A method and system for intelligent evaluation of Chinese character writing ability

Abstract

The present application relates to a method and system for intelligent assessment of Chinese character writing ability, comprising the following steps: when a recognition start command is received, writing data sent by a writing recognition terminal is received in real time; the writing data is sent to a writing analysis model, the writing analysis model compares the stroke order data of the Chinese character text with pre-stored standard stroke order information, and outputs a stroke order comparison result; the stroke order comparison result includes correct stroke order information and incorrect stroke order information; the Chinese character text is compared with a preset standard text for contour similarity, and a comparison score is output; if incorrect stroke order information exists, the Chinese character text associated with the incorrect stroke order information is screened out, and the standard stroke order information corresponding to the Chinese character text is associated with the Chinese character text to generate a stroke order correction data packet; the comparison score and the stroke order correction data packet of the Chinese character text are sent to the writing recognition terminal. The present application has the effect of facilitating the judgment of the correctness of each student's writing stroke order.

Description

Technical Field

[0001] The present application relates to the technical field of Chinese character recognition, and in particular to a method and system for intelligently evaluating Chinese character writing ability. Background Art

[0002] At present, Chinese language education in primary schools focuses on the writing of Chinese characters. The Chinese character writing standards involve many aspects such as strokes, radicals, structure, and stroke order. It cultivates students' observation, memory, and concentration, and is conducive to improving their thinking and learning abilities.

[0003] Traditional Chinese character writing assessments mostly rely on manual evaluation, which makes it difficult to judge the correctness of each student's writing stroke order, especially when there are a large number of students. Therefore, it is difficult for teachers to develop targeted training based on the problems found in the test for each student, which needs to be improved. Summary of the Invention

[0004] In order to facilitate the judgment of the correctness of each student's writing stroke order and formulate targeted training based on the problems found in the test for each student, this application provides an intelligent assessment method and system for Chinese character writing ability.

[0005] The above-mentioned invention objective of this application is achieved through the following technical solutions:

[0006] A method for intelligent evaluation of Chinese character writing ability, comprising the steps of:

[0007] When a recognition start instruction is received, the writing data sent by the writing recognition terminal is received in real time, the writing data including the Chinese character text written by the student and the stroke order data of the Chinese character text;

[0008] Sending the writing data to the writing analysis model, the writing analysis model compares the stroke order data of the Chinese character text with the pre-stored standard stroke order information, and outputs the stroke order comparison result, which includes the correct stroke order information and the incorrect stroke order information;

[0009] If there is incorrect stroke order information, the Chinese character text associated with the incorrect stroke order information is filtered out, and the standard stroke order information corresponding to the Chinese character text is associated to generate a stroke order correction data packet;

[0010] Compare the contour similarity of the Chinese text with the preset standard text, and output the writing score based on the stroke order comparison results;

[0011] The writing score and stroke order correction data package of the Chinese character text are sent to the writing recognition terminal.

[0012] By adopting the above technical solution, the writing recognition terminal is responsible for capturing and storing the Chinese text written by students and their stroke order data in real time. When the recognition start instruction is received, the writing data is sent to the writing analysis model. The writing analysis model compares the stroke order data with the pre-stored standard stroke order information corresponding to the Chinese text based on the received Chinese text and the corresponding stroke order data, and thus outputs a stroke order comparison result containing correct stroke order information and incorrect stroke order information, that is, it can pick out the incorrect stroke order and correct stroke order in the complete stroke order sequence of the Chinese text; while checking the stroke order data, the Chinese text written by the stroke order is compared for contour similarity, compared with the preset standard text, and then outputs a writing score combining stroke order and contour. After scoring, the system will identify the Chinese characters associated with incorrect stroke order information and associate them with the standard stroke order information. This information will then be used to generate a stroke order correction data package. This data package and the writing score will be sent back to the writing recognition terminal, allowing students to promptly identify any errors in writing and stroke order. This targeted correction data will assist each student in correcting their incorrect characters and stroke order. This system can determine the correctness of each student's writing stroke order and provide targeted training based on the problems identified during the test.

[0013] In a preferred embodiment of the present application, a handwriting recognition terminal is pre-set with a plurality of writing areas. When a recognition start instruction is received, the handwriting recognition terminal receives writing data in real time, wherein the writing data includes Chinese characters written by a student and stroke order data of the Chinese characters. Specifically, the steps include:

[0014] When the stroke order data of any writing area is picked up, the Chinese character text and stroke order data in the previous writing area of ​​the writing area are stored in the to-be-sent list;

[0015] When a writing completion instruction is received, the Chinese character text and stroke order data in the list to be sent are packaged;

[0016] When a recognition start instruction is received, the Chinese character text and stroke order data sent by the waiting list are received;

[0017] The contour information of each Chinese character is extracted from the received Chinese character text and stroke order data.

[0018] By adopting the above technical solution and setting up multiple writing areas on the writing recognition terminal, data can be collected word by word when students write. This area-based collection method helps to extract contour information more accurately, providing more accurate data support for subsequent scoring and feedback.

[0019] In a preferred embodiment of the present application, when the writing completion instruction is received, before the step of packaging the Chinese character text and stroke order data in the to-be-sent list, the following steps are further included:

[0020] When a writing modification instruction is received, the Chinese character text and stroke order data in the selected writing area are deleted;

[0021] When the modification completion instruction is received, the Chinese character text and stroke order data of the writing area are picked up again;

[0022] Store the newly picked up Chinese character text and stroke order data into the waiting list;

[0023] Based on the writing area corresponding to the Chinese character text, identifying the corresponding Chinese character text to be replaced and stroke order data in the to-be-sent list;

[0024] Replace the Chinese character text and stroke order data.

[0025] By adopting the above technical solution, a convenient way is provided for students to modify written texts. When students find writing errors, they can correct them in time. When students confirm that the writing is correct, the modification completion instruction is sent, and then the next step of writing analysis is carried out. This provides students with an opportunity for self-examination and improves the accuracy of writing scoring judgments.

[0026] In a preferred embodiment of the present application, the step of sending the writing data to the writing analysis model, the writing analysis model comparing the stroke order data of the Chinese character text with pre-stored standard stroke order information, and outputting a stroke order comparison result, wherein the stroke order comparison result includes correct stroke order information and incorrect stroke order information, specifically includes:

[0027] Sending the writing data to the writing analysis model, after which the writing analysis model extracts the Chinese characters in the writing data and the standard stroke order information corresponding to the Chinese characters;

[0028] The stroke order data in the writing data includes a plurality of stroke order texts arranged in a preset order, and the plurality of stroke order texts are compared one by one with the standard stroke order information in the preset order;

[0029] Associating errors where the stroke order text does not match the standard stroke order information to obtain error stroke order information;

[0030] Match the stroke order text with the standard stroke order information to obtain the correct stroke order information;

[0031] A stroke order comparison result is generated, where the stroke order comparison result includes stroke order texts arranged in a preset order and associated with correct or incorrect markers.

[0032] By adopting the above technical solution, the writing analysis model can accurately compare the stroke order, identify and record in detail whether each stroke order is correct or not, providing a clear basis for subsequent corrections. At the same time, by marking incorrect and correct stroke orders, it is convenient to combine the comparison results with the subsequent contour comparison scores, and help students understand the differences between their own writing habits and standards.

[0033] In a preferred example of the present application, if there is erroneous stroke order information, the steps of screening out the Chinese character text associated with the erroneous stroke order information, associating the standard stroke order information corresponding to the Chinese character text, and generating a stroke order correction data packet specifically include:

[0034] Based on the stroke order comparison results, filter out the stroke order texts with incorrect markings;

[0035] Based on the stroke order text with error identification, the corresponding standard stroke order information is extracted from the database;

[0036] A stroke order correction data package is generated, which includes animation data of the standard stroke order of each Chinese character.

[0037] By adopting the above technical solution, a stroke order correction data package is generated according to the comparison results. The stroke order correction data package includes animation data of the standard stroke order of each Chinese character. The animation data package can help students learn the correct stroke order intuitively, thereby improving learning efficiency. In addition, this visual learning material is easier for students to understand and imitate.

[0038] In a preferred example of the present application, the steps of comparing the contour similarity of the Chinese character text with a preset standard text and outputting a writing score based on the stroke order comparison result specifically include:

[0039] Obtaining the area occupied by each Chinese character in the writing area in the writing data, and outputting writing area data based on the ratio of the area of ​​each Chinese character to the area of ​​the writing area;

[0040] When the writing area data is larger than a preset threshold, the Chinese character is corrected by reducing its size and a correction mark is associated with the Chinese character text;

[0041] When the writing area data is smaller than a preset threshold, the size of the Chinese character is corrected by enlarging it, and the Chinese character text is associated with a correction mark;

[0042] Obtain the position of each Chinese character in the writing area, and output writing position data based on the blank area ratio on both sides of the diagonal line in the writing area;

[0043] When the writing position data is greater than or less than a preset threshold, the position of the Chinese character is corrected, the Chinese character text is associated with a correction mark, and a corrected writing sample is output;

[0044] Compare the contour information of each Chinese character in the corrected writing sample with the contour information of the standard text for contour similarity, and output a contour similarity score;

[0045] The contour similarity score is combined with the stroke order comparison result to output the writing score.

[0046] By adopting the above technical solution, before performing contour similarity comparison, the size and position of each Chinese character are first corrected to reduce misrecognition and improve the standardization and fairness of contour similarity scoring. Students can understand their own writing status through the writing scores received by the writing recognition terminal, and make self-corrections accordingly to improve their writing skills.

[0047] In a preferred example of the present application, the step of combining the contour similarity score with the stroke order comparison result to output the writing score specifically includes:

[0048] Get the number of correction marks in each Chinese character text and the number of error marks in the stroke order text of each Chinese character. The number of error marks of Chinese characters associated with the stroke order text is considered to be 0;

[0049] Generate a scoring coefficient based on the number of correction marks in each Chinese character text and the number of error marks in the stroke order text of each Chinese character;

[0050] Combine the contour similarity score of each Chinese character with the scoring coefficient to output the writing score.

[0051] By adopting the above technical solution, a comprehensive writing score is given for each Chinese character by combining the contour similarity score and the stroke order comparison results. This scoring method can more comprehensively evaluate students' writing ability and provide directions for improvement.

[0052] The second object of the present invention is achieved through the following technical solutions:

[0053] An intelligent evaluation system for Chinese character writing ability, comprising:

[0054] The recognition module is used to recognize the student's handwriting data and receive the handwriting scoring and stroke order correction data packets;

[0055] An analysis module is used to compare the stroke order data of the Chinese character text with the pre-stored standard stroke order information and output the stroke order comparison result;

[0056] The correction module is used to correct the contour information of the Chinese text, compare the contour similarity with the preset standard text, and output the contour similarity score;

[0057] The prompt module is used to generate a stroke order correction data packet.

[0058] By adopting the above technical solution, the recognition module is used to recognize the student's writing data and receive the writing score and stroke order correction data package; the analysis module is used to compare the stroke order data of the Chinese character text with the pre-stored standard stroke order information and output the stroke order comparison result; the correction module is used to correct the contour information of the Chinese character text and compare the contour similarity with the preset standard text to output the contour similarity score; the prompt module is used to generate the stroke order correction data package; through the above four modules, the writing data can be quickly recognized, accurately analyzed, and corrected in time, and effective feedback prompts can be provided, which improves the teaching interactivity, facilitates students' self-testing and teachers' guidance, and improves the overall writing teaching effect.

[0059] In summary, this application includes at least one of the following beneficial technical effects:

[0060] 1. The writing recognition terminal is responsible for capturing and storing the Chinese text and stroke order data written by students in real time. When the recognition start command is received, the writing data is sent to the writing analysis model. The writing analysis model compares the stroke order data with the pre-stored standard stroke order information corresponding to the Chinese text based on the received Chinese text and the corresponding stroke order data, and then outputs a stroke order comparison result containing correct stroke order information and incorrect stroke order information, that is, it can pick out the incorrect stroke order and correct stroke order in the complete stroke order sequence of the Chinese text; while checking the stroke order data, the Chinese text written by the stroke order is compared for contour similarity, compared with the preset standard text, and then outputs a writing score combining stroke order and contour. After the scoring is completed, the Chinese characters associated with the incorrect stroke order information will be screened out and associated with the corresponding standard stroke order information to generate a stroke order correction data package. The stroke order correction data package and the writing score are sent back to the writing recognition terminal so that students can promptly know the specific errors in writing and stroke order. Through targeted correction data, each student can be assisted in correcting their incorrect Chinese characters and stroke order. The correctness of each student's writing stroke order can be judged, and targeted training can be immediately formulated based on the problems found in the test for each student.

[0061] 2. It provides students with a convenient way to modify their written texts. When students find errors, they can correct them immediately. When they confirm that the writing is correct, they will receive a modification completion instruction, and then proceed to the next step of writing analysis. This provides students with an opportunity for self-examination and improves the accuracy of writing scoring.

[0062] 3. Generate a stroke order correction data package based on the comparison results. The stroke order correction data package includes animation data of the standard stroke order of each Chinese character. The animation data package can help students learn the correct stroke order intuitively, improving learning efficiency. This visual learning material is easier for students to understand and imitate.

[0063] 4. Before performing contour similarity comparison, the size and position of each Chinese character are corrected to reduce misrecognition and improve the standardization and fairness of contour similarity scoring. Students can understand their own writing conditions through the writing scores received by the writing recognition terminal, and make self-corrections accordingly to improve their writing skills. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 This is a flow chart of an embodiment of an intelligent evaluation method for Chinese character writing ability of the present application;

[0065] Figure 2 This is a flowchart for implementing step S10 in an embodiment of an intelligent evaluation method for Chinese character writing ability of the present application;

[0066] Figure 3 This is a flowchart for implementing step S40 in an embodiment of an intelligent evaluation method for Chinese character writing ability of the present application;

[0067] Figure 4 This is a schematic diagram of an intelligent evaluation system for Chinese character writing ability in this application. DETAILED DESCRIPTION

[0068] The present application is further described in detail below with reference to the accompanying drawings.

[0069] In the following embodiments, Figure 1-3 As shown, this application discloses an intelligent evaluation method for Chinese character writing ability, which specifically includes the following steps:

[0070] S10: When a recognition start instruction is received, writing data sent by the writing recognition terminal is received in real time, where the writing data includes the Chinese character text written by the student and the stroke order data of the Chinese character text;

[0071] In this embodiment, the writing recognition terminal refers to a smart mobile terminal or PC terminal bound to the student's identity information. After the student logs in using the identity information, the writing recognition terminal obtains the student's writing data in real time. The writing data includes the Chinese character text currently written by the student and the stroke order data of the Chinese character text.

[0072] Specifically, students log in to the writing recognition terminal using their identity information and perform Chinese character writing training on the writing recognition terminal. When receiving a recognition start instruction, they receive in real time the Chinese character text currently written by the student and the stroke order data of the Chinese character text from the writing recognition terminal.

[0073] S20: Sending the writing data to the writing analysis model, which compares the stroke order data of the Chinese character text with pre-stored standard stroke order information and outputs a stroke order comparison result, which includes correct stroke order information and incorrect stroke order information;

[0074] In this embodiment, the writing data includes the Chinese text written by the student and the stroke order data of the Chinese text. The writing analysis model will compare the input stroke order data with the pre-set standard stroke order information. This process involves checking the accuracy of each stroke sequence and the starting, turning, and ending positions of the strokes. After the comparison is completed, the writing analysis model will output the stroke order comparison result, which includes correct stroke order information and incorrect stroke order information.

[0075] Specifically, the Chinese character text written by the student and the stroke order data of the Chinese character text are sent to the writing analysis model. The writing analysis model checks the accuracy of the stroke sequence of each Chinese character and the starting, turning and ending positions of the strokes. For incorrect stroke orders, the writing analysis model will point out the wrong part and provide an error mark. Based on the comparison results, the student can get feedback and correct his writing according to the feedback.

[0076] S30: If there is incorrect stroke order information, the Chinese character text associated with the incorrect stroke order information is screened out, and the standard stroke order information corresponding to the Chinese character text is associated with the corrected stroke order information to generate a stroke order correction data packet;

[0077] In this embodiment, in order to provide a correct correction solution, a database containing standard stroke order information is required. This database contains the standard stroke order of all commonly used Chinese characters.

[0078] Specifically, after the writing analysis model identifies incorrect stroke order information, it associates these incorrect stroke orders with the Chinese character text in which they appear, and combines the associated incorrect stroke order information, the corresponding Chinese character text, and the correct standard stroke order information in the database into a stroke order correction data packet. This data packet can be used to generate correction guidance feedback to the user, such as showing the writing errors to the user through the writing recognition terminal, and providing correct stroke order animation demonstrations to help the user correct them.

[0079] S40: Comparing the contour similarity of the Chinese character text with a preset standard text, and outputting a writing score based on the stroke order comparison result;

[0080] In this embodiment, the stroke order comparison result refers to the incorrect stroke order information identified by the writing analysis model, and the writing score is a comprehensive score for the outline and stroke order of the Chinese character. This score can be a simple numerical value or a grading system, such as "excellent", "good", "average", or "poor".

[0081] Specifically, contour information is extracted from the Chinese character text through image processing techniques, such as edge detection or contour recognition algorithms, and the extracted contour information is compared with the standard text in the database for contour similarity. Some shape similarity metrics, such as Euclidean distance, can be used here to evaluate the similarity of the contours. Combined with the incorrect stroke order information associated with the Chinese character, a writing score is output.

[0082] S50: Sending the writing score and stroke order correction data packet of the Chinese character text to the writing recognition terminal.

[0083] In this embodiment, a suitable transmission mode is selected according to the connection mode of the terminal, such as TCP / IP, Bluetooth, Wi-Fi, etc.

[0084] Specifically, the writing score and stroke order correction data packets are packaged into one data packet. This data packet should contain all necessary information so that the writing recognition can understand and process it. The data packet is sent to the writing recognition terminal through the selected transmission method. After receiving the data, the writing recognition terminal parses the data and performs corresponding processing based on the scoring results and stroke order correction information, such as adjusting the display, providing feedback, etc.

[0085] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0086] In one embodiment, step S10 specifically includes the following steps:

[0087] S11: The handwriting recognition terminal is preset with several handwriting areas. When the stroke order data of any handwriting area is picked up, the Chinese character text and stroke order data of the previous handwriting area are stored in a to-be-sent list;

[0088] S12: When the writing completion instruction is received, the Chinese character text and stroke order data in the to-be-sent list are packaged;

[0089] S13: When the recognition start instruction is received, the Chinese character text and stroke order data sent by the to-be-sent list are received;

[0090] S14: Extracting the outline information of each Chinese character from the received Chinese character text and stroke order data.

[0091] In this embodiment, when the user writes on the terminal, he writes each Chinese character in the corresponding writing area; when the writing completion instruction is issued, the writing recognition terminal will package all the stroke order data and Chinese character text in the list to be sent. This process may include sorting, formatting data, etc. to ensure that the sent data is orderly and easy to process.

[0092] Specifically, when the writing recognition terminal detects that the user starts writing in a certain writing area, it will collect the stroke order data and corresponding Chinese character text of the writing area before the area. This information will be temporarily stored in a to-be-sent list, waiting for subsequent processing. When the user completes writing and issues an instruction to complete writing, the writing recognition terminal will package all the stroke order data and Chinese character text in the to-be-sent list. When the recognition start instruction is received, the writing recognition terminal will extract the previously packaged data from the to-be-sent list, and analyze the received stroke order data to extract the outline information of each Chinese character. The whole process is a cyclic iterative process. The user writes in the writing area, the terminal continuously collects data, and processes and recognizes it when the user issues an instruction.

[0093] In one embodiment, the step S102 further includes the following steps:

[0094] S111: When a writing modification instruction is received, the Chinese character text and stroke order data in the selected writing area are deleted;

[0095] S112: When the modification completion instruction is received, re-pick up the Chinese character text and stroke order data of the writing area;

[0096] S113: storing the newly picked up Chinese character text and stroke order data into a waiting list;

[0097] S114: Based on the writing area corresponding to the Chinese character text, identifying the corresponding Chinese character text to be replaced and stroke order data in the to-be-sent list;

[0098] S115: Replace the Chinese character text and stroke order data.

[0099] In this embodiment, when the writing area selected by the modification instruction is the current writing area, that is, the Chinese text to be modified has not yet been stored in the to-be-sent list, steps S114 and S115 do not need to be executed.

[0100] Specifically, after the user issues a modification instruction, the system will identify the writing area that needs to be modified and delete the Chinese text and stroke order data in the writing area. Once the user completes the modification and issues an instruction to complete the modification, the system will start collecting the writing data in the area again, including Chinese text and stroke order data. The system will identify the original writing data that needs to be replaced based on the newly picked up data, and replace the original data in the list to be sent with the newly collected data. The whole process allows the user to make modifications during the writing process and can update the data in the system to reflect the latest writing content.

[0101] In one embodiment, step S20 includes the following steps:

[0102] S21: Send the writing data to the writing analysis model. After receiving the writing data, the writing analysis model extracts the Chinese character text in the writing data and the standard stroke order information corresponding to the Chinese character text.

[0103] S22: The stroke order data in the writing data includes a number of stroke order texts arranged in a preset order. Compare each of the number of stroke order texts with the standard stroke order information in the preset order.

[0104] S23: Associate an error flag with the stroke order text that does not match the standard stroke order information to obtain the incorrect stroke order information.

[0105] S24: Associate a correct flag with the stroke order text that matches the standard stroke order information to obtain the correct stroke order information.

[0106] S25: Generate a stroke order comparison result. The stroke order comparison result includes stroke order texts arranged in a preset order and associated with correct or incorrect flags.

[0107] In this embodiment, the number of stroke order texts arranged in a preset order refers to the contour information of a Chinese character after each additional stroke. These contour information are arranged in the order of writing. For example, the stroke order texts of the character "个" when writing are "丿、人、个"; the standard stroke order text refers to the contour information of a Chinese character after each additional stroke in the standard stroke order information. For example, the standard stroke order text of "个" is "丿、人、个". Compare each stroke order text of this Chinese character with the standard stroke order text one by one, and associate a correct flag to obtain the correct stroke order information.

[0108] Specifically, the system receives the data written by the user, which usually includes Chinese character text and the stroke order information recorded when writing these Chinese characters. The writing analysis model extracts the Chinese character text and the corresponding stroke order information from these data. The system compares the extracted stroke order text with the preset standard stroke order information, which is completed by comparing each step of the stroke order text with the corresponding step of the standard stroke order information one by one. During the comparison process, if it is found that a certain step of the stroke order text does not match the standard stroke order information, the system will associate an error flag with this unmatched stroke order text to mark the incorrect stroke order information. On the contrary, if the step of the stroke order text matches the standard stroke order information, the system will associate a correct flag with this matching stroke order text to mark the correct stroke order information. Finally, the system will generate a stroke order comparison result, which contains all the compared stroke order texts, and each stroke order text is associated with a correct or incorrect flag.

[0109] In one embodiment, step S30 includes the steps:

[0110] S31: Based on the stroke order comparison result, screen out the stroke order texts with incorrect flags.

[0111] S32: Extract the corresponding standard stroke order information from the database based on the stroke order text with error markings;

[0112] S33: Generate a stroke order correction data packet, where the stroke order correction data packet includes the animation data of the standard stroke order for each Chinese character.

[0113] In this embodiment, the stroke order correction data packet includes a stroke order animation area and a practice tracking area. The stroke order animation area will display the correct stroke order. For example, the animation shows the correct writing order at a speed of 2 strokes per second. The practice tracking area records the student's practice history. In addition, if the system finds that the student repeatedly makes mistakes in the stroke "勺" of the character "的", it will automatically adjust the stroke thickness from 2 mm to 5 mm to help the student more clearly identify and practice.

[0114] Specifically, filter out all the stroke order texts marked as incorrect, retrieve the standard stroke order information of each corresponding Chinese character from the database. By matching the Chinese characters in the incorrect stroke order text, the system can find the corresponding standard stroke order and create a stroke order correction data packet, which contains the standard stroke order animation data of each Chinese character that the user needs to correct. These animation data are usually presented in an easy-to-understand form, such as through a series of images or videos, to guide the user on how to correctly write each stroke.

[0115] In one embodiment, step S40 includes the steps: [[ID=​​​​​​​​​​​​​​​​​​S47: Combine the contour similarity score with the stroke order comparison result to output a writing score.

[0123] In this embodiment, if the writing area of ​​a Chinese character is a rectangle of 100x100 pixels and the Chinese character occupies a space of 75x50 pixels, then the area ratio of the Chinese character is 0.75 (75 pixels / 100 pixels). This ratio can be used to compare with other thresholds or standards to determine whether the Chinese character is written too large or too small, and make appropriate corrections accordingly; if the center point of a Chinese character is closer to the left side of the diagonal of the writing area than to the right side, then the blank area ratio may indicate that the Chinese character is written too far to the left. This ratio can be used to compare with other thresholds or standards to determine whether the Chinese character is written too offset, and make appropriate corrections accordingly.

[0124] Specifically, each Chinese character written by the user is analyzed, and the area occupied by the character in the writing area is calculated. The ratio of this area to the total area of ​​the writing area is the writing area data. If the writing area data exceeds the threshold value preset by the system, it means that the Chinese character may be written too large. In this case, the system will adjust the size of the Chinese character to reduce it to an appropriate ratio, and associate a correction mark with the Chinese character text to record this adjustment. If the writing area data is lower than the preset threshold, it may mean that the Chinese character is written too small. The system will enlarge the Chinese character to ensure that it occupies an appropriate area, and similarly associate a correction mark with the Chinese character text; obtain the position of each Chinese character in the writing area, and calculate the blank area ratio on both sides of the diagonal. This ratio is the writing position data. If the writing position data does not meet the preset threshold, the system will correct the position of the Chinese character to ensure that it is located in the appropriate position within the writing area. The corrected Chinese character text will be associated with a correction mark and the corrected writing sample will be output; the contour information of each Chinese character in the corrected writing sample is compared with the contour information of the corresponding Chinese character in the standard text, and a contour similarity score is output. Combining the contour similarity score and the stroke order comparison results, a comprehensive writing score is calculated. This score can be used to evaluate the user's writing quality and provide overall feedback. The score includes a comprehensive consideration of size, position, stroke order and contour to help users understand their writing performance and identify areas that need improvement.

[0125] In one embodiment, step S47 includes the following steps:

[0126] S471: Obtain the number of correction marks in each Chinese character text and the number of error marks in the stroke order text of each Chinese character. A Chinese character with a correct mark associated with its stroke order text is deemed to have a number of error marks of 0;

[0127] S472: generating a scoring coefficient based on the number of correction marks in each Chinese character text and the number of error marks in the stroke order text of each Chinese character;

[0128] S473: Combine the outline similarity score of each Chinese character with the scoring coefficient and output the writing score.

[0129] In this embodiment, the writing score is the product of the contour similarity score and the scoring coefficient. The scoring coefficient is a percentage. The scoring coefficient can reflect the accuracy of Chinese character writing and the correctness of the stroke order. The more correction marks and error marks there are, the smaller the scoring coefficient is, and the final writing score will decrease as the number of correction marks and error marks increases.

[0130] Specifically, based on the text of each Chinese character, correction marks and error marks are searched. Correction marks are used to mark Chinese characters that the system has recognized and adjusted in size, position, etc., while error marks are used to mark errors in stroke order. If a Chinese character has a correct mark in its stroke order text, then the number of its error marks will be regarded as 0 because the stroke order has been correctly executed. According to the number of correction marks in each Chinese character text and the number of error marks in the stroke order text, a scoring coefficient is calculated. The contour similarity score of each Chinese character is multiplied by the generated scoring coefficient to generate a comprehensive writing score. This score can comprehensively reflect the writing quality of Chinese characters in terms of size, position, stroke order, etc.

[0131] In one embodiment, a system for intelligent assessment of Chinese character writing ability is provided, which corresponds one-to-one to the method for intelligent assessment of Chinese character writing ability in the above embodiment. As shown in the figure, the system includes:

[0132] The recognition module is used to recognize the student's handwriting data and receive the handwriting scoring and stroke order correction data packets;

[0133] An analysis module is used to compare the stroke order data of the Chinese character text with the pre-stored standard stroke order information and output the stroke order comparison result;

[0134] The correction module is used to correct the contour information of the Chinese text, compare the contour similarity with the preset standard text, and output the contour similarity score;

[0135] The prompt module is used to generate a stroke order correction data packet.

[0136] For the specific definition of a Chinese character writing ability intelligent assessment system, please refer to the definition of a Chinese character writing ability intelligent assessment method above, which will not be repeated here. Each module in the above-mentioned Chinese character writing ability intelligent assessment system can be implemented in whole or in part by software, hardware, and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0137] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. An intelligent evaluation method for Chinese character writing ability, characterized by: Including steps: When a recognition start instruction is received, the writing data sent by the writing recognition terminal is received in real time, the writing data including the Chinese character text written by the student and the stroke order data of the Chinese character text; Sending the writing data to the writing analysis model, the writing analysis model compares the stroke order data of the Chinese character text with the pre-stored standard stroke order information, and outputs the stroke order comparison result, which includes the correct stroke order information and the incorrect stroke order information; If there is incorrect stroke order information, the Chinese character text associated with the incorrect stroke order information is filtered out, and the standard stroke order information corresponding to the Chinese character text is associated to generate a stroke order correction data packet; The Chinese character text is compared with the preset standard text for contour similarity, and the writing score is output based on the stroke order comparison result. The stroke order comparison result refers to the incorrect stroke order information identified by the writing analysis model. The writing score is a comprehensive score for the contour and stroke order of the Chinese character. The stroke order correction data packet and the writing score are sent back to the writing recognition terminal so that the student can promptly know the specific errors of irregular writing and irregular stroke order, and can assist the student in correcting the incorrectly written Chinese characters and stroke order through targeted corrections made by the stroke order correction data packet; The step of sending the writing data to the writing analysis model, the writing analysis model comparing the stroke order data of the Chinese character text with pre-stored standard stroke order information, and outputting a stroke order comparison result, wherein the stroke order comparison result includes correct stroke order information and incorrect stroke order information, specifically includes: Sending the writing data to the writing analysis model, after which the writing analysis model extracts the Chinese characters in the writing data and the standard stroke order information corresponding to the Chinese characters; The stroke order data in the writing data includes a plurality of stroke order texts arranged in a preset order, and the plurality of stroke order texts are compared one by one with the standard stroke order information in the preset order; Associating errors where the stroke order text does not match the standard stroke order information to obtain error stroke order information; Match the stroke order text with the standard stroke order information to obtain the correct stroke order information; generating a stroke order comparison result, the stroke order comparison result including stroke order texts arranged in a preset order and associated with correct identifiers or incorrect identifiers; The step of comparing the contour similarity of the Chinese character text with the preset standard text and outputting a writing score based on the stroke order comparison result specifically includes: Obtaining the area occupied by each Chinese character in the writing area in the writing data, and outputting writing area data based on the ratio of the area of ​​each Chinese character to the area of ​​the writing area; When the writing area data is larger than a preset threshold, the Chinese character is corrected by reducing its size and a correction mark is associated with the Chinese character text; When the writing area data is smaller than a preset threshold, the size of the Chinese character is corrected by enlarging it, and the Chinese character text is associated with a correction mark; Obtain the position of each Chinese character in the writing area, and output writing position data based on the blank area ratio on both sides of the diagonal line in the writing area; When the writing position data is greater than or less than a preset threshold, the position of the Chinese character is corrected, the Chinese character text is associated with a correction mark, and a corrected writing sample is output; Compare the contour information of each Chinese character in the corrected writing sample with the contour information of the standard text for contour similarity, and output a contour similarity score; The contour similarity score is combined with the stroke order comparison result to output the writing score.

2. The intelligent evaluation method for Chinese character writing ability according to claim 1, characterized in that: The handwriting recognition terminal is pre-set with several writing areas. When a recognition start instruction is received, the handwriting recognition terminal receives writing data in real time, wherein the writing data includes Chinese characters written by the student and stroke order data of the Chinese characters. Specifically, the steps include: When the stroke order data of any writing area is picked up, the Chinese character text and stroke order data in the previous writing area of ​​the writing area are stored in the to-be-sent list; When a writing completion instruction is received, the Chinese character text and stroke order data in the list to be sent are packaged; When a recognition start instruction is received, the Chinese character text and stroke order data sent by the waiting list are received; The contour information of each Chinese character is extracted from the received Chinese character text and stroke order data.

3. The intelligent evaluation method for Chinese character writing ability according to claim 2, characterized in that: The step of packaging the Chinese character texts and stroke order data in the to-be-sent list upon receiving the writing completion instruction further includes the following steps: When a writing modification instruction is received, the Chinese character text and stroke order data in the selected writing area are deleted; When the modification completion instruction is received, the Chinese character text and stroke order data of the writing area are picked up again; Store the newly picked up Chinese character text and stroke order data into the waiting list; Based on the writing area corresponding to the Chinese character text, identifying the corresponding Chinese character text to be replaced and stroke order data in the to-be-sent list; Replace the Chinese character text and stroke order data.

4. The intelligent evaluation method for Chinese character writing ability according to claim 3, characterized in that: If there is erroneous stroke order information, the steps of screening out Chinese character texts associated with the erroneous stroke order information, associating the Chinese character texts with the standard stroke order information corresponding to the erroneous stroke order information, and generating a stroke order correction data packet specifically include: Based on the stroke order comparison results, filter out the stroke order texts with incorrect markings; Based on the stroke order text with error identification, the corresponding standard stroke order information is extracted from the database; A stroke order correction data package is generated, which includes animation data of the standard stroke order of each Chinese character.

5. The intelligent evaluation method for Chinese character writing ability according to claim 1, characterized in that: The step of combining the contour similarity score with the stroke order comparison result to output a writing score specifically includes: Get the number of correction marks in each Chinese character text and the number of error marks in the stroke order text of each Chinese character. The number of error marks of Chinese characters associated with the stroke order text is considered to be 0; Generate a scoring coefficient based on the number of correction marks in each Chinese character text and the number of error marks in the stroke order text of each Chinese character; Combine the contour similarity score of each Chinese character with the scoring coefficient to output the writing score.

6. A Chinese character writing ability intelligent assessment system, which adopts the Chinese character writing ability intelligent assessment method according to any one of claims 1 to 5, characterized in that: include: The recognition module is configured to receive writing data sent by the writing recognition terminal in real time upon receiving a recognition start instruction, recognize the student's writing data, and receive a writing score and stroke order correction data packet, wherein the writing data includes the Chinese character text written by the student and the stroke order data of the Chinese character text; An analysis module is used to send the writing data to a writing analysis model, which compares the stroke order data of the Chinese character text with pre-stored standard stroke order information and outputs a stroke order comparison result, which includes correct stroke order information and incorrect stroke order information; A correction module is used to obtain the area occupied by each Chinese character in the writing data in the corresponding writing area, output the writing area data based on the ratio of the area of ​​each Chinese character to the area of ​​the corresponding writing area, obtain the position of each Chinese character in the writing area, output the writing position data based on the ratio of the blank areas on both sides of the diagonal in the writing area, correct the size of the Chinese character by reducing it when the writing area data is greater than a preset threshold, correct the size of the Chinese character by enlarging it when the writing area data is less than the preset threshold, correct the position of the Chinese character when the writing position data is greater than or less than the preset threshold, and associate the Chinese character text with a correction identifier, compare the contour information of each Chinese character in the corrected writing sample with a preset standard text for contour similarity, and output a contour similarity score; The prompt module is used to filter out Chinese character texts associated with incorrect stroke order information, associate the standard stroke order information corresponding to the Chinese character texts, and generate a stroke order correction data package.

Images