present invention.

In FIG. 8, an automatic handwriting note generation function for each image is disclosed.

Referring to FIG. 8, an automatic handwritten note may be generated in consideration of a handwriting area of a learning image.

The primary automatic handwritten note 810 for the handwritten image itself is generated, and the secondary automatic handwritten note 820 may be generated by extracting the text based on the image analysis of the handwritten image.

In order to generate the primary automatic handwriting note 810 for the handwriting image itself, a segmentation point for determining the handwriting note generation unit may be determined. The segmentation point may be a generation point for generating a handwritten note. First, a case in which a handwritten note is basically erased may be set as one segment point.

Also, a point at which the writing position is changed in a specific direction may be a segment point. For example, a point at which a writing direction that deviates from a general writing direction (left to right, top to bottom) may also be set as a segment point. The segmentation point may be determined by additionally considering characteristics of an object (for example, a blackboard or a notebook) in which writing is currently performed, and a user's existing writing characteristics. In the case of the blackboard, since the description area is relatively wide, the segmentation point is formed relatively wide, and in the case of the notebook, since the description area is relatively narrow, the segmentation point may be formed relatively narrow. In addition, when the writing space is widely used in consideration of the user's existing handwriting pattern, the segmentation point may be set wide, and when the writing space is used narrowly, the segmentation point may be set narrowly.

A handwritten image image may be captured and generated based on the segmentation point.

Further, according to an embodiment of the present invention, it may be possible to edit the handwritten content according to a change in the handwriting format. The user can split the sub-image on which the handwriting is performed and move it to a desired area. Alternatively, the content written in the notebook may be moved to the blackboard to maintain the content of the handwriting, but re-adjustment may be performed on the area where the handwriting is performed.

In addition, according to an embodiment of the present invention, a secondary automatic handwriting note 820 may be generated through character recognition and image recognition for a handwritten image. In the case of character recognition, it is recognition of a handwritten character, and image recognition may be recognition of a figure, figure, etc., excluding characters. Based on such character recognition and image recognition, handwriting can be arranged once again. Recognition accuracy for each character may be determined. When the recognition accuracy is greater than or equal to the threshold, a handwritten note is generated without a separate trial, and when the recognition accuracy is less than the threshold, the character may be judged. In order to perform a judgment on the text, the audio information at the time the text is generated may be analyzed and the text information and audio information may be compared based on the audio information to determine whether the recognized text is correct. The audio information corresponding to the group (word or word bundle) containing the character may be extracted, and the audio information may be compared with the corresponding portion of the character to perform a judgment on the accuracy of the character.

In addition, according to an embodiment of the present invention, in the case of a figure, a change in line may also be generated in consideration of encounters between lines. The determination as to whether the line is straight may be determined as a straight line by determining whether the slope is less than or equal to the critical slope in consideration of the slope of the line and the degree of change of the line.

In the case of a curve, whether a cause or an ellipse is determined as a circle when a line enters a certain threshold range based on the center of the circle, or is determined as an ellipse if the line does not fall within a certain threshold range based on the center of the circle. You can.

In addition, it may be determined whether or not to meet the line based on the distance between the lines in order to determine whether the line-to-line encounter. When the distance between the lines is less than or equal to the threshold distance, it is determined that the two lines met, and when the distance between the lines exceeds the threshold distance, the two lines may be determined not to meet.

The threshold distance for the generation of such a shape may be adjusted in consideration of the existing shape creation habit of the scriber.

9 is a conceptual diagram illustrating a method of recommending a learning image based on an image receiver device according to an embodiment of the present invention.

9, a method of evaluating an image based on an image reception result of the image receiver device is disclosed.

Referring to FIG. 9, information on a video reception pattern of the video receiver device may be collected.

The image reception pattern information may include hourly image departure information 910, skip information 920, and linked subscription information 930.

The hourly image departure information 910 is about the departure of the image according to time, and may be determined by collecting information on when to stop subscribing to the image. The video departure rate by time is a rate at which the viewing of the video is stopped according to time, and may be collected when the subscription of the video is stopped. If the viewing of the video is disengaged due to a malfunction or malfunction of the device, and when viewing the video again, the hourly video information may be generated without considering the case where the viewing of the video is deviated due to the malfunction or malfunction of the device. have.

The skip information 920 may include information about skipping of an image. The skip information 920 may include information about how many skips are generated for an image, skip start points, and skip stop points.

The linked subscription information 930 may include information on a case where a subscription is generated by being linked between uploaded videos. Information about whether a subscription occurs in association with one video receiver device for video information uploaded from one video sender device may be included in the linked subscription information 930.

Based on the above information, information on a region having high image concentration and a region having low image concentration may be determined, and information on the image region determined based on the image concentration may be transmitted to the image transmitter device. The video sender device may receive the video reception pattern information and produce an image based on this when uploading the video in the future.

In addition, an analysis on a learning image showing characteristics different from the existing one may be performed in consideration of the image reception pattern for each user. For example, when a video receiver mainly having a first reception pattern (image departure characteristic, skip characteristic) shows a first reception pattern and when a video receiver having a first reception pattern shows a different reception pattern than the first reception pattern Can be classified. Such a reception pattern may be classified into patterns by setting an image deviation range and a skip range. An image evaluation may be performed by classifying a video showing a first reception pattern and an image showing a different reception pattern, and comparing the result with users having the same first reception pattern.

When the first user to the tenth user are users having a first reception pattern, when a critical number (for example, a majority) or more of the first user and the tenth user shows a reception pattern other than the first reception pattern, The video may acquire additional ratings or have additional deductions. For example, if the user has a relatively faster departure based on the first threshold time than the first reception pattern and a longer skip based on the first threshold skip period, additional deductions are given to the corresponding video, and the first reception pattern is If a relatively slow departure based on the second threshold time and a shorter skip based on the threshold skip period, an additional rating may be given to the corresponding video. That is, a rating for an image may be assigned in consideration of the behavioral characteristics of an existing user.

The embodiments according to the present invention described above may be implemented in the form of program instructions that can be executed through various computer components and can be recorded in a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, or the like alone or in combination. The program instructions recorded on the computer-readable recording medium may be specially designed and configured for the present invention or may be known and usable by those skilled in the computer software field. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical recording media such as CD-ROMs and DVDs, and magneto-optical media such as floptical disks. medium), and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes produced by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware device can be changed to one or more software modules to perform the processing according to the present invention, and vice versa.

In the above, the present invention has been described by specific matters such as specific components, etc. and limited examples and drawings, which are provided to help the overall understanding of the present invention, but the present invention is not limited to the above embodiments, and Those skilled in the art to which the invention pertains may seek various modifications and changes from these descriptions.

Accordingly, the spirit of the present invention is not limited to the above-described embodiments, and should not be determined, and the scope of the spirit of the present invention as well as the claims to be described later, as well as all ranges that are equivalent to or equivalently changed from the claims Would belong to

Claims (8)

The image-based mutual learning method using the image meta index,
Uploading an image to the mutual learning server by the video transmission user device;
The mutual learning server transmitting the image to the image receiving user device according to the selection of the image receiving user device; And
And transmitting, by the image receiving user device, evaluation information on the image to the mutual learning server.
The mutual learning server performs generation of a second automatic handwritten note by extracting text based on the generation of a primary automatic handwritten note for the handwritten image itself included in the image and the video analysis for the handwritten image,
The primary automatic handwritten note is generated based on a segmentation point for determining the generation unit of the handwritten note,
The segmentation point is a generation point for generating the handwritten note and includes a point where the handwritten note is erased or a point where the handwriting position is changed in a specific direction,
The segmentation point is a method characterized in that it is determined by additionally considering the characteristics of the object to which the current writing is performed, and the existing writing characteristics of the user. ◈ Claim 2 was abandoned when payment of the registration fee was set.◈ According to claim 1,
The secondary automatic handwritten note is generated based on character recognition and image recognition for the handwritten image included in the primary automatic handwritten note,
If the recognition accuracy for the character recognition and the image recognition is greater than or equal to a threshold, the second automatic handwritten note is generated without a separate trial, and when the recognition accuracy is less than the threshold, the trial for the text or image is performed Become,
The judging panel for the text is a judging panel for determining whether the recognized text is correct by analyzing the audio information at the time when the text is generated and comparing the text information with the audio information.
The image recognition method includes linear recognition based on whether the degree of change of a line, figure recognition based on line-to-line encounters, a slope of a line, or a degree of change of a line is a critical slope or a threshold change or less. . ◈ Claim 3 was abandoned when payment of the set registration fee was made.◈ According to claim 2,
The mutual learning server determines an image meta-cognition index for the image,
The image meta-cognition index includes a meta learning preparation index, a meta learning habit index, and a meta learning understanding index,
The meta-learning readiness index is an index value for indicating the learning readiness of the video transmission user of the video transmission user device,
The meta learning habit index is an index value for indicating the learning habit of the video transmission user,
And the meta-learning understanding index is an index value related to an evaluation of an image receiving user for the image uploaded by the image transmitting user. ◈ Claim 4 was abandoned when payment of the set registration fee was made.◈ According to claim 3,
The image meta-cognitive index is a meta-learning index, meta-learning habits index, meta-learning understanding method. ◈ Claim 5 was abandoned when payment of the set registration fee was made.◈ The image-based mutual learning system using the image meta index,
An image transmission user device implemented to upload an image to the mutual learning server;
A mutual learning server implemented by a mutual learning server to transmit the video to the video receiving user device according to the selection of the video receiving user device; And
The video receiving user device is implemented to transmit the evaluation information for the video to the mutual learning server,
The mutual learning server performs generation of a second automatic handwritten note by extracting text based on the generation of a primary automatic handwritten note for the handwritten image itself included in the image and the video analysis for the handwritten image,
The primary automatic handwritten note is generated based on a segmentation point for determining the generation unit of the handwritten note,
The segmentation point is a generation point for generating the handwritten note and includes a point where the handwritten note is erased or a point where the handwriting position is changed in a specific direction,
The segmentation point is a mutual learning system characterized in that it is determined by additionally considering the characteristics of the object to which the current writing is performed and the existing writing characteristics of the user. ◈ Claim 6 was abandoned when payment of the registration fee was set.◈ The method of claim 5,
The secondary automatic handwritten note is generated based on character recognition and image recognition for the handwritten image included in the primary automatic handwritten note,
If the recognition accuracy for the character recognition and the image recognition is greater than or equal to a threshold, the second automatic handwritten note is generated without a separate trial, and when the recognition accuracy is less than the threshold, the trial for the text or image is performed Become,
The judging panel for the text is a judging panel for determining whether the recognized text is correct by analyzing the audio information at the time when the text is generated and comparing the text information with the audio information.
The image recognition includes a line recognition based on whether the degree of change of the line, a figure recognition based on the encounter between the lines, the slope of the line, and the degree of change of the line is based on whether the threshold is less than or equal to the critical slope. Learning system. ◈ Claim 7 was abandoned when payment of the set registration fee was made.◈ The method of claim 6,
The mutual learning server determines an image meta-cognition index for the image,
The image meta-cognition index includes a meta learning preparation index, a meta learning habit index, and a meta learning understanding index,
The meta-learning readiness index is an index value for indicating the learning readiness of the video transmission user of the video transmission user device,
The meta learning habit index is an index value for indicating the learning habit of the video transmission user,
The meta-learning understanding index is a mutual learning system, characterized in that the index value associated with the evaluation of the image receiving user for the image uploaded by the image transmitting user. ◈ Claim 8 was abandoned when payment of the set registration fee was made.◈ The method of claim 7,
The image meta-cognitive index is a cross learning system, characterized in that the sum of the meta-learning readiness index, meta-learning habit index, meta-learning understanding index.

Images