Answer statistics method, device, equipment and storage medium

Abstract

An embodiment of the present application discloses a method, device, equipment and storage medium for answering question statistics. The method includes: obtaining an exercise image to be answered, where the exercise image is obtained after a paper exercise material is photographed by a photographing device of a learning device; identifying the title of each exercise in the exercise image and the corresponding answering area of the title; when it is confirmed that answering starts, controlling the photographing device to photograph the answering operation, and during the photographing process, obtaining the answering content and answering duration of each exercise according to the answering operation; and determining the knowledge mastery level of the corresponding exercise according to the answering content and answering duration. The above solution can solve the technical problem in the prior art that the knowledge mastery level cannot be reasonably determined based on the user's answering process.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of learning devices, and in particular, to a method, device, equipment and storage medium for answering question statistics. Background Art

[0002] At present, artificial intelligence technology is widely used in all walks of life. For example, in the education industry, artificial intelligence technology is applied to learning devices. When a user answers questions on an exercise book or a test paper, the learning device uses a front camera to take pictures of the exercises and the answering content written by the user, and uses artificial intelligence technology to identify the accuracy of the answers. This can not only save the link of manual marking of the answering content, but also automatically count the wrong questions and find out the weak knowledge points, enabling the user to learn targeted.

[0003] However, it is relatively inaccurate to determine the user's knowledge mastery degree through the answering content. For example, the user enters the correct answering content, but due to the user's insufficient mastery of the knowledge point, it takes a long time for the user to answer the question. Another example is that when the user answers the question, due to the insufficient mastery of the knowledge point, the user just writes the answer casually. However, if the user writes the correct answer, it is impossible to determine that the user has insufficient mastery of the knowledge point.

[0004] In summary, using the user's answering process to reasonably determine the user's knowledge mastery degree has become a technical problem that needs to be solved urgently. Summary of the Invention

[0005] The embodiments of the present application provide a method, device, equipment and storage medium for answering question statistics to solve the technical problem that the existing technology cannot reasonably determine the knowledge mastery degree based on the user's answering process.

[0006] In the first aspect, the embodiments of the present application provide a method for answering question statistics, including:

[0007] Obtain an exercise image to be answered, where the exercise image is obtained after a shooting device of a learning device shoots a paper exercise material;

[0008] Identify the title of each exercise in the exercise image and the answering area corresponding to the title;

[0009] When it is confirmed that the answering starts, control the shooting device to shoot the answering operation, and during the shooting process, obtain the answering content and answering duration of each exercise according to the answering operation;

[0010] Determine the knowledge mastery degree of the corresponding exercise according to the answering content and the answering duration.

[0011] In the second aspect, the embodiments of the present application further provide an answering question statistics device, including:

[0012] An image acquisition unit, configured to acquire an exercise image to be answered, where the exercise image is obtained by a photographing device of a learning device photographing a paper exercise material;

[0013] A region recognition unit, configured to recognize the title of each exercise in the exercise image and the corresponding answer region of the title;

[0014] An operation photographing unit, configured to control the photographing device to photograph the answering operation when starting to answer, and during the photographing process, obtain the answering content and answering duration of each exercise according to the answering operation;

[0015] A proficiency determination unit, configured to determine the knowledge proficiency of the corresponding exercise according to the answering content and the answering duration.

[0016] Thirdly, an embodiment of the present application further provides an answering statistics device, including:

[0017] One or more processors;

[0018] A photographing device, configured to photograph according to the instruction of the processor;

[0019] A memory, configured to store one or more programs;

[0020] When the one or more programs are executed by the one or more processors, the one or more processors implement the answering statistics method as described in the first aspect.

[0021] Fourthly, an embodiment of the present application further provides a storage medium containing computer-executable instructions, where the computer-executable instructions are used to execute the answering statistics method as described in the first aspect when executed by a computer processor.

[0022] In the above answering statistics method, device, equipment and storage medium, after controlling the photographing device to photograph the paper exercise material, the title of the exercise and the answer region in the paper exercise material are recognized. Then, when starting to answer is determined, the photographing device is controlled to photograph the answering operation, so as to obtain the answering content and answering duration of each exercise according to the answering operation, and the technical solution of determining the knowledge proficiency of the corresponding exercise by combining the answering content and the answering duration solves the technical problem in the prior art that the knowledge proficiency cannot be reasonably determined based on the user's answering process. By adding the answering duration, the knowledge proficiency can be more reasonably determined. Description of the Drawings

[0023] Figure 1 A schematic diagram of an application of a learning device provided by an embodiment of the present application;

[0024] Figure 2Flowchart of a method for answering question statistics provided by an embodiment of the present application;

[0025] Figure 3 The first frame of captured image provided by an embodiment of the present application;

[0026] Figure 4 The second frame of captured image provided by an embodiment of the present application;

[0027] Figure 5 The third frame of captured image provided by an embodiment of the present application;

[0028] Figure 6 The fourth frame of captured image provided by an embodiment of the present application;

[0029] Figure 7 Schematic diagram of answering questions provided by an embodiment of the present application;

[0030] Figure 8 Flowchart of a method for answering question statistics provided by an embodiment of the present application;

[0031] Figure 9 Schematic structural diagram of an answering question statistics device provided by an embodiment of the present application;

[0032] Figure 10 Schematic structural diagram of an answering question statistics device provided by an embodiment of the present application. Detailed implementation manners

[0033] The present application will be further described in detail below with reference to the drawings and embodiments. It can be understood that the specific embodiments described herein are used to explain the present application, rather than limiting the present application. Additionally, it should be noted that for the sake of convenience of description, only parts related to the present application rather than all structures are shown in the drawings.

[0034] An embodiment of the present application provides a method for answering question statistics. This method for answering question statistics can be executed by an answering question statistics device. The answering question statistics device can be implemented in software and / or hardware and integrated in an answering question statistics device. Among them, the answering question statistics device can be composed of two or more physical entities, or can be composed of one physical entity. The embodiment does not make any limitations in this regard. The answering question statistics device can be a tablet computer, a notebook computer, a mobile phone, a learning device, etc. Currently, taking the answering question statistics device as a learning device as an example. Among them, as an auxiliary device for users to study, the learning device can formulate a study plan, display teaching courses, recommend exercises, capture the learning process or the process of doing questions of the user, and correct the exercises answered by the user, etc. The learning device can also be recorded as a learning machine, a learning terminal, etc.

[0035] In one embodiment, the learning device further includes a photographing device. Optionally, refer toFigure 1 The learning device 11 is fixed on the desktop 12 of a desk (which can also be referred to as a table, a school desk, etc.). The photographing device 13 is located in the middle above the side of the learning device 11 away from the desktop 12 and can function as a front camera. In one embodiment, the photographing device can photograph the desktop. When the user places the paper-based learning materials used for learning on the desktop, the photographing device can photograph the paper-based learning materials on the desktop. Among them, the paper-based learning materials can be paper-based writing materials (such as textbooks) or paper-based exercise materials (such as test papers, workbooks, etc.). The components and structure of the photographing device can be set according to the actual situation. For example, the photographing device includes a rotatable telephoto camera and a fixed wide-angle camera. The wide-angle camera has a larger photographing range and can generally photograph the complete paper-based learning materials on the desktop. The telephoto camera supports a zoom function and has a relatively small field of view (FOV). It can be understood that when the resolution of the camera is fixed, the smaller the field of view, the smaller the area that the camera can photograph, and the more detailed the information expressed by each pixel in the photographed image, that is, the higher the clarity of the image and the reduction of image distortion. Exemplarily, the telephoto camera captures the user's writing process on the paper-based learning materials in real time with high definition and tracks the writing area. The wide-angle camera photographs the complete paper-based learning materials to assist in positioning the writing area. For another example, the photographing device includes a single rotatable camera. At this time, by rotating the camera, different areas of the desktop can be photographed, and then a complete image can be obtained through stitching. Moreover, writing tracking can also be achieved. For still another example, the photographing device includes multiple fixed cameras, and each camera can photograph different areas of the desktop, and then a complete image can be obtained through stitching. In one embodiment, in addition to photographing the desktop, the photographing device can also photograph the user in front of the desk to obtain the picture of the user during the learning process. At this time, different cameras can be used to photograph the desktop and the user. For example, the photographing device includes a rotatable telephoto camera, a fixed wide-angle camera, and a rotatable ordinary camera. The telephoto camera and the wide-angle camera cooperate to photograph the desktop, and the ordinary camera photographs the user.

[0036] Exemplarily, when the learning device executes the answer statistics method, Figure 2 is a flowchart of an answer statistics method provided by an embodiment of the present application. Refer to Figure 2 The answer statistics method includes:

[0037] Step 110, obtain an exercise image to be answered, where the exercise image is obtained after the photographing device of the learning device photographs the paper-based exercise materials.

[0038] Currently, taking the user's answering questions scenario as an example for description. At this time, there are paper exercise materials placed on the fixed desktop of the learning device. The learning device controls the photographing device to photograph the desktop, obtaining an image containing the paper exercise materials. In the embodiment, this image is denoted as the exercise image. It can be understood that the paper exercise materials in the exercise image are in the state where the user has not answered them.

[0039] Step 120: Identify the title of each exercise in the exercise image and the corresponding answering area for the title.

[0040] Exemplarily, perform text recognition on the exercise image to obtain the title of each exercise in the paper exercise materials. Then, combine the titles of each exercise in the paper exercise materials and the blank areas to determine the answering area.

[0041] Among them, the technical means of text recognition can be set according to the actual situation. For example, process the exercise image through Optical Character Recognition (OCR) to obtain the text content included in the exercise image. Then, based on the text content, obtain the title of each exercise. When there are exercise numbers, based on the numbers of each exercise, determine the text content belonging to the same exercise as the title of this exercise. Another example is to pre-construct a neural network that can perform text recognition, and then obtain the title of each exercise in the exercise image through the neural network. It should be noted that identifying the exercise titles of the paper learning materials in the image is a technical means that has been implemented and will not be described in detail currently.

[0042] Exemplarily, determine the answering area of the exercise according to the title of the exercise and the blank area. The answering area is the area for the user to write answers and answering steps. It can be understood that when there is a blank area in the title, then use the blank area in the title as the answering area, or use the blank area in the set symbols or identifiers in or after the title as the answering area. For example, if there is a "()" symbol in or after the title, then use the blank area in the "()" symbol as the answering area. Another example is that after semantic recognition of the title, it is known that the answer needs to be written in the circle identifier "○", so use the blank area in the circle identifier "○" as the answering area. Or use the blank area below the title as the answering area. For example, in the case of application problems or subjective questions, use the blank area below the title as the answering area. An optional method is that after obtaining the exercise title, the type of the exercise can be determined. If it is an exercise type that directly writes the result, such as fill-in-the-blank, multiple-choice, or true / false questions, then the blank area where the title is located can be combined to determine the answering area. If it is an exercise type that requires writing answering steps, such as subjective questions or application problems, then the blank area below the title can be combined to determine the answering area. After determining the answering area, the corresponding relationship between each exercise and its answering area is known.

[0043] Step 130: When it is confirmed that the answering starts, control the photographing device to photograph the answering operation, and during the photographing process, obtain the answering content and answering duration of each exercise according to the answering operation.

[0044] The answering operation refers to the writing operation performed by the user on the paper test materials during answering. In one embodiment, after the answering starts, the learning device photographs the answering operation to obtain the user's answering process.

[0045] Exemplarily, control the photographing device to photograph each answering area. When a pen used for writing and / or the user's hand is detected in a certain answering area, it is determined that the answering starts. Or, whether the answering starts is detected by setting an action. At this time, control the photographing device to photograph each answering area or the paper exercise materials. When a certain set action is detected in a certain answering area, it is determined that the answering starts. Or, before answering, the user notifies the learning device by touching the learning device or inputting voice to the learning device, so that the learning device determines that the answering starts.

[0046] In one embodiment, when it is confirmed that the answering starts, controlling the photographing device to photograph the answering operation may include: when a set action is detected in the image photographed by the photographing device, determining that the answering starts, and controlling the photographing device to photograph the answering operation.

[0047] Wherein, the set action is a pre-specified action, which can be understood as the starting action of answering. That is, before answering, the user first makes this action to prompt the learning device to start answering. Correspondingly, the learning device determines that the answering starts by detecting this action. Optionally, when the user performs the set action, no writing trace will be left on the paper exercise materials. Exemplarily, the user makes a set action (such as an action of ticking in the air) within the photographing range of the photographing device. Then, the learning device can analyze the trajectory of the action made by the user according to the photographed content of the photographing device. Then, compare this trajectory with the trajectory of the set action. If the trajectory coincidence rate reaches the set threshold (which can be set according to the actual situation), it is determined that the set action is detected. Then, control the photographing device to photograph the answering operation.

[0048] In one embodiment, when it is confirmed that the answering starts, controlling the photographing device to photograph the answering operation may include: when receiving an answering start instruction, confirming that the answering starts, and controlling the photographing device to photograph the answering operation, and the answering start instruction is input by the user's voice or touch.

[0049] Among them, the answering start instruction is an instruction issued by the user for the learning device to determine the start of answering questions. The answering start instruction can be input by the user's voice or touch. When input by voice, the learning device is also equipped with a microphone. Optionally, keywords corresponding to the answering start instruction are set, such as the keyword being "start" or "answer questions", etc. After the user inputs voice through the microphone, the learning device analyzes whether the voice contains the keyword of the answering start instruction. If it does, it determines that the answering start instruction has been received. Additionally, the user inputs voice through the microphone, and the learning device analyzes the semantics corresponding to the voice input by the user. If the semantics express the meaning of starting to answer questions, it determines that the answering start instruction has been received. When input by touch, the learning device displays a virtual button indicating the start of answering questions on the display screen. When it detects that the virtual button has received a touch operation, it determines that the answering start instruction has been received. In practical applications, the answering start instruction can also be executed through the physical buttons of the learning device. For example, when it detects that a set physical button has received a set operation, it determines that the answering start instruction has been received. After receiving the answering start instruction, control the shooting device to shoot the answering operation. It should be noted that the current user can be the user answering the questions or other users (such as parents supervising the answering, etc.).

[0050] Before controlling the shooting device to shoot the answering operation, first let the shooting device detect the answering operation. Among them, detecting the answering operation can also be considered as detecting the writing operation, which can be achieved by detecting the writing action or detecting the writing pen and / or hand required for writing. Currently, the writing operation located in the answering area is determined as the answering operation. That is, after starting to answer questions, first instruct the shooting device to shoot each answering area. If a writing operation is detected in a certain answering area, it is determined that the answering operation has been detected. Or, after starting to answer questions, first instruct the shooting device to shoot the paper exercise materials. If a writing operation is detected in the paper exercise series, it is determined whether the writing position where the writing operation is located is in the answering area. If it is in the answering area, it is determined that the detected answering operation has been detected. After detecting the answering operation, control the shooting device to shoot the answering operation. The shooting process can also be considered as a process of tracking the answering.

[0051] When controlling the shooting device to shoot the answering operation, multiple shooting images can be obtained, and the shooting frequency of the images can be set according to the actual situation.

[0052] In one embodiment, through multiple captured images, the answering process can be obtained, that is, the answering content can be obtained. The answering content may include an answering result and answering steps. For exercises that do not require answering steps, the answering result can be directly used as the answering steps, or it can be determined that the answering steps are empty. In one embodiment, the stroke order is obtained according to the writing trajectory in multiple captured images, and then the answering content is determined based on the writing trajectory and the stroke order. In one embodiment, after processing multiple captured images by OCR, the content written by the user in the captured images is obtained as the answering content. In one embodiment, the answering content is obtained by combining the writing trajectory, the stroke order, and OCR technology. At this time, obtaining the answering content for each exercise according to the answering operation includes steps 131 to 134:

[0053] Step 131: Determine the writing trajectory and stroke order generated by the answering operation according to multiple captured images obtained when the answering operation is captured by the capturing device.

[0054] Exemplarily, after each captured image is obtained, the captured image is compared with the previous captured image to determine the newly written writing trajectory in the current captured image. For example, refer to Figures 3 - 6 , which are four frames of captured images obtained by capturing the answering operation during the user's answering process. After parsing the first frame of captured image (which can be done through writing recognition or by comparing with the blank answering area), it is determined that the writing trajectory currently written by the user is "one". Then, after comparing the second frame of captured image with the first frame of captured image, it can be determined that the writing trajectory in the second frame of captured image is "T", and the stroke order is "one", "vertical stroke". Then, after performing the same processing on the third frame of captured image and the fourth frame of captured image, it can be determined that the user's writing trajectory is "king", and the stroke order is "one", "vertical stroke", "one", "one".

[0055] Step 132: Determine the first answering content corresponding to the answering operation according to the writing trajectory and the stroke order.

[0056] Exemplarily, the answer content corresponding to the answer operation can be obtained according to the writing trajectory and the stroke sequence, and the currently determined answer content is recorded as the first answer content. It can be understood that this process is the process of giving semantics to the writing trajectory, that is, determining the user's answer steps and answer results. Optionally, a text library is constructed, and the trajectory and stroke sequence of each text are recorded in the text library. After that, the currently recognized writing trajectory and stroke sequence are compared with the trajectory and stroke sequence of each text in the text library to determine the text represented by the currently recognized writing trajectory and stroke sequence, and then the first answer content is obtained. It can be understood that when the user continuously inputs multiple texts, it can be determined whether to write a new text based on the position interval and pause time of the writing trajectory. For example, when the user writes "we", after the writing of "I" is completed, there is a certain position interval between the first stroke of "we" and the last stroke of "I" and a pause time will be generated. Therefore, based on the position interval and pause time, it can be determined that the new text "we" is written. Optionally, a neural network for text recognition is constructed, and after the writing trajectory and stroke sequence are input into the neural network, the first answer content output by the neural network can be obtained. It is understandable that based on the stroke sequence, it is also possible to determine whether the stroke sequence written by the user is correct.

[0057] Step 133: Process the captured image using optical character recognition technology to obtain a second answer content corresponding to the answer operation;

[0058] Exemplarily, each captured image is processed by OCR to obtain the answer content corresponding to each captured image. Currently, the answer content is recorded as the second answer content. Optionally, after each captured image is obtained, the corresponding first answer content and second answer content can be obtained.

[0059] It is understandable that the first answer content and the second answer content can be identified simultaneously or successively, and there is no current limitation.

[0060] Step 134: Obtain the final answer content of the corresponding question according to the first answer content and the second answer content.

[0061] Exemplarily, the way to determine the final answer content can be to compare the first answer content and the second answer content. If the two are the same, the final answer content is determined to be the first answer content or the second answer content. If the two are different, the different texts are found. Then, semantic recognition is performed on the first answer content and the second answer content to determine the accurate text among the different texts, or it is displayed to the user, and the user selects the accurate text, thereby obtaining the final answer content. Or, a neural network for outputting the final answer content is constructed. Then, the first answer content and the second answer content are input into the neural network to obtain the final answer content. It should be noted that each exercise has a corresponding final answer content. The answer content is the text input by the user by hand and can be easily distinguished from the printed text.

[0062] Exemplarily, when controlling the shooting device to shoot the answering operation, the answering duration can also be counted based on the captured image. Among them, the answering duration can be determined according to the duration of the user's answering operation in the answering area.

[0063] In one embodiment, obtaining the answering duration of each exercise according to the answering operation includes: step 135 - step 136.

[0064] Step 135: Start the answering timing and determine the user's current target answering area according to the answering operation.

[0065] Exemplarily, when it is confirmed that the answering starts, the answering timing is started. Then, the answering operation is detected, and the answering area where the detected answering operation acts is determined. Currently, the answering area is recorded as the target answering area, and the exercise corresponding to the target answering area is the exercise being answered currently. It can be understood that the answering operation may exceed the answering area. At this time, the target answering area can be determined in combination with the distance from the answering area. For example, Figure 7 is a schematic diagram of answering questions provided by an embodiment of the present application. Refer to Figure 7 , which can be considered as a frame image captured by the shooting device based on the answering operation. Among them, for the third sub-question in the exercise labeled "5", the user's answer content exceeds the answering area. At this time, the target answering area can be determined according to the distance between the answering operation and each answering area. It should be noted that compared with starting the timing from when the answering operation is first detected, starting the answering timing from the start of answering in the embodiment can count the reading and thinking time before the user answers, thereby ensuring the accuracy of the answering duration.

[0066] Step 136: When it is detected that the answering operation in the target answering area stops, use the current timing result as the answering duration of the exercise corresponding to the target answering area.

[0067] Exemplarily, after determining the target answering area, the process of tracking and photographing the answering operation can be considered as a process of continuously detecting the answering operation. If no answering operation is detected in the target answering area at a certain moment, it is determined that the answering has stopped currently. It can be understood that there may be a situation where the answering operation exceeds the target answering area. For example, Figure 7 in the second sub-question of the exercise with the label "4" in the middle, the user's answering content exceeds the target answering area, that is, the corresponding answering operation exceeds the target answering area. At this time, in order to avoid the influence of this situation on the statistics of the answering duration, when tracking and photographing the answering operation, if the answering operation exceeds the target answering area, the distance between the answering operation and the target answering area is determined. When the distance is relatively close (which can be compared through a threshold), it is considered that the answering operation acts on the target answering area. Further, when it is determined that the answering has stopped currently, the current timing result is obtained and used as the answering duration of the exercise corresponding to the target answering area. After determining that the answering operation has stopped, the photographing device is continuously controlled to photograph the paper exercise materials to continue detecting new answering operations.

[0068] When the answering operation stops, there are two situations. One is that the user is thinking about the answering content, and the other is that the user starts to read a new question. When the user is thinking about the answering content, although the answering operation has stopped, it still belongs to the answering stage of the target answering area. Therefore, it is necessary to continue to count the answering duration of the current exercise. Based on this, in one embodiment, after using the current timing result as the answering duration of the exercise corresponding to the target answering area, it further includes: starting to count the stop duration of the answering operation; if a new answering operation is detected and the new answering operation is within the target answering area, the currently counted stop duration is obtained; the stop duration is added to the answering duration, and the answering duration is updated according to the new answering operation.

[0069] Exemplarily, the stop duration is the continuous duration when the answering operation stops, and no answering operation is received in each answering area during the stop duration. When counting the stop duration, the learning device continues to use the photographing device to detect the answering operation and stops counting the stop duration when a new answering operation is detected. Then, it can be judged which answering area the new answering operation acts on. If the answering area is the target answering area, it is determined that the user continues to answer the current exercise, and the stop duration is used as the thinking time of the user for the current exercise and added to the currently counted answering duration. Then, according to the new answering operation, the answering duration is continued to be counted. At this time, the answering timing can be restarted, and the counted timing result is added to the original answering duration to update the answering duration. It can be understood that when it is detected that the new answering operation stops again, it can be continued to determine whether to add the stop duration to the answering duration until it is determined that the user answers other exercises.

[0070] It can be understood that when the answering area affected by the new answering operation is another answering area, it can be considered that the user is answering a new question. At this time, after starting to count the stop duration of the answering operation, it further includes: if a new answering operation is detected and the answering operation is within another answering area, then update the other answering area as the target answering area; obtain the answering duration of the exercise corresponding to the target answering area according to the new answering operation, and add the stop duration to the answering duration.

[0071] Exemplarily, if the answering area affected by the new answering operation is another answering area, it is determined that the user is answering a new question. At this time, the other answering area can be updated as the target answering area. When a new answering operation is detected, restart the answering timing. At this time, the obtained timing result can be considered as the answering duration of the new exercise. The statistical method for the answering duration corresponding to the new target answering area is the same as the aforementioned statistical method for the answering duration, which will not be elaborated here. Optionally, when a new answering operation is detected, obtain the currently counted stop duration, and add the stop duration to the answering duration corresponding to the current new target answering area. At this time, the stop duration can be considered as the duration for reading the question.

[0072] For example, when the answering operation in the target answering area A1 stops, the timing result is T1, start counting the stop duration, and when a new answering operation is detected, the stop duration is T2. Then, when it is determined that the new answering operation is still within the target answering area, restart the answering timing, and when the answering operation stops, obtain the timing result T3. Then, start counting the stop duration, and when a new answering operation is detected, the stop duration is T4. Then, when it is determined that the new answering operation acts in another answering area B1, determine that the answering duration of the exercise corresponding to the target answering area A1 is T1 + T2 + T3. Update the answering area B1 as the target answering area, start the answering timing, and add T4 to the obtained answering duration to obtain the answering duration corresponding to the target answering area B1.

[0073] In one embodiment, there is a situation where after a user answers a certain exercise and then returns to answer another exercise that has already been answered. At this time, taking the current timing result as the answering duration of the exercise corresponding to the target answering area includes: determining whether there is already a corresponding answering duration for the target answering area, and the existing answering duration is statistically obtained according to the historical answering operations in the target answering area; if there is already an answering duration, then use the superposition result of the existing answering duration and the current timing result as the answering duration of the exercise corresponding to the target answering area.

[0074] Historical answering operations refer to the answering operations that have been performed in the target answering area during the current answering process (i.e., the process of answering the current paper exercise materials). After the answering operation stops, new answering operations are detected in other answering areas. During the operation of the historical answering operation, the answering duration has been counted in the target answering area. Therefore, the currently counted answering duration needs to be added to the already counted answering duration. Specifically, after determining the answering duration, first judge whether the exercise corresponding to the target answering area already has an answering duration. If there is an answering duration, it means that the exercise has been answered. Therefore, the sum of the existing answering duration and the currently counted answering duration is used as the new answering duration. If there is no answering duration, the currently counted answering duration is recorded. It can be understood that referring to the foregoing content, the process of determining the answering duration needs to consider the timing result of the answering timer and the stop duration.

[0075] After obtaining the answering content and the answering duration, perform step 140.

[0076] Step 140: Determine the knowledge mastery level of the corresponding exercise according to the answering content and the answering duration.

[0077] Exemplarily, the knowledge mastery level refers to the mastery level of the knowledge points corresponding to the exercise. Each exercise has corresponding exercise points, and the division of the exercise points is not limited at present. For example, referring to Figure 7 , the knowledge points corresponding to the exercise numbered "8" are first-level operations and second-level operations. Currently, the knowledge mastery level can include three categories: proficient, generally proficient, and unproficient. In practical applications, the knowledge mastery level can also be divided into more detailed levels, which are not limited at present.

[0078] Currently, the calculation rules for the knowledge mastery level are not limited. Optionally, the answering content and the answering duration can be used as influencing factors for the knowledge mastery level. Then, the knowledge mastery level is determined according to the influencing factors. For example, when the answering content is correct and the answering duration is short, the knowledge mastery level is determined to be proficient after calculation. When the answering content is correct and the answering duration is long, the knowledge mastery level is determined to be generally proficient after calculation.

[0079] In one embodiment, when calculating the knowledge mastery level, step 140 may include steps 141 - 143:

[0080] Step 141: Obtain the answering result and the answering steps of the corresponding exercise according to the answering content.

[0081] Exemplarily, for exercises that do not require answer steps, the answer content can be directly used as the answer result, and it is determined that the answer steps are empty. For exercises that require answer steps, the answer steps and the answer result can be determined based on the semantic recognition result of the answer steps, or other methods can be used to obtain the answer steps and the answer result, which are not limited at present. It should be noted that it is possible to determine whether answer steps and an answer result are required through the exercise type. For example, fill-in-the-blank questions, multiple-choice questions, true or false questions, etc. do not require answer steps, while calculation questions, application questions, etc. require answer steps. It is also possible to directly perform semantic recognition on the answer content to determine whether it contains answer steps. If it contains answer steps, the answer steps and the answer result are obtained; otherwise, only the answer result is obtained.

[0082] Step 142: Determine the grading result of the corresponding exercise based on the answer result, answer steps, reference result, and reference steps.

[0083] Exemplarily, the reference result and reference steps refer to the correct answer result and answer steps. In one embodiment, the reference result and reference steps can be found in a pre-constructed exercise database. At this time, after identifying the title of each exercise in the exercise image and the corresponding answer area for the title, it further includes: searching for the corresponding reference result and reference steps in the exercise database according to the title of the exercise.

[0084] The exercise database contains the titles of a large number of exercises and the corresponding reference results and reference steps. The exercise database can be stored in the learning device or in the background server of the learning device. Optionally, the exercise database can be classified according to subject and grade. It is understandable that for exercises with multiple solution methods, the exercise database can contain multiple reference steps. Exemplarily, after obtaining the titles of each exercise, access the exercise database to find the exercise with the same title in the exercise database. In one embodiment, by calculating the similarity of the titles, find one or more exercises in the exercise database that are most similar to the title, and then obtain the reference result and reference steps corresponding to the found exercise. At this time, the obtained reference result and reference steps can be used to judge the accuracy of the current answer. Generally, the reference result and reference steps can be found in the exercise database for each title. If a certain exercise is not entered into the exercise database, making it impossible to find the reference result and reference steps, the title of the exercise can be pushed to the administrator of the exercise database, and the administrator can incorporate the title, reference result, and reference steps of the exercise into the exercise database.

[0085] After obtaining the reference answer and reference steps, the reference answer and the answer result can be compared to judge whether the answer result is accurate, and the reference steps and the answer steps can be compared to determine whether the answer steps are accurate. Furthermore, based on the two comparison results, the grading result is obtained, and the grading result records whether the answer result and the answer steps are accurate.

[0086] Step 143: Obtain the knowledge mastery level based on the interval to which the answering duration belongs and the marking result.

[0087] Exemplarily, intervals are pre-divided for the answering duration, and different intervals correspond to different answering duration ranges. The basis for the division and the corresponding answering duration ranges can be set in combination with actual requirements. For example, currently, fast intervals, normal intervals, slow intervals, and invalid regions are divided. Among them, the fast interval means that the answering duration is less than or equal to the first duration threshold. When the answering duration is in the fast interval, it can be considered that the user may be answering randomly. The slow interval means that the answering duration is greater than or equal to the second duration threshold and less than the third duration threshold. When the answering duration is in the slow interval, it can be considered that the user does not firmly master the corresponding knowledge points. The normal interval means that the answering duration is greater than the first duration threshold and less than the second duration threshold. When the answering duration is in the normal region, it can be considered that the user is proficient in the corresponding knowledge points. The invalid region means that the answering duration exceeds the third duration threshold. When the answering duration is in the invalid region, it can be considered that the user has not been answering continuously, and the corresponding knowledge mastery level cannot be determined. The first duration threshold, the second duration threshold, and the third duration threshold can all be determined in combination with the answering durations of all users in the network. For example, the first duration threshold is less than or equal to the shortest answering duration required by the user, the second duration threshold is greater than or equal to the longest answering duration required by the user under the condition of proficient mastery, and the third duration threshold is greater than or equal to the longest answering duration required by the user. It can be understood that the answering duration ranges corresponding to the regions used for different types of exercises can be different.

[0088] After that, obtain the knowledge mastery level based on the interval to which the answering duration belongs and the marking result. For example, determine in advance the intervals and marking results corresponding to different knowledge levels, and then obtain the knowledge mastery level based on the interval and the marking result. Another example is that each interval corresponds to an influence weight, and different marking results also correspond to different influence weights. For example, the marking results include four categories: accurate answering result and accurate answering steps, accurate answering result and inaccurate answering steps, inaccurate answering result and accurate answering steps, and inaccurate answering result and inaccurate answering steps. Different marking results correspond to different influence weights. The greater the influence weight, the greater the influence degree when determining the knowledge mastery level. After that, input the currently determined interval, marking result, and the corresponding influence weight into the trained neural network to output the knowledge mastery level through the neural network.

[0089] Optionally, after obtaining the intervals to which the answering durations of all exercises belong, if most of the answering durations (such as 80% of the answering durations) belong to the fast interval, it is determined that the user has not answered carefully. At this time, the image containing the answering content and the question can be sent to the supervision device (such as the electronic device used by the student's parents) so that the user of the supervision device can assist in the inspection.

[0090] Optionally, after determining the knowledge mastery level, learning suggestions can be generated based on the knowledge mastery level for the user to study targeted according to the learning suggestions. Or, after determining the knowledge mastery level, exercises can also be recommended to the user in a targeted manner so that the user can conduct consolidation exercises on the knowledge points that are not mastered proficiently.

[0091] Optionally, during the execution of each of the above steps, after the user answers each exercise, the user can obtain the answer content and the answering duration, and then obtain the corresponding knowledge mastery level, that is, the knowledge mastery level can be determined in real time. Or, after the user has completed all the answers, the answer content and the answering duration of each exercise are obtained, and then the corresponding knowledge mastery level is obtained.

[0092] Optionally, the server can also determine the exercise mastery level. At this time, the answer content and the answering duration can be sent to the server, and the server determines the knowledge mastery level of the corresponding exercise.

[0093] In the above, after controlling the shooting device to shoot the paper exercise materials, the questions and the answering areas of the exercises in the paper exercise materials are recognized. Then, when it is determined to start answering, the shooting device is controlled to shoot the answering operation, so as to obtain the answer content and the answering duration of each exercise according to the answering operation, and the knowledge mastery level of the corresponding exercise is determined in combination with the answer content and the answering duration. The technical solution solves the technical problem in the prior art that the knowledge mastery level cannot be reasonably determined based on the user's answering process. By adding the answering duration, the knowledge mastery level can be determined more reasonably. And when determining the answering duration, by counting the stop duration, the thinking time and the question reading time can be added to the answering duration, ensuring the accuracy of the answering duration. And the answer content can be determined by combining the writing trajectory, the stroke order and OCR, ensuring the accuracy of the answer content.

[0094] Figure 8 This is a flowchart of an answering statistics method provided by an embodiment of the present application. The answering statistics method provided by this embodiment adds an emotion detection process on the basis of the above answering statistics method. In this embodiment, the shooting device not only shoots the paper exercise materials, but also shoots the user answering the questions. At this time, different cameras of the shooting device are used to shoot the paper exercise materials and the user respectively. Refer to Figure 8 and the answering statistics method includes:

[0095] Step 210: Obtain the exercise image to be answered, where the exercise image is obtained by the shooting device of the learning device shooting the paper exercise materials.

[0096] Step 220: Identify the questions of each exercise in the exercise image and the answering area corresponding to the questions.

[0097] Step 230: When it is confirmed that the answering starts, control the shooting device to shoot the answering operation. During the shooting process, obtain the answering content and answering duration of each exercise according to the answering operation, and control the shooting device to shoot the user who is currently answering, and determine the answering posture information and / or answering expression information during the user's answering process based on the obtained user answering image. The user and the answering operation are respectively shot by different cameras in the shooting device.

[0098] When it is confirmed that the answering starts, control the shooting device to respectively shoot the answering operation and the user who is currently answering. Optionally, the shooting device shoots the user at a certain frequency to obtain multiple images. Currently, the images obtained by shooting the user are recorded as user answering images. Multiple user answering images can reflect the state of the user during the answering process. It can be understood that since the answering operation and the user answering image are shot simultaneously, based on the exercise to which the answering operation belongs, the exercise to which the user answering image shot at the same time as the answering operation belongs can be determined. In one embodiment, the answering posture information and / or answering expression information corresponding to the exercise are obtained through each user answering image corresponding to the same exercise. Exemplarily, the answering posture information includes the answering posture and the number of postures. The answering postures include lowering the head, raising the head, propping the chin, scratching the head and ears, etc. It should be noted that parsing the human posture based on the image is an already implemented technical means and will not be described separately here. The number of postures refers to the number of times the user changes the posture during the answering process of the same exercise. The answering expression information refers to the facial expression of the user when answering. It should be noted that obtaining the facial expression of the human face based on the image is an already implemented technical means and will not be described separately here.

[0099] Step 240: Determine the answering emotion information of the user according to the answering posture information and / or answering expression information.

[0100] The answering emotion information can reflect the emotion of the user when answering. The types of emotions included in the answering emotion information can be set according to the actual situation. For example, the answering emotion information includes calm, anxious, irritable, etc. Optionally, when determining the answering emotion information according to the answering posture information, the answering postures and the number of postures corresponding to each answering emotion information can be determined in advance. For example, if the answering postures of a certain exercise currently include raising the head, propping the chin and lowering the head, and the number of postures changes relatively much, the corresponding answering emotion information includes irritable and anxious. Optionally, when determining the answering emotion information according to the answering expression information, the expressions corresponding to each answering emotion information can be determined in advance. Optionally, when determining the answering emotion information according to the answering expression information and the answering posture information, the expressions, answering postures and the number of postures corresponding to each answering emotion information can be determined in advance. It can be understood that in practical applications, other methods can also be used to determine the answering emotion information, such as constructing a neural network, inputting the answering posture information and / or answering expression information into the neural network, and then obtaining the answering emotion information.

[0101] It is understandable that the emotional information of answering questions can also be obtained through other methods, for example, through detection by a wearable device associated with the learning device.

[0102] Step 250: Determine the knowledge mastery level of the corresponding exercise based on the answering emotion information, the answering content and the answering time.

[0103] Exemplarily, when determining the knowledge mastery level, the influence factor also adds the answer emotion information. It can be understood that the current calculation method for determining the knowledge mastery level is the same as the calculation method for the knowledge mastery level described above, except that the answer emotion information is added during the calculation process.

[0104] It is understandable that when photographing the user answering the question, it is also possible to determine whether the user has left the table where the learning device is placed. At this time, the answer statistics method also includes: in the process of photographing the user answering image, continuously detecting whether there is a user currently answering the question in the user answering image; if there is no user currently answering the question, then stop photographing the user currently answering the question and stop photographing the answering operation until the user currently answering the question is detected again, and then resume photographing the user currently answering the question and the answering operation.

[0105] Exemplarily, when shooting the image of the user answering the question, image analysis is performed to determine whether the user who is currently answering the question is included. If included, it means that the user is answering the question. At this time, continue to shoot the user answering the question image, and continue to determine the answering content and answering time. If not included, it means that the user has left. At this time, stop shooting the current answering user and the answering operation, and stop determining the answering content and answering time. Control the shooting device to continue to detect whether the current answering user reappears. If it reappears, resume shooting the current answering user and the answering operation and continue to count the answering content and answering time. Optionally, detecting whether the user answering the question image contains the current answering user can be to detect whether the area occupied by the head of the current answering user in the user answering question image exceeds a set ratio. If it exceeds the set ratio, it is determined that the current answering user is included. In actual applications, other methods can also be used to determine whether the current answering user is included.

[0106] Optionally, in actual applications, a distance sensor may be provided in the learning device to determine whether the user is answering questions at the correct position (such as the position in front of the learning device) through the distance sensor.

[0107] As mentioned above, by adding emotion detection, the knowledge mastery level can be determined more reasonably and accurately, thereby ensuring subsequent effective targeted learning. In addition, by detecting whether the user answering the question image contains the current answering user, the shooting and processing related to the answer statistics can be stopped when the user leaves, saving the equipment resources of the answering device.

[0108] Figure 9The structural schematic diagram of an answering statistics device provided by an embodiment of the present application. Refer to Figure 9 , the device includes: an image acquisition unit 301, a region recognition unit 302, an operation shooting unit 303, and a proficiency determination unit 304.

[0109] The image acquisition unit 301 is configured to acquire an exercise image to be answered, which is obtained by the shooting device of the learning device shooting a paper exercise material; the region recognition unit 302 is configured to recognize the title of each exercise and the corresponding answering region in the exercise image; the operation shooting unit 303 is configured to control the shooting device to shoot the answering operation when starting to answer is confirmed, and during the shooting process, obtain the answering content and answering duration of each exercise according to the answering operation; the proficiency determination unit 304 is configured to determine the knowledge proficiency of the corresponding exercise according to the answering content and answering duration.

[0110] In one embodiment, the operation shooting unit 303 is specifically: when a set action is detected in the picture shot by the shooting device, it is confirmed that the answering starts, and the shooting device is controlled to shoot the answering operation, and during the shooting process, the answering content and answering duration of each exercise are obtained according to the answering operation; or, when a answering start instruction is received, it is confirmed that the answering starts, and the shooting device is controlled to shoot the answering operation, and during the shooting process, the answering content and answering duration of each exercise are obtained according to the answering operation, and the answering start instruction is input by user voice or touch input.

[0111] In one embodiment, the operation shooting unit 303 includes: a shooting confirmation subunit, configured to control the shooting device to shoot the answering operation when starting to answer is confirmed; an answering timing start subunit, configured to start answering timing during the shooting process and determine the current target answering region of the user according to the answering operation; a duration determination subunit, configured to use the current timing result as the answering duration of the exercise corresponding to the target answering region when the answering operation in the target answering region stops being detected; a content determination subunit, configured to obtain the answering content of each exercise according to the answering operation.

[0112] In one embodiment, it further includes: a stop timing unit, configured to start counting the stop duration of the answering operation after using the current timing result as the answering duration of the exercise corresponding to the target answering region; a stop duration acquisition unit, configured to acquire the currently counted stop duration if a new answering operation is detected and the new answering operation is within the target answering region; a duration update unit, configured to add the stop duration to the answering duration and update the answering duration according to the new answering operation.

[0113] In one embodiment, the duration determination subunit includes: an existing determination sub-unit for determining whether there is a corresponding answering duration for the target answering area, and the existing answering duration is statistically obtained based on historical answering operations within the target answering area; an overlay sub-unit for, if there is an existing answering duration, using the overlay result of the existing answering duration and the current timing result as the answering duration for the exercise corresponding to the target answering area.

[0114] In one embodiment, it further includes: a region update unit for, after starting to count the stop duration of the answering operation, if a new answering operation is detected and the new answering operation is in other answering areas, updating the other answering areas as the target answering area; a duration overlay unit for obtaining the answering duration for the exercise corresponding to the target answering area according to the new answering operation and overlaying the stop duration onto the answering duration.

[0115] In one embodiment, the operation shooting unit 303 includes: a shooting confirmation sub-unit for, when confirming the start of answering, controlling the shooting device to shoot the answering operation; a trajectory determination sub-unit for, during the shooting process, determining the writing trajectory and stroke order generated by the answering operation based on multiple shooting images obtained when the shooting device shoots the answering operation; a first content determination sub-unit for determining the first answering content corresponding to the answering operation according to the writing trajectory and stroke order; a second content determination sub-unit for using optical character recognition technology to process the shooting images to obtain the second answering content corresponding to the answering operation; a third content determination sub-unit for obtaining the final answering content of the corresponding exercise according to the first answering content and the second answering content; an answering timing sub-unit for determining the answering duration for each exercise according to the answering operation.

[0116] In one embodiment, it further includes: a search unit for, after identifying the title of each exercise and the corresponding answering area in the exercise image, searching for the corresponding reference result and reference steps in the exercise library according to the title of the exercise; the degree determination unit 304 includes: a step determination sub-unit for obtaining the answering result and answering steps of the corresponding exercise according to the answering content; a grading determination sub-unit for determining the grading result of the corresponding exercise according to the answering result, answering steps, reference result, and reference steps; a mastery determination sub-unit for obtaining the knowledge mastery degree according to the interval to which the answering duration belongs and the grading result.

[0117] In one embodiment, the operation shooting unit 303 is further configured to: control the shooting device to shoot the user who is answering the current question, and determine the answering posture information and / or answering expression information during the user's answering process according to the captured user answering image, where the user and the answering operation are respectively shot by different cameras in the shooting device; the device further includes: an emotion determination unit, configured to determine the user's answering emotion information according to the answering posture information and / or answering expression information; the degree determination unit 304 is specifically configured to: determine the knowledge mastery degree of the corresponding exercise according to the answering emotion information, the answering content, and the answering duration.

[0118] In one embodiment, it further includes: a user detection unit, configured to continuously detect whether the user who is answering the current question exists in the user answering image during the process of shooting the user answering image; a stop unit, configured to stop shooting the user who is answering the current question and stop shooting the answering operation if the user who is answering the current question does not exist, and resume shooting the user who is answering the current question and the answering operation until the user who is answering the current question is detected again.

[0119] The answering statistics device provided in this embodiment is included in the answering statistics device, and is configured to execute the answering statistics method provided in any of the above embodiments, and has corresponding functions and beneficial effects.

[0120] It should be noted that in the embodiments of the above answering statistics device, the included units are only divided according to the functional logic, but are not limited to the above division, as long as the corresponding functions can be realized.

[0121] Figure 10 This is a schematic structural diagram of an answering statistics device provided in an embodiment of the present application. As Figure 10 shown, the answering statistics device includes a processor 30, a memory 31, an input device 32, an output device 33, and a shooting device 34; the number of processors 30 in the answering statistics device may be one or more, Figure 10 taking one processor 30 as an example; the processor 30, the memory 31, the input device 32, the output device 33, and the shooting device 34 in the answering statistics device may be connected through a bus or other means, Figure 10 taking the connection through the bus as an example.

[0122] The memory 31, being a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules in the answering statistics method in the embodiments of the present application (for example, the image acquisition unit 301, region recognition unit 302, operation shooting unit 303, and degree determination unit 304 in the answering statistics device). By running the software programs, instructions, and modules stored in the memory 31, the processor 30 executes various functional applications and data processing of the answering statistics device, that is, implements the answering statistics method provided in any of the above embodiments.

[0123] The memory 31 mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the answering statistics device, etc. In addition, the memory 31 can include high-speed random access memory and can also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage devices. In some instances, the memory 31 can further include a memory remotely set relative to the processor 30, and these remote memories can be connected to the answering statistics device through a network. Examples of the above network include but are not limited to the Internet, enterprise intranet, local area network, mobile communication network, and their combinations.

[0124] The input device 32 can be used to receive input digital or character information and generate key signal inputs related to the user settings and function controls of the answering statistics device. The output device 33 can include devices such as a display screen and a speaker. The shooting device 34 conducts shooting according to the instructions of the processor 30. The learning device can also include a communication device (not shown in the figure), which can be used for data communication with other devices.

[0125] The above answering statistics device includes the answering statistics device provided in the foregoing embodiments, can be used to execute the answering statistics method provided in any embodiment, and has corresponding functions and beneficial effects.

[0126] In addition, the embodiments of the present application further provide a storage medium containing computer-executable instructions, and the computer-executable instructions are used to execute relevant operations in the answering statistics method provided in any embodiment of the present application when executed by a computer processor, and have corresponding functions and beneficial effects.

[0127] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product.

[0128] Therefore, the present application can be implemented in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can be in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in one Figure 1 flow or multiple flows and / or blocks Figure 1 block or multiple blocks. These computer program instructions can also be stored in a computer-readable memory capable of guiding a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implements the functions specified in one Figure 1 flow or multiple flows and / or blocks Figure 1 block or multiple blocks. These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one Figure 1 flow or multiple flows and / or blocks Figure 1 block or multiple blocks.

[0129] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and a memory. The memory may include non-permanent memory in the computer-readable medium, in the form of random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.

[0130] A computer-readable medium includes permanent and non-permanent, removable and non-removable media that can implement information storage by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transitory medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer-readable medium does not include transitory computer-readable media, such as modulated data signals and carrier waves.

[0131] Note that the above is only the preferred embodiment of the present application and the technical principles applied. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein, and various obvious changes, re-adjustments, and substitutions can be made by those skilled in the art without departing from the protection scope of the present application. Therefore, although the present application has been described in more detail through the above embodiments, the present application is not limited to the above embodiments. Without departing from the concept of the present application, it can also include more other equivalent embodiments, and the scope of the present application is determined by the scope of the appended claims.

Claims

1. A method for answering question statistics, characterized in that, Including: Obtaining an exercise image to be answered, where the exercise image is obtained by a photographing device of a learning device photographing a paper exercise material; Identifying the question of each exercise in the exercise image and the corresponding answer area of the question; When it is confirmed to start answering questions, controlling the photographing device to photograph the answering operation, where, including: when a set action is detected in the picture photographed by the photographing device, confirming to start answering questions, and controlling the photographing device to photograph the answering operation, and during the photographing process, obtaining the answering content and answering duration of each exercise according to the answering operation. Obtaining the answering content of each exercise according to the answering operation includes: determining the writing trajectory and stroke order generated by the answering operation according to multiple photographed images obtained when the photographing device photographs the answering operation, determining the first answering content corresponding to the answering operation according to the writing trajectory and stroke order, processing the photographed images by using optical character recognition technology to obtain the second answering content corresponding to the answering operation, and obtaining the final answering content of the corresponding exercise according to the first answering content and the second answering content; Determining the knowledge mastery level of the corresponding exercise according to the answering content and the answering duration.

2. The answer statistics method according to claim 1, characterized in that When it is confirmed to start answering questions, controlling the photographing device to photograph the answering operation includes: When receiving an answering start instruction, confirming to start answering questions, and controlling the photographing device to photograph the answering operation, where the answering start instruction is input by user voice or touch input.

3. The answering statistics method according to claim 1, wherein Obtaining the answering duration of each exercise according to the answering operation includes: Starting an answering timer, and determining the current target answering area of the user according to the answering operation; When it is detected that the answering operation in the target answering area stops, using the current timing result as the answering duration of the exercise corresponding to the target answering area.

4. The answering statistics method according to claim 3, characterized in that, After using the current timing result as the answering duration of the exercise corresponding to the target answering area, it further includes: Starting to count the stop duration of the answering operation; If a new answering operation is detected and the new answering operation is in the target answering area, obtaining the currently counted stop duration; Adding the stop duration to the answering duration, and updating the answering duration according to the new answering operation.

5. The answer statistics method according to claim 3, characterized in that Using the current timing result as the answering duration of the exercise corresponding to the target answering area includes: Determining whether there is a corresponding answering duration for the target answering area, and the existing answering duration is statistically obtained according to the historical answering operations in the target answering area; If there is an existing answering duration, using the superposition result of the existing answering duration and the current timing result as the answering duration of the exercise corresponding to the target answering area.

6. The answer statistics method according to claim 4, wherein After starting to count the stop duration of the answering operation, it further includes: If a new answering operation is detected and the new answering operation is in other answering areas, updating the other answering areas as the target answering area; Obtaining the answering duration of the exercise corresponding to the target answering area according to the new answering operation, and adding the stop duration to the answering duration.

7. The answer statistics method according to claim 1, wherein After identifying the question of each exercise in the exercise image and the corresponding answer area of the question, it further includes: Search for the corresponding reference results and reference steps in the exercise question bank according to the title of the exercise question; The determination of the knowledge mastery level of the corresponding exercise question according to the answer content and the answer duration includes: Obtain the answer result and answer steps of the corresponding exercise question according to the answer content; Determine the marking result of the corresponding exercise question according to the answer result, the answer steps, the reference result and the reference steps; Obtain the knowledge mastery level according to the interval to which the answer duration belongs and the marking result.

8. The answer statistics method according to claim 1, wherein When confirming the start of answering questions, control the shooting device to shoot the answering operation, and during the shooting process, when obtaining the answer content and answer duration of each exercise question according to the answering operation, it also includes: Control the shooting device to shoot the user who is currently answering questions, and determine the answering posture information and / or answering expression information during the user's answering process according to the captured user answering image. The user and the answering operation are respectively shot by different cameras in the shooting device; Determine the answering emotion information of the user according to the answering posture information and / or answering expression information; The determination of the knowledge mastery level of the corresponding exercise question according to the answer content and the answer duration includes: Determine the knowledge mastery level of the corresponding exercise question according to the answering emotion information, the answer content and the answer duration.

9. The answering statistics method according to claim 8, characterized in that It also includes: During the process of shooting the user answering image, continuously detect whether the user who is currently answering questions exists in the user answering image; If the user who is currently answering questions does not exist, stop shooting the user who is currently answering questions and stop shooting the answering operation, and resume shooting the user who is currently answering questions and the answering operation until the user who is currently answering questions is detected again.

10. A question answering statistics device, characterized in that, It includes: An image acquisition unit, configured to acquire an exercise question image to be answered, where the exercise question image is obtained after the shooting device of the learning device shoots a paper exercise material; A region recognition unit, configured to recognize the title of each exercise question in the exercise question image and the answering region corresponding to the title; An operation shooting unit, configured to control the shooting device to shoot the answering operation when confirming the start of answering questions, and during the shooting process, obtain the answer content and answer duration of each exercise question according to the answering operation. Specifically, the operation shooting unit is configured to: when a set action is detected in the picture shot by the shooting device, confirm the start of answering questions, control the shooting device to shoot the answering operation, determine the writing trajectory and stroke order generated by the answering operation according to multiple shooting images obtained when the shooting device shoots the answering operation, determine the first answer content corresponding to the answering operation according to the writing trajectory and stroke order, process the shooting images by using optical character recognition technology, obtain the second answer content corresponding to the answering operation, and obtain the final answer content of the corresponding exercise question according to the first answer content and the second answer content; A degree determination unit, configured to determine the knowledge mastery level of the corresponding exercise question according to the answer content and the answer duration.

11. A question answering statistics device, characterized in that, It includes: One or more processors; A shooting device, configured to shoot according to the instructions of the processor; A memory, configured to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the answering statistics method according to any one of claims 1-9.

12. A storage medium containing computer-executable instructions, characterized in that, The computer-executable instructions are used to execute the answering statistics method according to any one of claims 1-9 when executed by a computer processor.

Images