Numerical control teaching system based on operation behavior analysis

Abstract

The invention relates to the technical field of numerical control simulation, in particular to a numerical control teaching system based on operation behavior analysis, which comprises a learning interface connection module, a learning recording module and an operation analysis module, and is characterized in that the operation analysis module comprises an operation interface judgment unit and a student operation behavior analysis unit; in the numerical control teaching system based on operation behavior analysis, the operation interface judgment unit judges the software operation interface of teaching software in the teaching video and the learning video, and the student operates the behavior analysis unit according to the operation change of the software operation interface corresponding to the teaching video and the learning video. The operation interface judgment unit judges whether the student operates along with the teaching video or not, and when the student does not operate along with the teaching video, the student operation behavior analysis unit feeds back the operation behavior of the student and adjusts the playing speed of the teaching video, and through the cooperation of the operation interface judgment unit and the student operation behavior analysis unit, a more personalized learning system is obtained.

Description

Technical Field

[0001] The invention relates to the technical field of numerical control simulation, in particular to a numerical control teaching system based on operation behavior analysis. Background Art

[0002] As one of the core technologies of modern manufacturing, CNC technology plays a key role in many fields such as mechanical processing, automobile manufacturing, aerospace, etc. With the continuous development and wide application of CNC technology, enterprises have an increasing demand for CNC technical talents and have put forward higher requirements for the professional skills and practical ability of talents.

[0003] Due to the breakthrough progress made in computer technology and virtual simulation technology, the emergence of these advanced technologies has provided new ideas and methods for solving the problems in CNC teaching. Through virtual simulation technology, a highly realistic CNC machine tool operating environment can be created on the computer, simulating various actual processing scenarios and fault conditions. Students can practice operations in a virtual environment through operating terminals and teaching videos provided by teachers, which not only avoids the risk of damage to real equipment, but also allows students to obtain a nearly real operating experience.

[0004] However, during the students' operation practice, some students are not proficient in the operation skills, which leads to their inability to follow the teaching video to practice in time. On the other hand, for teachers, it takes extra time and energy to help these backward students, disrupting the original teaching plan. In view of this, we propose a CNC teaching system based on operation behavior analysis. Summary of the invention

[0005] The purpose of the present invention is to solve the problem that the teaching video cannot be adjusted in time during the process of students following the teaching video to practice, resulting in the students being trained.

[0006] To achieve the above-mentioned purpose, the present invention provides a numerical control teaching system based on operation behavior analysis, comprising a learning interface connection module, a learning record module and an operation analysis module; The learning interface connection module is connected to the operation terminal through the system interface to verify the student's identity; The learning record module senses the source code of the operation terminal in the learning interface connection module, reads the source code through the parsing library to sense the corresponding operation elements and the operation text corresponding to the operation elements, identifies the attributes of the operation elements as event processing triggers, and senses the operation behavior of the students in the operation terminal during the student learning process, and stores the operation behavior regularly with the timestamp as the index; The operation analysis module includes an operation interface determination unit and a student operation behavior analysis unit; The operation interface determination unit extracts the operation text in the operation video and the teaching video by optical text recognition technology, determines the operation text in the operation video and the teaching video corresponding to the operation text in the operation element by an accurate matching method, and determines that the area corresponding to the multiple operation texts is the software operation interface; The student operation behavior analysis unit uses an edge feature extraction method to extract the image features of each frame of the corresponding area of ​​the software operation interface in the teaching video, analyzes the operation changes in the video according to the image features of each frame, and records the time nodes of the operation changes; If the operation changes are different, a reminder signal is output to the student operation terminal, and the teaching video is adjusted to the time node with the same operation change. If the operation time nodes are different, the proportional scaling method is used to reduce the playback speed of the teaching video.

[0007] As a further improvement of the technical solution, the learning interface connection module generates a connection request signal in a specific format through a system interface, and sends the connection request signal through a network line; The receiving device of the operating terminal continuously monitors the connection request of the network line. After detecting the connection request signal from the learning interface connection module, it calls out the connection information in the request signal and the connection information stored in itself. If the operation text in the two connection information is exactly the same, it is determined that the learning interface connection module is successfully connected to the operating terminal.

[0008] As a further improvement of the present technical solution, the working principle of the simple operation text string matching calculation is as follows: The login information stored in the operation terminal is randomly selected, starting from the first operation text in the selected login information and the login information in the request signal, and the operation texts are compared one by one until unmatched operation texts are found and the end of the login information is reached. If all the operation texts in the two login information are the same one by one, then the two login information are determined to match; otherwise, they do not match; and the selected login information is marked, and the unmarked login information continues to be selected until all the login information stored in the operation terminal is selected.

[0009] As a further improvement of the technical solution, the parsing library in the learning record module is used to parse the source code into an operable object structure. The specific working principle is as follows: After sensing the source code of the operation terminal in the learning interface connection module, the parsing library scans line by line from the beginning of the source code according to the grammatical rules of HTML, identifies the operation element tags in the code as different operation element nodes, and identifies the attributes as the attributes of the nodes; For the nested relationship between operation elements, the parsing library builds a tree-like hierarchical structure based on the structure of HTML, in which the parent operation element contains the child operation elements. When students interact with the software operating system, the corresponding event learning record module will be triggered. The operation element structure and event processing trigger information established by the parsing library will be monitored to perceive the operation elements clicked by the students.

[0010] As a further improvement of the technical solution, the working principle of the optical character recognition technology in the operation interface determination unit is as follows: Convert each color image in the video into a grayscale image, and each grayscale image into a black and white binary image. The binary formula is based on the threshold value. Method: ; in, It's a pixel The gray value of is the pixel value after binarization; Use median filtering to eliminate noise in each frame of the image, sort the pixel values ​​in the neighborhood around the pixel point, and take the median value as the pixel point value; The histogram equalization method is used to enhance the contrast of the image, and the projections of each frame of the image in the horizontal and vertical directions are calculated to determine the position and boundary of the operation text. The row range and column range of each operation text found are the operation text in the operation video and teaching video.

[0011] As a further improvement of the present technical solution, the precise matching method in the operation interface determination unit compares the operation texts in the operation video, teaching video, and operation elements word by word. When all aspects of the operation texts, such as the character sequence and content, are consistent, they are determined to be the same operation texts.

[0012] As a further improvement of the technical solution, the edge feature extraction method in the student operation behavior analysis unit includes a Sobel operator, which performs convolution operations with the image through convolution kernels in the horizontal and vertical directions to calculate the horizontal and vertical gradients of each pixel point of each frame image in the operation video and the teaching video of the software operation interface; Perception Operation Video Frames and tutorial videos The pixel points in the edge area of ​​the frame are and , the gradient amplitudes of the corresponding pixels are calculated by edge feature extraction method and are and ; like ≠ , it is determined that the operation text clicked in the operation video is different from that in the teaching video.

[0013] As a further improvement of the technical solution, the scaling method in the student operation behavior analysis unit adjusts the playback speed of the teaching video in proportion to the degree of difference in the operation time nodes. The working principle is as follows: The time point for perceiving the same operation changes in the teaching video and the student's terminal operation is and , total duration of teaching videos , Current teaching video playback speed ; like , then it is determined that the teaching video > the time node when the student's operation terminal changes at the same time, That is, the duration of the teaching video being faster than the student operating the terminal during the same operation, so the new playback speed is: .

[0014] As a further improvement of the technical solution, the operation analysis module also includes an operation behavior mining module, which senses the different signals of the operation changes and operation time nodes corresponding to the operation videos and teaching videos in the student operation behavior analysis unit, calls out the operation texts and corresponding time nodes clicked by the students on the operation terminal in the learning record module, and sorts the operation texts according to the order of the time nodes, which are used as keywords for the web crawler search; The web crawler simulates the behavior of real users, sets the request header, sends HTTP to multiple network platforms, obtains the web page content in the network platform, and feeds back to the student operation terminal.

[0015] Compared with the prior art, the present invention has the following beneficial effects: In the numerical control teaching system based on operation behavior analysis, the operation interface determination unit determines the software operation interface of the teaching software in the teaching video and the learning video, and the student operation behavior analysis unit determines whether the student follows the teaching video according to the operation changes of the software operation interface corresponding to the teaching video and the learning video. If the student does not follow the teaching video, the student operation behavior analysis unit gives feedback on the student's operation behavior and adjusts the playback speed of the teaching video. Through the collaboration of the operation interface determination unit and the student operation behavior analysis unit, a more personalized learning experience can be obtained. Students with different skill levels can learn at an appropriate speed according to their own learning progress and ability. Students with slower learning progress will not feel excessive pressure because they cannot keep up, and students with faster learning progress can also get appropriate challenges to avoid boredom due to simple repetition, thereby improving the overall learning effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is the overall module principle diagram of the present invention; Figure 2This is a schematic diagram of the operation analysis module of the present invention.

[0017] The meaning of each number in the figure is: 100. Learning interface connection module; 200. Learning record module; 300. Operation analysis module; 310. Operation interface determination unit; 320. Student operation behavior analysis unit; 330. Operation behavior mining module. DETAILED DESCRIPTION

[0018] The following will be combined with the accompanying drawings in the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] like Figure 1-Figure 2 As shown, a numerical control teaching system based on operation behavior analysis includes a learning interface connection module 100, a learning record module 200 and an operation analysis module 300; The CNC teaching system based on operation behavior analysis aims to provide a highly simulated, low-cost and fully functional training platform for CNC machine tool operation and maintenance teaching through virtual technology, so as to cultivate professional CNC technical talents; In the early stage of CNC technology teaching, practical training mainly relied on real CNC machine tools, and students practiced operation directly on the machine tools. However, real machine tools are expensive, limited in number, and have safety risks, making it difficult to meet the needs of large-scale teaching and diversified teaching scenarios. With the rise of computer technology and virtual technology, in order to solve these problems, CNC teaching systems based on operation behavior analysis came into being. The operation terminal is the key part of the interaction between the CNC teaching system based on operation behavior analysis and the user. The teacher extracts and enters the simulation rules, teaching videos and login information of each student in the operation terminal. The teaching video contains multiple knowledge points, and each knowledge point is closely related to the simulation rules (for example, the knowledge point of explaining the correct operation process corresponds to the normal simulation rule, and the knowledge point of analyzing the consequences of improper operation corresponds to the fault simulation rule). This provides students with a good learning environment on the basis of reducing costs.

[0020] The learning interface connection module 100 is connected to the operation terminal through the system interface, obtains the simulation rules, teaching videos, source code of the operation terminal and login information of each student in the operation terminal, receives the login information input by the students, and compares the login information input with the obtained login information using a simple operation text string matching calculation method. If the login information is the same, the verification is passed and the student's identity is confirmed. If the match fails, the student is prompted to re-enter the login information.

[0021] The learning interface connection module 100 generates a connection request signal in a specific format through the system interface. The request signal includes information such as the student's login information and the purpose of connection, and sends the connection request signal through the network line; The receiving device of the operating terminal continuously monitors the connection request of the network line. After detecting the connection request signal from the learning interface connection module 100, it calls out the connection information in the request signal and the connection information stored in itself. If the operation text in the two connection information is exactly the same, it is determined that the learning interface connection module 100 is successfully connected to the operating terminal.

[0022] The working principle of the simple operation text string matching calculation is as follows: randomly select the login information stored in the operation terminal, start from selecting the login information and the first operation text in the login information in the request signal, and compare the operation texts one by one until unmatched operation texts are found and the end of the login information is reached. If all the operation texts in the two login information are the same one by one, then the two login information are determined to match; otherwise, they do not match; and mark the selected login information, and continue to select unmarked login information until all the login information stored in the operation terminal is selected.

[0023] The learning interface connection module 100 verifies the student identity to ensure that only legitimate students can use the corresponding learning resources. By matching the login information entered by the student with the login information obtained from the operating terminal, it prevents unauthorized users from accessing teaching resources, thereby ensuring the security of the learning system and the rational use of resources.

[0024] When students learn CNC machine tools with the help of teaching videos, the existing technology generally calls out the teaching videos that match the learning chapter information selected by the students for students to use for learning. However, in this process, if the students are not clear about the directory information of the chapter they want to learn, it will cause many inconveniences. This may cause students to fall into confusion in learning, not knowing where to start, and thus wasting a lot of time on finding suitable learning resources. Therefore, the learning record module 200 senses the source code of the operating terminal in the learning interface connection module 100, and reads the source code through the parsing library to sense the corresponding operating elements and the operating text corresponding to the operating elements, and identifies the attributes of the operating elements as event processing triggers. During the student's learning process, the learning record module 200 continuously senses the student's operating behavior in the operating terminal (operating behavior, the behavior of clicking on the operating element), and stores the operating behavior regularly with the timestamp as the index to ensure that the student can continue learning from the last interrupted position when logging in next time; When students interact with the software operating system (for example, clicking a button, entering content in an input box, scrolling a page, etc.), corresponding events are triggered. Each event corresponds to a specific operation element. By monitoring the event, the learning record module 200 senses that the student clicks the operation element and obtains the timestamp corresponding to the operation element. In order to ensure that the student can continue learning from the last interrupted position when logging in next time, the learning record module 200 associates and stores the operation button with the corresponding timestamp. The timestamp is used as the index of the operation, and the operation element and the timestamp are associated and stored, which provides a basis for the student to continue learning from the last interrupted position when logging in next time. It can record the order and time of operation, which is convenient for subsequent data query and recovery of learning progress.

[0025] The learning record module 200 allows students to study according to their own learning rhythm and habits without being affected by interruptions. They can continue previous learning at different times and on different devices, thereby improving the flexibility and convenience of learning. At the same time, after the student's identity is authenticated by the learning interface connection module 100, the learning record module 200 can obtain the source code of the operating terminal, thereby knowing the student's operating behavior in the operating terminal, avoiding the leakage of the source code.

[0026] The parsing library in the learning record module 200 is used to parse the source code into an operable object structure, similar to the process of a browser parsing HTML into a DOM tree, which is performed in a server-side environment. The specific working principle is as follows: After the source code of the operating terminal in the sensing learning interface connection module 100 is read, the parsing library scans the source code line by line starting from the beginning of the source code according to the grammatical rules of HTML, and converts the operation element tags (such as 、 <button>, <input> etc.) as different operation element nodes, and attributes (such as id, class, inline event attribute onclick, etc.) as node attributes; For the nested relationship between operation elements, the parsing library will establish a tree-like hierarchical structure based on the structure of HTML, in which the parent operation element contains the child operation elements. For example, < / button> <button>< / button> , it will <button>Operation elements as Sub-operation elements of an operation element; When students interact with the software operating system, such as clicking a button, inputting content, or scrolling a page, the corresponding event learning recording module 200 will be triggered by parsing the operation element structure and event processing trigger information established by the library, and sensing the operation element clicked by the student by monitoring the event.

[0027] The operation analysis module 300 includes an operation interface determination unit 310 and a student operation behavior analysis unit 320; The operation interface determination unit 310 extracts the operation text in the operation video and the teaching video through the optical text recognition technology, senses the operation text corresponding to the operation element in the learning record module 200, uses the precise matching method to determine the operation text in the operation video and the teaching video corresponding to the operation text in the operation element, marks the operation text in the operation video and the teaching video, and determines that the area corresponding to the multiple operation texts is the software operation interface; The student operation behavior analysis unit 320 uses an edge feature extraction method to extract the image features of each frame of the corresponding area of ​​the software operation interface in the teaching video (the image features are the horizontal and vertical gradients of each pixel of each frame image), analyzes the operation changes in the video according to the image features of each frame (the operation changes are the changes of the operation elements in the software operation interface), and records the time nodes of the operation changes; If the operation changes are different, a reminder signal is output to the student operation terminal, and the teaching video is adjusted to the time node with the same operation change. If the operation time nodes are different, the proportional scaling method is used to reduce the playback speed of the teaching video.

[0028] At the same time, a reminder signal is output to the student operation terminal, which enhances the interactivity between students and teaching videos. The playback speed of the teaching video is adjusted according to the students' operation conditions, making the teaching video more adaptable and able to meet the learning speed and learning habits of different students.

[0029] The working principle of the optical character recognition technology in the operation interface determination unit 310 is as follows: the color image of each frame in the video is converted into a grayscale image for subsequent processing. The purpose is to simplify the color information in each frame of the image into a single grayscale value. The formula is: ; in Respectively represent the red, green, and blue components of the pixels in each frame of the image. Represents the gray value of a pixel; Convert each grayscale image in the video into a black and white binary image, so that the operation text and the background form a sharp contrast, which is convenient for operation text recognition. The binary formula is based on the threshold Method: ; in, It's a pixel The gray value of is the pixel value after binarization; Median filtering is used to eliminate noise in each frame of the image and improve the image quality of each frame. The principle is to sort the pixel values ​​in the neighborhood around the pixel point and take the median value as the value of the pixel point. The formula is: ; in is the pixel value in each frame of the image, is the size of the filter window, is the pixel value after denoising; Use the histogram equalization method to enhance the contrast of the image and make the operation text clearer. The calculation formula is: ; in is the grayscale of the original image, Is grayscale The probability density function of is the gray level after transformation, is the total number of gray levels; Then, the position and boundary of the operation text are determined by calculating the horizontal and vertical projections of each frame of the image; The horizontal projection formula is: ,in is the pixel value of the binary image, It is The horizontal projection value of the row; the vertical projection formula is: ,in It is The vertical projection value of the column; Traversal , find continuous satisfaction The row is the row range where the operation text is located; traverse , find continuous satisfaction The column is the column range where the operation text is located; The row range and column range of each operation text found are the operation text in the operation video and teaching video.

[0030] The precise matching method in the operation interface determination unit 310 compares the operation texts in the operation video, the teaching video, and the operation elements word by word. When the operation texts are consistent in all aspects such as the character sequence and content, they are determined to be the same operation texts. The specific calculation formula is as follows: ; in, Indicates the operation text in the operation element. It refers to the operation text in the operation video or teaching video.

[0031] The edge feature extraction method in the student operation behavior analysis unit 320 includes a Sobel operator, which performs convolution operations with the image through convolution kernels in the horizontal and vertical directions, and calculates the horizontal and vertical gradients of each frame image at each pixel point in the operation video and the teaching video of the software operation interface; Each frame of the image is , the Sobel operator in the horizontal direction is , the Sobel operator in the vertical direction is , then at the pixel The horizontal gradient and vertical gradient They are: ; ; in, and The convolution kernel is defined as: , ; By calculating the magnitude of the gradient and direction Get edge information, the gradient amplitude reflects the strength of the edge, and the direction reflects the direction of the edge: ; ; In operation videos and teaching videos, different operation behaviors will cause the position, shape or outline of the object in the image to change, and the changes will be reflected in the edge features; Perception Operation Video Frames and tutorial videos The pixel points in the edge area of ​​the frame are and , the gradient amplitudes of the corresponding pixels are calculated by edge feature extraction method and are and ; like ≠ , it is determined that the operation text clicked in the operation video is different from that in the teaching video.

[0032] The scaling method in the student operation behavior analysis unit 320 adjusts the playback speed of the teaching video in proportion to the difference in the operation time nodes. The working principle is as follows: The time point for perceiving the same operation changes in the teaching video and the student's terminal operation is and , total duration of teaching videos , Current teaching video playback speed ; like , then it is determined that the teaching video > the time node when the student's operation terminal changes at the same time, That is, the duration of the teaching video being faster than the student operating the terminal during the same operation, so the new playback speed is: .

[0033] The operation analysis module 300 also includes an operation behavior mining module 330. The operation behavior mining module 330 senses different signals of operation changes and operation time nodes corresponding to the operation video and teaching video in the student operation behavior analysis unit 320, calls out the operation texts and corresponding time nodes clicked by the student on the operation terminal in the learning record module 200, and sorts the operation texts according to the order of the time nodes, which are used as keywords for the web crawler search; Web crawlers simulate the behavior of real users, set request headers, including User-Agent, Referer and other information, send HTTP to multiple network platforms, obtain web page content (reasons for changes in operation text) in the network platform, and feed it back to the student operation terminal, mining operation text changes from the network platform to provide students with an additional source of knowledge. Since teaching videos and operation terminals may not cover detailed explanations of all operations, web crawlers can search for more extensive information, enrich students' learning resources, and help students better understand the principles and reasons for the operation, and improve the depth and breadth of students' understanding of the operation. When students encounter difficulties during the operation or are confused about the principles of the operation, the information searched on the Internet can help them overcome the difficulties.

[0034] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents. < / button>

Claims

1. A numerical control teaching system based on operation behavior analysis, characterized in that: It comprises a learning interface connection module (100), a learning record module (200) and an operation analysis module (300); The learning interface connection module (100) is connected to the operation terminal via a system interface to verify the identity of the student; The learning recording module (200) senses the source code of the operation terminal in the learning interface connection module (100), reads the source code through a parsing library to sense the corresponding operation elements and the operation text corresponding to the operation elements, identifies the attributes of the operation elements as event processing triggers, senses the operation behavior of the students in the operation terminal during the student learning process, and stores the operation behavior regularly with the timestamp as the index; The operation analysis module (300) comprises an operation interface determination unit (310) and a student operation behavior analysis unit (320); the operation interface determination unit (310) extracts operation texts in operation videos and teaching videos by using optical text recognition technology, uses an accurate matching method to determine the operation texts in the operation videos and teaching videos corresponding to the operation texts in the operation elements, and determines that the area corresponding to the multiple operation texts is the software operation interface; The student operation behavior analysis unit (320) uses an edge feature extraction method to extract image features of each frame of the teaching video corresponding to the area of ​​the software operation interface, analyzes the operation changes in the video according to the image features of each frame, and records the time nodes of the operation changes. If the operation changes are different, a reminder signal is output to the student operation terminal, and the teaching video is adjusted to the time nodes with the same operation changes. If the operation time nodes are different, a scaling method is used to reduce the playback speed of the teaching video.

2. The numerical control teaching system based on operation behavior analysis according to claim 1 is characterized in that: The learning interface connection module (100) generates a connection request signal in a specific format through a system interface, and sends the connection request signal through a network line; The receiving device of the operating terminal continuously monitors the connection request of the network line, and after detecting the connection request signal from the learning interface connection module (100), calls out the connection information in the request signal and the connection information stored in the device itself, and if the operation text in the two connection information is exactly the same, it is determined that the learning interface connection module (100) is successfully connected to the operating terminal.

3. The numerical control teaching system based on operation behavior analysis according to claim 2 is characterized in that: The working principle of the learning interface connection module (100) is as follows: The login information stored in the operation terminal is randomly selected, starting from the first operation text in the selected login information and the login information in the request signal, and the operation texts are compared one by one until unmatched operation texts are found and the end of the login information is reached. If all the operation texts in the two login information are the same one by one, then the two login information are determined to match; otherwise, they do not match; and the selected login information is marked, and the unmarked login information continues to be selected until all the login information stored in the operation terminal is selected.

4. The numerical control teaching system based on operation behavior analysis according to claim 3 is characterized in that: The parsing library in the learning record module (200) is used to parse the source code into an operable object structure. The specific working principle is as follows: After sensing the source code of the operation terminal in the learning interface connection module (100), the parsing library scans the source code line by line starting from the beginning according to the grammatical rules of HTML, identifies the operation element tags in the code as different operation element nodes, and identifies the attributes as the attributes of the nodes; For the nested relationship between operation elements, the parsing library builds a tree-like hierarchical structure based on the structure of HTML, in which the parent operation element contains the child operation elements. When the student interacts with the software operating system, the corresponding event learning recording module (200) will be triggered to monitor the event and sense the operating element clicked by the student through the operation element structure and event processing trigger information established by the parsing library.

5. The numerical control teaching system based on operation behavior analysis according to claim 1 is characterized in that: The working principle of the optical character recognition technology in the operation interface determination unit (310) is as follows: Convert each color image in the video into a grayscale image, and each grayscale image into a black and white binary image. The binary formula is based on the threshold value. Method: ； in, It's a pixel The gray value of is the pixel value after binarization; Use median filtering to eliminate noise in each frame of the image, sort the pixel values ​​in the neighborhood around the pixel point, and take the median value as the pixel point value; The histogram equalization method is used to enhance the contrast of the image, and the projections of each frame of the image in the horizontal and vertical directions are calculated to determine the position and boundary of the operation text. The row range and column range of each operation text found are the operation text in the operation video and teaching video.

6. The numerical control teaching system based on operation behavior analysis according to claim 5 is characterized in that: The precise matching method in the operation interface determination unit (310) compares the operation texts in the operation video, the teaching video, and the operation elements word by word, and when all aspects such as the character sequence and content of the operation texts are consistent, they are determined to be the same operation texts.

7. The numerical control teaching system based on operation behavior analysis according to claim 6 is characterized in that: The edge feature extraction method in the student operation behavior analysis unit (320) includes a Sobel operator, which performs convolution operations with the image through convolution kernels in the horizontal and vertical directions to calculate the horizontal and vertical gradients of each frame image at each pixel point in the operation video and the teaching video of the software operation interface; Perception Operation Video Frames and tutorial videos The pixel points in the edge area of ​​the frame are and , the gradient amplitudes of the corresponding pixels are calculated by edge feature extraction method and are and ; like ≠ , it is determined that the operation text clicked in the operation video is different from that in the teaching video.

8. The numerical control teaching system based on operation behavior analysis according to claim 1 is characterized in that: The scaling method in the student operation behavior analysis unit (320) proportionally adjusts the playback speed of the teaching video according to the degree of difference in the operation time nodes. The working principle is as follows: The time point for perceiving the same operation changes in the teaching video and the student's terminal operation is and , total duration of teaching videos , Current teaching video playback speed ; like , then it is determined that the teaching video > the time node when the student's operation terminal changes at the same time, That is, the duration of the teaching video being faster than the student operating the terminal during the same operation, so the new playback speed is: .

9. The numerical control teaching system based on operation behavior analysis according to claim 8 is characterized in that: The operation analysis module (300) further comprises an operation behavior mining module (330), wherein the operation behavior mining module (330) senses different signals of operation changes and operation time nodes corresponding to the operation video and the teaching video in the student operation behavior analysis unit (320), retrieves the operation texts and corresponding time nodes clicked by the student on the operation terminal in the learning record module (200), and sorts the operation texts according to the sequence of the time nodes, and uses the operation texts as keywords for the web crawler search; The web crawler simulates the behavior of real users, sets the request header, sends HTTP to multiple network platforms, obtains the web page content in the network platform, and feeds back to the student operation terminal.

Images