Information processing device and program

The information processing device addresses the inefficiencies of existing learning systems by generating review sheets that focus on incorrect answers, reducing instructor workload, and enhancing learner understanding through tailored review options.

JP2024156279APending Publication Date: 2024-11-06DAI NIPPON PRINTING CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
JP2023070612
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-04-24
Publication Date
2024-11-06

Smart Images

  • Figure 2024156279000001_ABST
    Figure 2024156279000001_ABST
Patent Text Reader

Abstract

To easily generate a review sheet image consisting of a plurality of questions corresponding to a scoring result of a test.SOLUTION: An information processing device acquires and displays an answer sheet image having a plurality of questions and answer columns in which an answer to each of the questions has been entered in each of the answer columns by a learner. The information processing device acquires correctness information of the answers. Then, the information processing device generates and outputs the review sheet image by deleting or cutting out the incorrect answers. Thus the information processing device can easily generate and output the review sheet image for learning by review to be used by the learner.SELECTED DRAWING: Figure 13
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a technique for generating information related to learning. [Background technology]

[0002] Conventionally, there has been known a learning system that provides review questions to a learner according to the test results of the learner. Patent Document 1 discloses a learning system that provides one review question per day according to the result of an assessment test by the learner. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2012-133251 A Summary of the Invention [Problem to be solved by the invention]

[0004] In reviewing, it is important for the learner to carefully check which questions in the test are correct and which are incorrect, and to answer them again. However, in the invention of Patent Document 1, questions that were incorrect in the assessment test were not necessarily provided as review questions, and since only one question was provided per day, if there were multiple incorrect questions, they could not be reviewed at once. In addition, since the marking results differed for each learner, it was difficult and burdensome for the instructor to manually generate a review test consisting of incorrect questions for each learner.

[0005] The present invention has been made to solve the above-mentioned problems, for example, and aims to provide an information processing device that can easily generate a review sheet image consisting of a plurality of questions according to the test scoring results. [Means for solving the problem]

[0006] In one aspect of the present invention, an information processing device includes an answer sheet image acquisition unit that acquires and displays an answer sheet image having a plurality of questions and answer columns, with each answer column filled in by a learner to answer each question, a correct / incorrect information acquisition unit that acquires correct / incorrect information on the answers, a review sheet image generation unit that deletes or cuts out incorrect answers to generate a review sheet image, and a review sheet image output unit that outputs the review sheet image. According to this aspect, the information processing device can easily generate and output a review sheet image that a learner uses for review learning by deleting or cutting out incorrect answers. This reduces the burden on instructors and other instructors.

[0007] In one aspect of the information processing device, the review sheet image generating unit generates the review sheet image by deleting the incorrect answers in the answer sheet image. According to this aspect, the information processing device generates a review sheet image that has the same layout as the answer sheet image, but deletes only the incorrect answers or answers that received partial points, leaving the answer columns blank. In this way, the correct answers are displayed together with the questions in the review sheet image, so that the student can clearly check which questions are correct and which are incorrect, and can deepen their understanding by reviewing the questions they got wrong.

[0008] In one aspect of the information processing device, the review sheet image generating unit generates the review sheet image by cutting out and extracting the questions with incorrect answers and their answer columns from the answer sheet image and rearranging them. According to this aspect, the review sheet image shows only the questions with incorrect answers and their answer columns, which is easy to use and allows the learner to concentrate on the questions that he or she got wrong.

[0009] In one aspect of the above information processing device, the problem is composed of a main question and sub-questions, the main question represents a prerequisite spanning multiple sub-questions, and the device includes a sample paper image acquisition unit that acquires a sample paper image that is an answer sheet image on which the learner has not written an answer, and an area detection unit that detects position coordinates indicating the areas of the main question, the sub-questions, and the answer columns in the sample paper image, and a correspondence relationship between the main question, the sub-questions, and the answer columns, and when the review paper image generation unit extracts and rearranges the sub-questions and their answer columns based on the position coordinates indicating the areas and the correspondence relationship, the review paper image is generated by extracting and rearranging the main question related to the sub-question and all the sub-questions and their answer columns related to the main question. According to this aspect, when a main question is composed of multiple sub-questions, the information processing device does not extract and rearrange only specific sub-questions and answer columns among them, but extracts and rearranges all the sub-questions and answer columns related to the main question to generate a review paper image. This makes it possible to prevent problems such as only specific sub-questions being displayed in the review sheet image and related sub-questions not being displayed, resulting in questions that are not valid.

[0010] In one aspect of the information processing device, the device further includes a question setting unit that sets questions that are answered correctly as questions that require caution based on the academic ability rank of the learner and the difficulty of the questions, and the review sheet image generating unit generates the review sheet image with the question that requires caution as a question that is answered incorrectly. According to this aspect, when a learner's academic ability rank is low but the answer to a difficult question is correct, the information processing device sets the question as a question that requires caution because it is highly likely that the question was a lucky hit. Then, by the information processing device treating the question that requires caution as a question that is answered incorrectly when generating the review sheet image, the learner can review even the questions that were lucky hits using the review sheet image.

[0011] In another aspect of the present invention, a program executed by an information processing device having a computer causes the computer to function as an answer sheet image acquisition unit that acquires and displays an answer sheet image having a plurality of questions and answer columns, with each answer column filled in by a learner to answer each question, a correct / incorrect information acquisition unit that acquires correct / incorrect information on the answers, a review sheet image generation unit that deletes or cuts out incorrect answers to generate a review sheet image, and a review sheet image output unit that outputs the review sheet image. By installing this program on a computer and causing it to execute, the information processing device of the present invention can be configured. Effect of the Invention

[0012] According to the information processing device of the present invention, it is possible to easily generate a review sheet image made up of a plurality of questions corresponding to the test scoring results. [Brief description of the drawings]

[0013] [Figure 1] 1 shows a configuration of a learning support system in a first embodiment. [Diagram 2] 1 is an example of an image of an answer sheet. [Diagram 3] FIG. 2 is a block diagram showing a hardware configuration of a server according to the first embodiment. [Figure 4] 13 is an example of a data configuration of a marking result DB. [Diagram 5] FIG. 2 is a block diagram showing a functional configuration of a server according to the first embodiment. [Figure 6] 13 is an example of an image generation screen. [Figure 7] 13 is an example of an image generation screen displaying a scoring result. [Figure 8] 13 is an example of a data configuration of a grading result DB corresponding to a target student. [Figure 9] 13 is an example of a sample sheet image, an answer sheet image, and a difference image. [Figure 10] FIG. 13 is a diagram illustrating identification of an answer to a hidden question. [Figure 11] 4 is an example of a review sheet image in the first embodiment. [Figure 12] 13 is an example of an image generation screen on which a review sheet image is displayed. [Figure 13] 4 is a flowchart of an image generation process according to the first embodiment. [Figure 14] 1 shows a configuration of a learning support system in a second embodiment. [Figure 15] FIG. 11 is a block diagram showing a hardware configuration of a server according to a second embodiment. [Figure 16] FIG. 11 is a block diagram showing a functional configuration of a server according to a second embodiment. [Figure 17] 4 is an example of a data configuration of a region detection result and region information. [Figure 18] An example of a display problem and related elements and a blank image. [Figure 19] 13 is an example of a review sheet image in the second embodiment. [Figure 20] 4 is a flowchart of an image generation process according to the first embodiment. [Figure 21] 1 is an example of incorrect and correct rearrangement. [Figure 22] 13 shows the configuration of a learning support system in a third modified example. [Figure 23] 13 shows the configuration of a learning support system in a fourth modified example. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. First Embodiment [Overall configuration] FIG. 1 shows the configuration of a learning support system to which the server of the present invention is applied. In the first embodiment, the learning support system 100a is a system that easily generates a review sheet image that can be used for reviewing a test by displaying or hiding answers according to the test score. For the sake of convenience, the learners are students and the instructors are teachers. In this embodiment, an answer sheet consisting of a plurality of questions and their answer columns is applied as an example of a test.

[0015] The learning support system 100a is configured such that a server 10a and a plurality of teacher terminals 20 can communicate with each other via a network 5 such as the Internet. The server 10a is an information processing device that processes, stores, transmits and receives various information, and is, for example, a server device, a personal computer (PC), or a general-purpose tablet. The server 10a is also connected to a grading result database (hereinafter, "database" will be abbreviated as "DB") 31, which will be described later. The server 10a is an example of an information processing device of the present invention.

[0016] The teacher terminal 20 is used by a teacher and is, for example, an information processing device such as a tablet or PC. Specifically, the teacher terminal 20, which will be described in detail later, is a terminal device that transmits sample paper images, answer sheet images, grading result information, condition information, etc. to the server 10a, and obtains and outputs review paper images corresponding to the grading results from the server 10a. The teacher terminal 20 is also connected to a scanner and a printer, and scans sample papers and answer sheets with the scanner to create images, and obtains grading result information from the answer sheet images using an OCR (Optical Character Reader) and transmits it to the server 10a.

[0017] In this embodiment, the teacher terminal 20 acquires the marking result information from the answer sheet image using OCR, but the present invention is not limited to this, and the server 10a may acquire the marking result information linked to the answer sheet image from a specified DB, etc.

[0018] Here, an answer sheet image will be described. FIG. 2 is an example of an answer sheet image. As shown in FIG. 2, the answer sheet image 51 is an image of an answer sheet composed of a plurality of questions and their answer columns, and the answers are written by the students. Since the answer sheet is an example of a test, the answer sheet image 51 has a test content area 52 that displays the test implementation time, the target grade, the subject, the required time, etc., a student identification information area 53 in which the grade, the class, the number, and the name, which are the identification information of the student, are written, and a question answer column area 54 composed of a plurality of questions, answer columns, and diagrams. In this embodiment, there are two types of questions, a main question and a sub-question, and the main question represents a prerequisite that spans a plurality of sub-questions, and the main question in the answer sheet image 51 is composed of four sub-questions and one diagram. Note that the answer sheet image 51 is an example of a test image, and the test applied to the present invention is not limited to this, and the type and layout of the questions can be set arbitrarily.

[0019] Note that the dotted lines surrounding the answer boxes for each question are included for the convenience of explanation, and do not actually exist; details will be described later.

[0020] [Server hardware configuration] 3 is a block diagram showing the hardware configuration of server 10a. Server 10a includes communication unit 11, control unit 12, storage unit 13, recording medium 14, display unit 15, and input unit 16. These components and scoring result DB 31 are connected to each other via bus 19.

[0021] The server 10a may be executed by a single computer, or may be executed in a distributed manner by a plurality of computers, or may be executed in a distributed manner by virtual machines.

[0022] The communication unit 11 is a communication unit for communicating with the teacher terminal 20 through the network 5. Specifically, the communication unit 11 receives sample paper images, answer sheet images, grading result information, condition information, etc. from the teacher terminal 20, and transmits review paper images according to the grading results to the teacher terminal 20.

[0023] The control unit 12 includes an arithmetic processing device such as a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), and a GPU (Graphics Processing Unit), and performs various information processing, control processing, and the like related to the server 10a by reading and executing programs stored in the storage unit 13. The programs can be deployed so as to be executed on a single computer, or on a single site, or distributed across multiple sites and on multiple computers interconnected by a communication network. Although the control unit 12 is described in FIG. 3 as being a single processor, it may be a multiprocessor.

[0024] The storage unit 13 includes memory elements such as a RAM (Random Access Memory) and a ROM (Read Only Memory), and stores programs or data required for the control unit 12 to execute processing. The storage unit 13 also temporarily stores data required for the control unit 12 to execute arithmetic processing.

[0025] The recording medium 14 is a non-volatile and non-temporary recording medium such as a disk-shaped recording medium or a semiconductor memory, and is configured to be detachable from the server 10a. The recording medium 14 records various programs executed by the control unit 12. When the server 10a executes the image generation process, the programs recorded on the recording medium 14 are loaded into the storage unit 13 and executed by the control unit 12.

[0026] The display unit 15 is a liquid crystal display or an organic EL (electroluminescence) display, etc., and displays various information according to instructions from the control unit 12. The input unit 16 is an input device such as a mouse, a keyboard, a touch panel, a button, etc., and outputs received operation information to the control unit 12.

[0027] Fig. 4 is an example of the data configuration of the marking result DB 31. As shown in Fig. 4, the marking result DB 31 stores marking result information in which a test ID for identifying an answer sheet, which is an example of a test, a grade, class, number, and name for identifying a student, a question number for identifying a question constituting the answer sheet, and a marking result are associated with each other.

[0028] For ease of explanation, the names of the students will be referred to as Learner 1, Learner 2, etc. Also, although details will be described later, the marking result information may be acquired by the teacher terminal 20 from the marked answer sheet image using OCR or the like, or may be acquired by the teacher grading the answer sheet image using the teacher terminal 20. In this embodiment, marking is performed using marking symbols such as a circle for a correct answer, a cross for an incorrect answer, and a triangle for partial points, but the present invention is not limited to this, and the marking method and marking symbols can be set arbitrarily, for example, marking only for correct and incorrect answers without partial points.

[0029] [Server Functionality Configuration] 5 is a block diagram showing the functional configuration of the server 10a. The server 10a is connected to the marking result DB 31, and functionally includes a screen output unit 40, a sample sheet image acquisition unit 41, an answer sheet image acquisition unit 42, a judgment unit 43, a marking result information acquisition unit 44, a condition information acquisition unit 45, a question specification unit 46, a difference image acquisition unit 47, a review sheet image generation unit 48, and a review sheet image output unit 49.

[0030] In addition, the screen output unit 40, sample paper image acquisition unit 41, answer sheet image acquisition unit 42, judgment unit 43, scoring result information acquisition unit 44, condition information acquisition unit 45, question identification unit 46, difference image acquisition unit 47, review paper image generation unit 48, and review paper image output unit 49 are realized by the control unit 12 executing a program.

[0031] In response to various screen requests, the screen output unit 40 transmits screen information of an image generation screen or a grading screen to the teacher terminal 20 to display it. For example, when a teacher uses the teacher terminal 20 to log in to a specific site or launch a specific application, the screen output unit 40 acquires a screen request for an image generation screen. The screen output unit 40 then transmits the image generation screen to the teacher terminal 20. Also, when a teacher uses the teacher terminal 20 to press a grading button 75 (described later) on the image generation screen, the screen output unit 40 acquires a screen request for a grading screen. The screen output unit 40 then transmits the grading screen to the teacher terminal 20.

[0032] FIG. 6 is an example of an image generation screen. The image generation screen is displayed on the teacher's terminal 20, and is a screen where the teacher generates a review sheet image by performing a specified operation. Specifically, the teacher uploads a sample sheet image and an answer sheet image, selects the conditions for the answers to be displayed, and the like. Here, a review sheet image is an image of a review sheet that is composed of questions and their answer columns according to the grading results of the answer sheet image of the target student. Also, a target student is a student for whom a review sheet image is to be generated.

[0033] As shown in Figure 6, the image generation screen has a grade display area 61, a class display area 62, a name display area 63, a subject display area 64, a teacher display area 65, condition selection boxes 66a to 66c, an image display area 70, a save button 71, a zoom in button 72, a zoom out button 73, a grading result display area 74, and a grading button 75.

[0034] The grade display area 61, the group display area 62, and the name display area 63 are areas that display the grade, group, and name of the target student. The subject display area 64 is an area that displays the subject that is the subject of the review sheet image, in other words, the subject of the answer sheet. The teacher display area 65 is an area that displays the teacher's name and identification information. The grade, group, name, and subject of the target student displayed in the grade display area 61, the group display area 62, the name display area 63, and the subject display area 64 may be displayed based on the teacher's input, or may be obtained from the uploaded answer sheet image by OCR or the like and then displayed. In addition, the dotted lines surrounding the grade display area 61, the group display area 62, the name display area 63, and the subject display area 64 are drawn for the convenience of explanation and do not actually exist.

[0035] The condition selection boxes 66a to 66c are boxes in which the teacher selects and checks the conditions for the answers to be displayed in the review sheet image. The condition selection boxes 66a, 66b, and 66c are checked when correct answers (circles), partial point answers (triangles), and incorrect answers (crosses) are displayed, respectively.

[0036] The image display area 70 is an area that displays sample paper images and answer sheet images uploaded by the teacher through a specified operation, and generated review sheet images. The save button 71, enlarge button 72, and reduce button 73 are buttons that can be clicked to save, enlarge, or reduce the displayed image. The marking result display area 74 is an area that displays the question numbers that identify each question in the answer sheet image, the answers, and the marking results. Pressing the marking button 75 sends a screen request for a marking screen to the server 10a. This causes the marking screen to be displayed on the teacher terminal 20, and the teacher can mark the answer sheet images through a specified operation.

[0037] The sample paper image acquisition unit 41 acquires a sample paper image, which is an image of a sample paper made by a scanner, from the teacher terminal 20. A sample paper is an answer sheet on which no answers have been written by students. Specifically, the teacher uses the teacher terminal 20 to upload the sample paper image by performing a predetermined operation on the image generation screen. This enables the sample paper image acquisition unit 41 to acquire the sample paper image from the teacher terminal 20 and store it in the memory unit 13 or the like.

[0038] The sample paper image may be stored in a predetermined DB in advance, and the sample paper image acquisition unit 41 may acquire the corresponding sample paper image from the DB based on the information of the answer sheet image acquired by the answer sheet image acquisition unit 42 described later. Here, the information of the answer sheet image is, for example, information on the test name and question text that can be recognized by OCR or the like, and the sample paper image is stored in the predetermined DB in association with the test name and question text.

[0039] The answer sheet image acquisition unit 42 acquires an answer sheet image, which is an image of the answer sheet on which the target student has written the answer, by a scanner, from the teacher terminal 20. Specifically, the teacher uses the teacher terminal 20 to upload the answer sheet image by performing a predetermined operation on the image generation screen. This enables the answer sheet image acquisition unit 42 to acquire the answer sheet image from the teacher terminal 20 and store it in the memory unit 13 or the like.

[0040] The determination unit 43 determines whether the answer sheet image has been graded. Specifically, the determination unit 43 first detects an answer column area for each question from the answer sheet image 51 as shown in FIG. 2. For example, a method of detecting an object in an image as a rectangular area using deep learning, such as R-CNN (Rich feature hierarchies for accurate object detection and semantic segmentation), can be applied as the area detection method. The area surrounded by a dotted line in the answer sheet image 51 shown in FIG. 2 is the answer column area for each question. At this time, the determination unit 43 may calculate which answer column area the detected answer column area is from the top and the left in the answer sheet image 51, and identify and label the corresponding question number. The determination unit 43 determines that the answer column area has been graded if a marking symbol (such as a circle, a cross, or a triangle) is written in the answer column area by pattern matching.

[0041] The area detection method is not limited to a method using deep learning such as R-CNN, and any method can be applied as long as it is an algorithm that detects an object included in an image. The method of identifying the question number corresponding to the answer area is not limited to a method based on the order of the answer area in the answer sheet image, and any method can be applied, such as identifying the question number written in parentheses by performing character recognition by OCR for each area. The method of determining whether the answer sheet image has been graded is not limited to a method using pattern matching, and any method can be applied.

[0042] The marking result information acquisition unit 44 acquires the marking result information of the answer sheet image and stores it in the marking result DB 31. The marking result information includes information on the correctness of the answers written by the target student for each question. Specifically, when the answer sheet image has been marked, the marking result information acquisition unit 44 recognizes the marking symbols (circle, cross, triangle, etc.) present in each answer column area of ​​the answer sheet image by pattern matching, and acquires the question number and marking result of each question. The acquired question number and marking result are associated with a test ID for identifying the answer sheet, and the grade, class, and name of the target student, and are stored in the marking result DB 31 as marking result information. Then, the marking result information acquisition unit 44 displays the question number and marking result in the marking result display area 74 of the image generation screen. FIG. 7 shows an example of an image generation screen displaying the marking result. As shown in Figure 7, the marking result information acquisition unit 44 may cut out and extract the answers in the answer column areas labeled with each question number from the answer sheet image, along with the question numbers and marking results, and display them in the marking result display area 74 in correspondence with the question numbers and marking results.

[0043] The test ID can be obtained in any manner, for example, by having a two-dimensional barcode indicating the test ID written on the answer sheet image.

[0044] On the other hand, if the answer sheet image has not been graded, the grading result information acquisition unit 44 displays a pop-up screen in a separate window on the grading screen to prompt the teacher to grade. The teacher uses the teacher's terminal 20 to press the grading button 75 on the image generation screen to send a screen request for the grading screen to the server 10a. Then, the teacher grades the answer sheet image using the grading screen displayed on the teacher's terminal 20. The grading result information acquisition unit 44 acquires the question numbers and grading results that identify each question on the answer sheet image by grading performed on the grading screen. The acquired question numbers and grading results are associated with the test ID that identifies the answer sheet, the grade, class, and name of the target student, and are stored in the grading result DB 31 as grading result information. Then, the grading result information acquisition unit 44 displays the question numbers, answers, and grading results in the grading result display area 74 as shown in FIG. 7.

[0045] The condition information acquisition unit 45 acquires condition information related to the conditions of the answer to be displayed on the review sheet image. Specifically, the teacher uses the teacher terminal 20 to check one of the condition selection boxes 66a to 66c on the image generation screen to transmit the condition information to the server 10a. This allows the condition information acquisition unit 45 to acquire the condition information of the answer to be displayed that the teacher has arbitrarily selected from the teacher terminal 20.

[0046] The problem identification unit 46 refers to the marking result DB 31 based on the condition information, and identifies problems whose answers do not match the conditions selected by the teacher as answers to be displayed as non-display questions whose answers are not displayed. FIG. 8 shows the data structure of the marking result DB 31a in which marking result information corresponding to the answer sheet image of the learner 1 who is the target student is stored. Specifically, when a check mark is entered in the condition selection box 66a for displaying the correct answer (circle), the problem identification unit 46 refers to the marking result DB 31a and recognizes that the answer of the problem whose marking result is the correct answer (circle) surrounded by a dotted line is to be displayed. Then, the problem identification unit 46 identifies the problems with the problem numbers 3 and 4 whose marking results are incorrect (cross) as non-display questions whose answers are not displayed.

[0047] The difference image acquisition unit 47 acquires a difference image between the sample paper image and the answer sheet image. Fig. 9 shows an example of the sample paper image, the answer sheet image, and the difference image. As shown in Fig. 9, the difference image acquisition unit 47 acquires the grade, class, number, and name written by the target student and the answers to each question written by the target student as a difference image 80 between the sample paper image 50 and the answer sheet image 51.

[0048] The review sheet image generating unit 48 generates a review sheet image by displaying or hiding the answers according to the grading result of the answer sheet image that the target student has worked on. FIG. 10 is a diagram for explaining the identification of the answers to the hidden questions. Specifically, as shown in FIG. 10, the review sheet image generating unit 48 detects the answer column area of ​​each question as shown by the dotted line from the sample sheet image 50 by, for example, a method using R-CNN. At this time, it calculates which answer column area the detected answer column area is on the sample sheet image 50, and labels the corresponding question number. Based on the information of each answer column area, the review sheet image generating unit 48 identifies answers 81 and 82 surrounded by dotted lines as answers to the hidden questions, question numbers 3 and 4, from the difference image 80. Then, the review sheet image generating unit 48 deletes or erases the identified answers to question numbers 3 and 4 in the answer sheet image 51 by filling them in with the background color, etc., to generate a review sheet image 90 as shown in FIG. 11. In this way, the review sheet image generating unit 48 generates a review sheet image 90 in which the answers to the non-display questions are left blank, without changing the layout of the answer sheet image 51.

[0049] This allows a review sheet image to be generated in which only the answers for the conditions selected by the teacher are displayed. In the above example, questions marked as incorrect or with partial points are hidden, but if the teacher selects a question to be displayed as a partial point (triangle) answer, the server 10a generates a review sheet image in which questions marked as correct or incorrect are hidden and the answers for the hidden questions are left blank. Also, if the teacher selects a question to be displayed as an incorrect (X) answer, the server 10a generates a review sheet image in which questions marked as correct or with partial points are hidden and the answers for the hidden questions are left blank.

[0050] In addition, the sample sheet image has high area detection accuracy because it does not contain unnecessary information such as answers written by students, compared to the answer sheet image. Therefore, the review sheet image generating unit 48 detects the answer column area of ​​each question from the sample sheet image, identifies the answers of the hidden questions from the difference image, and generates the review sheet image, but the present invention is not limited to this, and if the determining unit 43 detects the answer column area of ​​each question from the answer sheet image, the review sheet image may be generated based on the detected answer column area.

[0051] The review sheet image output unit 49 transmits the generated review sheet image to the teacher terminal 20 and displays it in the image display area 70 of the image generation screen. FIG. 12 is an example of an image generation screen displaying a review sheet image 90. Using the teacher terminal 20, the teacher can save the review sheet image 90 displayed on the image generation screen as shown in FIG. 12 as a PDF (Portable Document Format) or the like. The teacher can also transmit the review sheet image 90 as data to a student terminal used by the target student, or print it on a paper medium using a printer and give it to the target student. Thus, the review sheet image 90 can be used for review learning according to the grading results of each student.

[0052] In the above configuration, the sample sheet image acquisition unit 41, answer sheet image acquisition unit 42, marking result information acquisition unit 44, condition information acquisition unit 45, review sheet image generation unit 48, and review sheet image output unit 49 of server 10a are examples of the sample sheet image acquisition unit, answer sheet image acquisition unit, correct / incorrect information acquisition unit, condition information acquisition unit, review sheet image generation unit, and review sheet image output unit of the present invention.

[0053] [Image generation processing] Next, the image generation process for generating a review sheet image by displaying or hiding the answers according to the grading result of the answer sheet image will be described. Fig. 13 is a flowchart of the image generation process. This process is realized by the server 10a executing a program prepared in advance.

[0054] When the teacher uses the teacher terminal 20 to log in to a specific site or launch a specific application, the server 10a acquires a screen request for an image generation screen. Then, the server 10a transmits screen information of the image generation screen as shown in Fig. 6 to the teacher terminal 20, and causes the teacher terminal 20 to display the image generation screen (step S10).

[0055] The teacher uploads a sample sheet image on the image generation screen by a predetermined operation using the teacher terminal 20. As a result, the server 10a acquires the sample sheet image (step S11). The teacher also uploads the answer sheet image of the target student on the image generation screen by a predetermined operation using the teacher terminal 20. As a result, the server 10a acquires the answer sheet image (step S12). Then, the server 10a judges whether the acquired answer sheet image has been graded or not (step S13). If it is judged that the answer sheet image has been graded (step S13; Yes), the server 10a proceeds to the process of step S15. On the other hand, if it is judged that the answer sheet image has not been graded (step S13; No), the server 10a displays a pop-up screen on the image generation screen to prompt grading. The teacher uses the teacher terminal 20 to press the grading button 75 on the image generation screen, and the server 10a acquires a screen request for the grading screen. Then, the server 10a transmits screen information of the marking screen to the teacher terminal 20, and causes the teacher terminal 20 to display the marking screen (step S14). This allows the teacher to use the marking screen to mark each question on the unmarked answer sheet image.

[0056] The server 10a acquires the question numbers and marking results that identify each question on the answer sheet image (step S15). The acquired question numbers and marking results are associated with information that identifies the target student and the subject, and are stored as marking result information in the marking result DB 31. Then, the teacher uses the teacher terminal 20 to enter a check mark on the image generation screen, thereby transmitting the conditions of the answer to be displayed as condition information to the server 10a. As a result, the server 10a acquires the condition information from the teacher terminal 20 (step S16).

[0057] The server 10a refers to the marking result DB 31 based on the condition information, and uses the difference image between the sample sheet image and the answer sheet image to generate a review sheet image in which the answers to the non-displayed questions have been deleted or erased (step S17). The server 10a then transmits the generated review sheet image to the teacher terminal 20 to display it on the image generation screen (step S18), and the image generation process ends. This allows the teacher to use the teacher terminal 20 to transmit the review sheet image to the student terminal of the target student, or to use a printer to print out the review sheet on paper and hand it to the target student. Therefore, the target student can use the review sheet image to thoroughly check the marking results of each question and review the questions that he or she got wrong.

[0058] The learning support system 100a of this embodiment can easily generate a review sheet image according to the grading results of each learner based on the conditions of the answer to be displayed selected by the instructor. For example, if the display target is the correct answer, a review sheet image can be generated in which partial marks and incorrect answers are left blank, as in the review sheet image 90 shown in FIG. 11. This allows the learner to review and study to deepen their understanding by using the digital data review sheet image or the paper review sheet. In addition, since the correct answer is displayed together with the question in the review sheet image, the student can confirm which question is correct or incorrect and deepen their understanding by reviewing the question they got wrong. This reduces the burden on the instructor to create review questions and review tests for review and study.

[0059] In this embodiment, the sample paper and answer sheet are paper media, and images of the sample paper and answer sheet are obtained by scanning them with a scanner, but the present invention is not limited to this, and the sample paper and answer sheet may be images of digital data from the beginning.

[0060] <Second embodiment> [Overall configuration] Fig. 14 shows the configuration of a learning support system to which the server of the present invention is applied. In the second embodiment, the learning support system 100b is a system that easily generates a review sheet image by cutting out and extracting a plurality of questions and answer columns according to the marking results from an answer sheet image and reconstructing them. For the sake of convenience of explanation, the learners are students and the instructors are teachers. In this embodiment, an answer sheet consisting of questions and their answer columns is applied as an example of a test.

[0061] The learning support system 100b is configured such that a server 10b and a plurality of teacher terminals 20 can communicate with each other via a network 5 such as the Internet. The server 10b is an information processing device that processes, stores, transmits and receives various information, and is connected to an area information DB 30 and a grading result DB 31. The server 10b is an example of the information processing device of the present invention. The teacher terminal 20 is the same as in the first embodiment, and therefore a description thereof will be omitted for the sake of convenience.

[0062] [Server hardware configuration] 15 is a block diagram showing the hardware configuration of server 10b. Like server 10a, server 10b includes communication unit 11, control unit 12, storage unit 13, recording medium 14, display unit 15, and input unit 16. These components, area information DB 30, and scoring result DB 31 are connected to each other via bus 19. Since the components other than area information DB 30 are the same as those in the first embodiment, the description will be omitted for convenience.

[0063] The area information DB 30 stores area information on each area of ​​the sample paper image in association with a test ID.

[0064] [Server Functionality Configuration] 16 is a block diagram showing the functional configuration of the server 10b. The server 10b is connected to the area information DB 30 and the marking result DB 31, and functionally includes a screen output unit 40, a sample sheet image acquisition unit 41x, an answer sheet image acquisition unit 42, a judgment unit 43, a marking result information acquisition unit 44, a condition information acquisition unit 45, a question specification unit 46x, a review sheet image generation unit 48x, and a review sheet image output unit 49.

[0065] The screen output unit 40, the sample sheet image acquisition unit 41x, the answer sheet image acquisition unit 42, the judgment unit 43, the marking result information acquisition unit 44, the condition information acquisition unit 45, the question identification unit 46x, the review sheet image generation unit 48x, and the review sheet image output unit 49 are realized by the control unit 12 executing a program. Also, the screen output unit 40, the answer sheet image acquisition unit 42, the judgment unit 43, the marking result information acquisition unit 44, the condition information acquisition unit 45, and the review sheet image output unit 49 are the same as those in the first embodiment, so their explanations will be omitted.

[0066] The sample paper image acquisition unit 41x acquires a sample paper image obtained by imaging the sample paper with a scanner from the teacher terminal 20. The sample paper image acquisition unit 41x then detects the areas of each question, its answer column, and the diagram from the sample paper image, and specifies the position coordinates indicating each area with the horizontal axis of the sample paper image as the x-axis and the vertical axis as the y-axis. The sample paper image acquisition unit 41x also estimates the correspondence relationship between the main question area, the small question area, the answer column area, and the diagram area. Here, the elements corresponding to the main question area, the small question area, the answer column area, and the diagram area are respectively designated as the main question, the small question, the answer column, and the diagram. Also, the main question and diagram that only have one are designated as main question 1 and diagram 1, respectively, and the multiple small questions and answer columns are respectively designated as small question 1 and answer column 1, etc., in combination with the question number. Then, the sample paper image acquisition unit 41x associates the elements with the test ID, and stores area information that associates the elements with areas consisting of the x-coordinate, y-coordinate, width, and height of the upper left point of the rectangle representing the area of ​​each element, and associated elements that are other elements associated with each element, in the area information DB 30. Note that the test ID can be acquired from the sample paper image by any method.

[0067] 17 is an example of the area detection result and the data configuration of the area information DB 30. Specifically, the sample paper image acquisition unit 41x detects the areas of each question, answer column, and diagram from the sample paper image by, for example, a method using R-CNN or the like. The sample paper image acquisition unit 41x also recognizes the question text, symbols, and question number by performing character recognition by OCR for each area. The sample paper image acquisition unit 41x then identifies whether each area is a main question area, a sub-question area, an answer column area, or a diagram area, and labels each area together with the question number.

[0068] Next, the sample paper image acquisition unit 41x performs area correspondence. For example, if the question number "1" in the small question area matches the question number "1" in the answer column area, it is found to be answer column 1 in which the answer to small question 1 is to be written, and the sample paper image acquisition unit 41x associates the small question area with the answer column area and links them as shown in FIG. 17. Also, when the sample paper image acquisition unit 41x detects a large question area with a black diamond symbol, it detects the small question area and the diagram area from the large question area to the next large question area or from the large question area to the bottom of the sample paper image. Then, the sample paper image acquisition unit 41x associates all the detected small question areas and diagram areas with the large question area and links them as shown in FIG. 17. This allows the sample paper image acquisition unit 41x to estimate the correspondence relationship between the large question area, the small question area, the answer column area, and the diagram area detected from the sample paper image. As a result, the sample paper image acquisition unit 41x stores in the area information DB30 area information that corresponds elements, areas consisting of the x coordinate, y coordinate, width and height of the upper left point of the rectangle representing the area of ​​each element, and related elements of each element, in correspondence with the test ID, as shown in Figure 17.

[0069] In addition, the question number recognized by OCR is used as a method of matching sub-question areas and answer box areas, but if the question number cannot be recognized, the correspondence between the areas may be estimated based on the distance between the areas or the order of the areas in the sample paper image, etc.

[0070] The question identification unit 46x refers to the marking result DB 31 based on the condition information, and identifies as display questions those questions whose answers match the conditions selected by the teacher as the answers to be displayed. Fig. 8 shows the data structure of the marking result DB 31a in which marking result information corresponding to the answer sheet image of the learner 1 who is the target student is stored. Specifically, when a check mark is entered in the condition selection box 66c for displaying incorrect answers (marked with a cross), the question identification unit 46x refers to the marking result DB 31a, and identifies questions No. 3 and 4, which have incorrect answers (marked with a cross), as display questions.

[0071] The review sheet image generating unit 48x generates a review sheet image by cutting out and extracting one or more questions and answer columns according to the marking results from the answer sheet image and reconstructing them. FIG. 18(a) is an example of a diagram showing the displayed questions and related elements. Specifically, when the displayed questions are questions with question numbers 3 and 4, the review sheet image generating unit 48x refers to the area information DB30 based on the elements "Sub-question 3" and "Sub-question 4", and recognizes the related element "answer column 3" of the element "Sub-question 3" and the related element "answer column 4" of the element "Sub-question 4". Then, the review sheet image generating unit 48x extracts the question and answer column of question number 3 and the question and answer column of question 4 from the sample sheet image, as shown in FIG. 18(a).

[0072] The review sheet image generating unit 48x also generates an image in which the questions, answer columns, and diagrams are all blank from the answer sheet image. The review sheet image generating unit 48x generates a review sheet image 95 by arranging the displayed questions and their related elements in the generated image without overlapping. FIG. 18(b) is an example of an image in which the questions, answer columns, and diagrams are blank. FIG. 19 is an example of a review sheet image. Specifically, the review sheet image generating unit 48x generates a review sheet image as shown in FIG. 19 by arranging the questions and answer columns of question number 3 and the questions and answer columns of question number 4 without overlapping in an image as shown in FIG. 18(b). The rules for arrangement can be set arbitrarily, such as arranging vertically or horizontally one by one.

[0073] In the above configuration, the sample sheet image acquisition unit 41x, the answer sheet image acquisition unit 42, the marking result information acquisition unit 44, the condition information acquisition unit 45, the review sheet image generation unit 48x, and the review sheet image output unit 49 of the server 10b are examples of the sample sheet image acquisition unit, the answer sheet image acquisition unit, the correct / incorrect information acquisition unit, the condition information acquisition unit, the review sheet image generation unit, and the review sheet image output unit of the present invention. Also, the sample sheet image acquisition unit 41x of the server 10b is an example of the area detection unit of the present invention.

[0074] [Image generation processing] Next, the image generation process will be described, which extracts one or more questions and their answer columns according to the marking results from the answer sheet image and reconstructs them to generate a review sheet image. Figure 20 is a flowchart of the image generation process. This process is realized by the server 10b executing a program prepared in advance.

[0075] When the teacher uses the teacher terminal 20 to log in to a specific site or launch a specific application, the server 10b acquires a screen request for an image generation screen. Then, the server 10b transmits screen information of the image generation screen as shown in Fig. 6 to the teacher terminal 20, and causes the teacher terminal 20 to display the image generation screen (step S20).

[0076] The teacher uploads a sample paper image on the image generation screen by a predetermined operation using the teacher terminal 20. This allows the server 10b to acquire the sample paper image (step S21). The server 10b detects each area from the acquired sample paper image and stores the position information of each area and area information related to related elements in the area information DB 30 (step S22). The teacher also uploads the answer sheet image of the target student on the image generation screen by a predetermined operation using the teacher terminal 20. This allows the server 10b to acquire the answer sheet image (step S23). Then, the server 10b judges whether the acquired answer sheet image has been graded or not (step S24). If the server 10b judges that the answer sheet image has been graded (step S24; Yes), the server 10b proceeds to the process of step S26. On the other hand, if the server 10b judges that the answer sheet image has not been graded (step S24; No), the server 10b displays a pop-up screen on the image generation screen to prompt the student to grade. When the teacher uses the teacher terminal 20 to press the marking button 75 on the image generation screen, the server 10b acquires a screen request for the marking screen. Then, the server 10b transmits screen information of the marking screen to the teacher terminal 20, and causes the teacher terminal 20 to display the marking screen (step S25). This allows the teacher to use the marking screen to mark each question on the unmarked answer sheet image.

[0077] Server 10b acquires the question numbers and marking results that identify each question in the answer sheet image (step S26). The acquired question numbers and marking results are associated with information that identifies the target student and the subject, and are stored as marking result information in marking result DB 31. Then, the teacher uses teacher terminal 20 to enter a check mark on the image generation screen, thereby transmitting the conditions of the answer to be displayed as condition information to server 10b. As a result, server 10b acquires the condition information from teacher terminal 20 (step S27).

[0078] The server 10b refers to the marking result DB 31 based on the condition information and identifies the display questions. The server 10b then refers to the area information DB and extracts the display questions and their related elements from the sample paper image. The server 10b generates an image in which the questions, answer columns, and diagrams are all blank from the answer sheet image, and generates a review paper image by arranging the display questions and their related elements without overlapping (step S28). The server 10b then transmits the generated review paper image to the teacher terminal 20 to display it on the image generation screen (step S29), and the image generation process ends. This allows the teacher to use the teacher terminal 20 to transmit the review paper image to the student terminal of the target student, or to hand over the review paper printed on a paper medium using a printer to the target student. Therefore, the target student can use the review paper image to review the questions that he or she got wrong.

[0079] The learning support system 100b of this embodiment can easily generate a review sheet image according to the grading results of each learner, based on the conditions of the answers to be displayed selected by the instructor. For example, if the display target is an incorrect answer (X), a review sheet image consisting of an incorrect question and its answer column, such as the review sheet image 95 shown in FIG. 19, can be generated. This allows learners to review to deepen their understanding by using the digital data review sheet image or the paper review sheet. In addition, since the review sheet image only displays the incorrect questions and their answer columns, it is easy to use and allows learners to concentrate on the questions they got wrong.

[0080] In addition, when a main question is composed of a diagram or multiple sub-questions as in this embodiment, the server 10b does not generate a review sheet image by extracting and rearranging only specific sub-questions from the main question. This is because the diagrams and sub-questions that compose the main question cannot be solved without the sub-questions before and after them, and there are relationships between the diagrams and sub-questions, making the main question invalid.

[0081] FIG. 21 shows examples of incorrect and correct rearrangement. Specifically, when the displayed questions are question numbers 3, 4, and 6 in the sample paper image shown in FIG. 17, question numbers 3 and 4 are independent sub-questions, and question number 6 is a sub-question that constitutes main question 1 together with question numbers 5, 7, and 8. Therefore, the server 10b does not generate a review paper image by extracting and arranging only the questions and answer columns of question numbers 3, 4, and 6, which are displayed questions, as in the example of incorrect rearrangement shown in FIG. 21. In this case, the server 10b refers to the area information DB 30, extracts the questions and answer columns of question numbers 3 and 4, which are displayed questions, and main question 1, and the questions and answer columns of FIG. 1 and sub-questions 5 to 8, which are related elements of main question 1, as in the example of correct rearrangement shown in FIG. 21, and generates a review paper image by arranging them.

[0082] At this time, the server 10b may extract and arrange the answers to questions 5, 7, and 8, which are not displayed questions, from the answer sheet image, as in the example of correct rearrangement shown in Figure 21, or may leave the answers to questions 5, 7, and 8 blank.

[0083] In addition, in this embodiment, the sample paper and answer sheet are paper media and images of the sample paper and answer sheet are obtained by scanning them with a scanner, but the present invention is not limited to this, and the sample paper and answer sheet may be images of digital data from the beginning.

[0084] <First Modification> In the first and second embodiments, the server 10a and the server 10b may set questions that are marked as correct as questions requiring caution based on the student's academic ability rank and the difficulty level of each question that constitutes the answer sheet. Here, the student's academic ability rank is an index that indicates the student's level of understanding, such as a deviation score.

[0085] Specifically, when a student has a low academic rank but a high-difficulty question is scored correctly, the server 10a and the server 10b set the question as a question requiring caution because it is highly likely that the answer was a fluke. Then, when generating a review sheet image based on the condition information, the server 10a and the server 10b treat the question requiring caution as a question that is scored incorrectly. This allows the target student to review even the questions that were a fluke using the review sheet image.

[0086] In addition, server 10a and server 10b may obtain the difficulty of the questions and the academic rank of the students from a specified server, etc., or may calculate them using Item Response Theory (IRT) or Latent Rank Theory (LRT) by referring to the scoring result DB31.

[0087] <Second Modification> In the above first and second embodiments, the teacher uses the teacher terminal 20 and generates a review sheet image using the image generation screen. However, the present invention is not limited to this, and a student may use a student terminal that can communicate with the server 10a or server 10b via the network 5 and perform the same process as the teacher terminal 20 to generate a review sheet image using the image generation screen.

[0088] <Third Modification> In the above first embodiment, the teacher uses the teacher terminal 20, but the present invention is not limited to this, and the teacher may use a teacher terminal 110a that has the functions of the server 10a. The teacher terminal 110a is, for example, a PC or a general-purpose tablet, just like the server 10a.

[0089] Fig. 22 shows an example of the configuration of the learning support system 200a in this case. The teacher terminal 110a is connected to the marking result DB 131. The teacher terminal 110a executes the image generation process that the server 10a performed, and can easily generate a review sheet image by displaying or hiding the answers in the answer sheet image according to the marking result.

[0090] <Fourth Modification> In the above second embodiment, the teacher uses the teacher terminal 20, but the present invention is not limited to this, and the teacher may use a teacher terminal 110b having the functions of the server 10b. The teacher terminal 110b is, for example, a PC or a general-purpose tablet, just like the server 10b.

[0091] 23 shows an example of the configuration of the learning support system 200b in this case. The teacher terminal 110b is connected to the area information DB 130 and the marking result DB 131. The teacher terminal 110b executes the image generation process that the server 10b performed, extracts a plurality of questions and their answer columns according to the marking results from the answer sheet image, and reconstructs them to easily generate a review sheet image. [Explanation of symbols]

[0092] 5. Network 10a, 10b servers 20, 110a, 110b Teacher's terminal 30, 130 Area information DB 31, 131 Scoring results DB 40 Screen output section 41, 41x Sample paper image acquisition section 42 Answer sheet image acquisition section 43 Judgment section 44 Scoring result information acquisition unit 45 Condition information acquisition section 46, 46x problem identification part 47 Differential image acquisition section 48, 48x Review Paper Image Generator 49 Review paper image output section 100a, 100b, 200a, 200b Learning Support System

Claims

1. an answer sheet image acquisition unit that acquires and displays an image of an answer sheet having a plurality of questions and answer columns, in which answers to the questions are written by a learner in each answer column; a correctness information acquisition unit for acquiring correctness information of an answer; a review sheet image generating unit for generating a review sheet image by deleting or cutting out incorrect answers; a review sheet image output unit for outputting the review sheet image; An information processing device comprising:

2. The information processing device according to claim 1 , wherein the review sheet image generating section generates the review sheet image by deleting incorrect answers in the answer sheet image.

3. The information processing device according to claim 1 , wherein the review sheet image generating section generates the review sheet image by cutting out and extracting questions with incorrect answers and their answer columns from the answer sheet image and rearranging them.

4. The problem is composed of a main question and a sub-question, and the main question represents a prerequisite that spans a plurality of sub-questions, a sample sheet image acquisition unit that acquires a sample sheet image, which is an image of an answer sheet on which no answers have been written by the learner; a region detection unit that detects position coordinates indicating the regions of the main question, the sub-question, and the answer column in the sample paper image, and a correspondence relationship between the main question, the sub-question, and the answer column; The information processing device described in claim 3, wherein when the review sheet image generation unit extracts and rearranges the sub-question and its answer column based on the position coordinates indicating the area and the correspondence, the review sheet image generation unit also extracts and rearranges the main question related to the sub-question and all sub-questions related to the main question and their answer columns, thereby generating the review sheet image.

5. A caution-required question setting unit sets a question whose answer is correct as a caution-required question based on the academic ability rank of the learner and the difficulty level of the question, The information processing apparatus according to claim 1 , wherein the review sheet image generating section generates the review sheet image by treating the caution questions as questions for which answers are incorrect.

6. A program executed by an information processing device having a computer, an answer sheet image acquisition unit which has a plurality of questions and answer columns and acquires and displays an image of an answer sheet on which a learner has written an answer to each question in each answer column; A correct / incorrect information acquisition unit for acquiring correct / incorrect information of an answer; A review sheet image generating unit that deletes or cuts out incorrect answers to generate a review sheet image; a review sheet image output unit for outputting the review sheet image; A program for causing the computer to function as a

Citation Information

Patent Citations

  • Learning system, learning method, server for learning, and program for learning server

    JP2012133251A