Teaching evaluation system based on sensing device

Abstract

The invention provides a teaching evaluation system based on a sensing device, and relates to the technical field of teaching evaluation. The data processing module is used for preprocessing the original sensing data; the feature extraction and analysis module is used for extracting various behavior features of teachers and students from the video data; the teaching evaluation module is used for evaluating the teaching process of the teacher, the learning state and effect of the students and the teaching effect and atmosphere of the whole classroom according to the extracted features and data; the system comprises a back-propagation (BP) neural network, a result display and feedback module and a system management and configuration module, the result display and feedback module is used for managing users of the system and allowing a system administrator to configure various parameters of the system, the teaching ability of teachers is evaluated through the BP neural network, and the BP neural network is trained based on a large amount of actual teaching data; the influence of subjective factors in manual evaluation is reduced, and the reliability and credibility of the evaluation result are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of teaching evaluation, and particularly to a teaching evaluation system based on a sensing device. Background Art

[0002] A teaching evaluation system based on a sensing device is a system that uses modern technical means, such as sensors, cameras, artificial intelligence, etc., to collect and analyze data on students' behaviors and performances during the learning process, so as to evaluate the teaching effect. This system aims to provide a more objective and comprehensive teaching evaluation, help teachers understand the learning status of students, optimize teaching methods, and improve teaching quality.

[0003] Currently, a teaching evaluation system based on a sensing device comprehensively evaluates teachers' teaching from multiple dimensions through a variety of advanced technologies and devices. Such a system uses sensing devices such as sensors, cameras, and microphones to collect data inside and outside the classroom, and mines valuable information through data analysis and machine learning algorithms to comprehensively evaluate teachers' teaching performances. And the sensing device can provide objective data. However, when determining the evaluation indicators and their weights, human participation is often required. For example, when deciding the importance ratio of indicators in aspects such as teaching interaction, content explanation, and classroom management, the personal experiences and educational concepts of relevant personnel such as education administrators and experts will have an impact. Personnel who tend to traditional teaching may attach more importance to the weight of knowledge explanation, while those who advocate innovative teaching may increase the proportion of classroom interaction, resulting in an increase in the influence of subjective factors in the artificial evaluation when evaluating teachers' teaching abilities, and reducing the reliability and credibility of the evaluation results. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides a teaching evaluation system based on a sensing device.

[0005] A teaching evaluation system based on a sensing device provided by the present invention includes a sensing data acquisition module for capturing sensing data on students' facial expression states, classroom performances, and interaction behaviors;

[0006] A data processing module for preprocessing the original sensing data;

[0007] A feature extraction and analysis module for extracting various behavior features of teachers and students from video data;

[0008] The teaching evaluation module is used to evaluate the teaching process of teachers, the learning status and effects of students, and the teaching effects and atmosphere of the entire classroom based on various extracted features and data. Among them, the teaching evaluation module includes a teacher teaching evaluation unit, a student learning evaluation unit, and a classroom overall evaluation unit. The teacher teaching evaluation unit is used to evaluate the teaching ability of teachers through a BP neural network. This teacher teaching evaluation unit sorts and quantifies the data on teachers' teaching behaviors in the collected perception data, converts it into input data suitable for processing by the BP neural network, and trains the BP neural network based on similar teaching scenarios. The student learning evaluation unit is used to evaluate the learning status and effects of students, and the classroom overall evaluation unit is used to evaluate the teaching effects and atmosphere of the entire classroom;

[0009] The result display and feedback module is responsible for presenting the analyzed and processed data to teachers, students, and administrators in an intuitive and easy-to-understand manner;

[0010] The system management and configuration module is responsible for managing the users of the system and allowing system administrators to configure the various parameters of the system.

[0011] Preferably, when the teacher teaching evaluation unit evaluates the teaching ability of teachers through a BP neural network, it first constructs a BP neural network, selects a suitable theoretical model, establishes the mapping relationship between various indicators and data, collects the corresponding original data, conducts data analysis and preprocessing, and then trains the BP neural network. These data are normalized, grouped according to the needs of neural network training, and the BP neural network parameters are set for training, testing, and verification. Finally, through the adjustment of the search algorithm, the evaluation model with the best evaluation effect is found.

[0012] Preferably, the student learning evaluation unit is used to evaluate the learning status and effects of students, and provides personalized learning suggestions and feedback for each student through comprehensive analysis of various aspects such as students' behaviors, expressions, and learning achievements.

[0013] Preferably, the classroom overall evaluation unit is used to evaluate the teaching effects and atmosphere of the entire classroom. The evaluation scope includes the quality of teacher-student interaction, classroom order, and learning atmosphere. By comprehensively analyzing various indicators, it gives an overall evaluation and suggestions for the classroom, providing a reference for teaching improvement.

[0014] Preferably, the structure display and feedback module includes a visualization display unit, a real-time feedback unit, and a report generation unit. The visualization display unit is used to display the results of teaching evaluation to teachers, students, and education administrators in the form of intuitive charts, graphs, and reports. The real-time feedback unit is used to provide feedback information to teachers and students in real time. The report generation unit is used to generate a detailed evaluation report based on the results of teaching evaluation.

[0015] Preferably, the visualization display unit is responsible for converting complex analysis data into an intuitive and easy-to-understand visual form to help teachers, students, and administrators quickly obtain valuable information.

[0016] Preferably, the real-time feedback unit is responsible for providing feedback information to teachers and students in real time during classroom teaching, enabling teachers to adjust teaching strategies in a timely manner and students to understand their learning situation and make adjustments in a timely manner.

[0017] Preferably, the report generation unit is responsible for generating a detailed evaluation report based on the results of teaching evaluation. The evaluation report includes the advantages and disadvantages of classroom teaching, an analysis of students' learning situations, and teaching suggestions, providing a basis for teachers' teaching reflection and professional development, and also providing a reference for education administrators' teaching management and decision-making.

[0018] Preferably, the perception data collection module includes an intelligent camera, a microphone array, and an environmental sensor. The intelligent camera is installed at the front and back corners or on the ceiling of the classroom to clearly record the facial expressions and body movements of teachers and students. The microphone array is distributed at different positions in the classroom.

[0019] Compared with related technologies, a teaching evaluation system based on a perception device provided by the present invention has the following beneficial effects:

[0020] The present invention evaluates the teaching ability of teachers through a BP neural network. First, a BP neural network is constructed, and a suitable theoretical model is selected to establish the mapping relationship between various indicators and data. The corresponding raw data is collected, and data analysis and preprocessing are carried out. Then, the BP neural network is trained, and the data is normalized and grouped according to the needs of neural network training, and the BP neural network parameters are set for training, testing, and verification. Finally, through the adjustment of the search algorithm, the evaluation model with the best evaluation effect is found, and the training effect of the model is judged by the coefficient of determination and the relative error. The BP neural network is trained based on a large amount of actual teaching data, reducing the influence of subjective factors in manual evaluation and improving the reliability and credibility of the evaluation results. Brief Description of the Drawings

[0021] Figure 1 is a structural block diagram of a teaching evaluation system based on a perception device provided by the present invention;

[0022] Figure 2 This is a flowchart for evaluating a teacher's teaching ability through a BP neural network provided by the present invention. Specific implementation manner

[0023] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0024] Please refer to Figure 1 and Figure 2 , where Figure 1 is a structural block diagram of a teaching evaluation system based on a perception device provided by the present invention; Figure 2 is a flowchart for evaluating a teacher's teaching ability through a BP neural network provided by the present invention.

[0025] In the specific implementation process, as Figure 1 and Figure 2 shown, a teaching evaluation system based on a perception device includes a perception data acquisition module for capturing perception data of students' facial expressions, classroom performance, and interaction behaviors;

[0026] It should be noted that the perception data acquisition module includes an intelligent camera, a microphone array, and an environmental sensor. The intelligent camera is installed at the front and back corners or on the ceiling of the classroom to clearly record the facial expressions and body movements of teachers and students. The microphone array is distributed at different positions in the classroom;

[0027] Moreover, the data acquisition content includes behavioral data, facial expression data, voice data, and environmental data. Behavioral data is used to collect the teaching behaviors of teachers through an intelligent camera, such as the gesture actions of teachers (gestures indicating blackboard content, emphasizing key points, etc.), body postures (standing, walking, bending down to guide students), and the position movement trajectory in the classroom. These behavioral data can be used to analyze the teaching style and teaching investment degree of teachers. At the same time, the collection of students' behaviors includes students' sitting postures (upright, bending, lying on the table), raising hands (raising hand frequency, raising hand duration), and classroom participation behaviors (actions in group discussions, actions interacting with teachers). These data can help evaluate students' learning enthusiasm and classroom participation;

[0028] Facial expression data, the camera captures the facial expressions of teachers, such as expression changes like smiling, frowning, and being serious. The expressions of teachers can reflect their teaching emotions and attitudes towards teaching content. At the same time, the facial expressions of students are captured, such as confused expressions (frowning, blank looks in the eyes, etc.), interested expressions (bright eyes, smiling faces, etc.), and tired expressions (yawning, dull looks in the eyes, etc.). These expressions can directly reflect students' understanding degree and learning interest in teaching content;

[0029] Voice data. The microphone collects the teacher's voice content, including voice features such as speech rate, intonation, and volume. The speech rate can reflect the rhythm of the teacher's explanation, the change in intonation can reflect the key points and difficulties of the teaching content, and the volume is related to the teacher's teaching passion and classroom control ability. The students' voice data mainly includes voices such as asking questions, answering questions, and group discussions. By analyzing the students' voice content and voice features, the thinking activity and language expression ability of the students can be understood;

[0030] Environmental data. The temperature and humidity data collected by environmental sensors can be used to evaluate the comfort of the classroom environment. An appropriate temperature and humidity range (generally, the temperature is between 20-25 degrees Celsius and the humidity is between 40%-60%) helps students concentrate on learning. The light intensity data is very important for evaluating the visual environment of the classroom. Too strong or too weak light will have an adverse impact on the students' eyesight and also affect the use effect of teaching equipment (such as projectors);

[0031] The data processing module is responsible for preprocessing the original perception data;

[0032] It should be noted that the specific process of data preprocessing includes the following steps:

[0033] Step 1: Data cleaning. During the data collection process, due to various reasons (such as equipment failures, environmental interference, etc.), noise data will be generated. The data cleaning process will identify and remove this noise data by setting thresholds and filtering algorithms. For video data, median filtering can be used to remove salt-and-pepper noise. For audio data, a low-pass filter can be used to remove high-frequency static noise;

[0034] Step 2: The data processing module needs to convert these different formats of data into a format that the system can uniformly process. Usually, the video is converted into a frame sequence format that is convenient for analysis (such as extracting and storing the video frame by frame as an image sequence), and the audio is converted into a digital signal array format. At the same time, the data is standardized so that different types of data are within the same dimension range;

[0035] The feature extraction and analysis module is used to extract various behavior features of teachers and students from the perception data;

[0036] It should be noted that the feature extraction and analysis module first performs behavior feature extraction and analysis. Through the video data collected by the intelligent camera, target tracking algorithms (such as Kalman filtering algorithm, target tracking algorithm based on deep learning) are used to track the position changes of the teacher in the classroom. At the same time, their position changes can be tracked through the intelligent camera;

[0037] Analysis of action frequency and duration, count the occurrence times of various actions of teachers and students in the classroom (such as raising hands, standing, walking, nodding, etc.), and calculate the action frequency per unit time. For teachers, frequent gesture actions may indicate that their explanations are vivid and vivid. For students, a higher frequency of raising hands usually means active participation in the classroom question-and-answer session;

[0038] Facial expression feature extraction and analysis, application of facial expression recognition algorithm, adopt advanced facial expression recognition models, such as models based on convolutional neural network (CNN). These models are trained with a large amount of face image data and can recognize various basic expressions, including happiness, sadness, anger, surprise, disgust, fear and neutral. In the teaching scenario, the facial images of teachers and students collected by the camera will be input into these models, and the model outputs the corresponding expression categories. At the same time, in order to improve the accuracy of expression recognition, factors such as lighting conditions and angles need to be considered;

[0039] Dynamic analysis of expression changes, analyze the relationship between expression changes and teaching links. At the same time, when the teacher explains difficult knowledge, pay attention to the occurrence frequency and duration of students' confused expressions in order to understand whether the teaching difficulties are effectively explained;

[0040] Speech feature extraction and analysis, basic speech parameter analysis, through speech processing technology, split the audio data into speech segments, calculate the speech rate of each segment, and the speech rate can be determined by counting the number of syllables per unit time, and analyze the intonation changes of the speech, that is, the change of fundamental frequency. Signal processing methods (such as autocorrelation function method, harmonic peak method) can be used to extract the fundamental frequency, and at the same time measure the volume of the speech, which can be measured by the amplitude of the audio signal;

[0041] Speech emotion and semantic analysis, using speech emotion recognition technology, classify speech emotions into categories such as positive, negative and neutral, which involves the comprehensive analysis of prosodic features (such as intonation, speech rate, volume) and timbre features (such as harmonic structure, formant, etc.) of the speech;

[0042] Teaching evaluation module, used to evaluate the teaching process of teachers, the learning status and effects of students, and the teaching effects and atmosphere of the entire classroom according to the extracted various features and data. Among them, the teaching evaluation module includes a teacher teaching evaluation unit, a student learning evaluation unit, and a classroom overall evaluation unit. The teacher teaching evaluation unit is used to evaluate the teaching ability of teachers through a BP neural network. The teacher teaching evaluation unit sorts out and quantifies the data on teachers' teaching behaviors in the collected perception data, converts it into input data suitable for BP neural network processing, and trains the BP neural network based on similar teaching scenarios. The student learning evaluation unit is used to evaluate the learning status and effects of students, and the classroom overall evaluation unit is used to evaluate the teaching effects and atmosphere of the entire classroom;

[0043] It should be noted that when the teacher teaching evaluation unit evaluates the teaching ability of teachers through the BP neural network, it first constructs the BP neural network, selects a suitable theoretical model, establishes the mapping relationship between each index and the data, collects the corresponding original data, analyzes and preprocesses the data, and then trains the BP neural network. These data are normalized and grouped according to the needs of neural network training, and the parameters of the BP neural network are set for training, testing and verification. Finally, through the adjustment of the search algorithm, the evaluation model with the best evaluation effect is found;

[0044] The student learning evaluation unit is used to evaluate the learning status and effects of students, and provides personalized learning suggestions and feedback for each student through comprehensive analysis of various aspects such as the behaviors, expressions and learning achievements of students.

[0045] The overall classroom evaluation unit is used to evaluate the teaching effect and atmosphere of the entire classroom. The evaluation scope includes the quality of teacher-student interaction, classroom order and learning atmosphere. Through comprehensive analysis of various indicators, it gives an overall evaluation and suggestions for the classroom, providing a reference for teaching improvement;

[0046] Moreover, the specific steps for evaluating the teaching ability of teachers through the BP neural network include the following:

[0047] S1. Data collection and preprocessing. Combining the teaching syllabus, educational objectives and actual teaching situations, determine the specific indicators for evaluating the teaching ability of teachers. At the same time, collect data related to the evaluation indicators through various channels. Finally, clean the collected data, remove duplicate, incorrect or incomplete data records to ensure the accuracy and integrity of the data;

[0048] It should be noted that the data related to the evaluation indicators include students' teaching evaluation data, classroom observation records, teaching achievement test scores and teachers' teaching reflection reports;

[0049] S2. Construct a BP neural network model. According to the quantity and complexity of the evaluation indicators, determine the structural parameters of the number of layers and neurons of the BP neural network, and then select a suitable activation function;

[0050] It should be noted that the number of neurons in the input layer is the same as the number of evaluation indicators, the number of neurons in the output layer is 1, representing the comprehensive evaluation result of the teaching ability of teachers, and the hidden layer can be set to 1-2 layers, and the number of neurons is determined according to experience and experiments;

[0051] S3. Model training: The preprocessed data is divided into a training set and a test set according to a ratio. The sample data in the training set is input into the BP neural network. Starting from the input layer, it passes through the hidden layers in sequence. Meanwhile, the data is normalized. The outputs of each neuron are calculated based on the current weights and thresholds. Finally, the prediction results of the output layer are obtained, and the prediction results of the output layer are compared with the actual teacher teaching ability evaluation values. The error between the two is calculated. Based on the error, starting from the output layer, the error is backpropagated, the error gradients of each layer of neurons are calculated, and the weights and thresholds of each layer are adjusted so that the error changes in the direction of decrease. The processes of forward propagation, error calculation, and backpropagation are repeated, and the weights and thresholds are continuously adjusted until the error reaches the preset accuracy requirement or the maximum number of iterations is reached;

[0052] It should be noted that data normalization is essentially a linear transformation. The good properties of linear transformation determine that the data will not be "invalidated" after being changed, but can improve the performance of the data. Therefore, normalization has a very important property. At the same time, the commonly used normalization methods are mainly the following two:

[0053] The first one is the maximum-minimum method, and its function form is as follows:

[0054] XK = (XK - X min ) / (X max - X min )

[0055] In the formula, X min is the minimum value in the training data, and X max is the maximum value in the training data;

[0056] The second one is the mean-variance method, and its function form is as follows:

[0057] XK = (XK - X mean ) / X var

[0058] In the formula, X mean is the mean value in the training data, and X var is the variance in the training data;

[0059] S4. Model evaluation and optimization: Use the test set data to evaluate the trained BP neural network model, and calculate the evaluation indicators such as the accuracy rate, recall rate, and mean square error of the model to determine whether the performance of the model meets the requirements;

[0060] S5. Teaching ability assessment: After processing the teacher-related data to be evaluated according to the preprocessing method, it is input into the optimized BP neural network model. After calculation by the model, the comprehensive evaluation value of the teacher's teaching ability is obtained at the output layer. According to the preset evaluation criteria, the evaluation value is converted into the corresponding evaluation level, and the evaluation results are analyzed to find out the advantages and disadvantages of the teacher's teaching ability, and provide targeted improvement suggestions and feedback for the teacher;

[0061] Result display and feedback module, which is responsible for presenting the analyzed and processed data to teachers, students and administrators in an intuitive and easy-to-understand way;

[0062] It should be noted that the structure display and feedback module includes a visualization display unit, a real-time feedback unit and a report generation unit. The visualization display unit is used to display the results of teaching evaluation to teachers, students and education administrators in the form of intuitive charts, graphs and reports. The real-time feedback unit is used to provide feedback information to teachers and students in real time. The report generation unit is used to generate a detailed evaluation report according to the results of teaching evaluation;

[0063] The visualization display unit is responsible for converting complex analysis data into an intuitive and easy-to-understand visual form to help teachers, students and administrators quickly obtain valuable information.

[0064] The real-time feedback unit is responsible for providing feedback information to teachers and students in real time during classroom teaching, enabling teachers to adjust teaching strategies in a timely manner and students to understand their learning situation in a timely manner and make adjustments;

[0065] The report generation unit is responsible for generating a detailed evaluation report according to the results of teaching evaluation. The evaluation report includes the advantages and disadvantages of classroom teaching, the analysis of students' learning situation, and teaching suggestions, providing a basis for teachers' teaching reflection and professional development, and also providing a reference for education administrators' teaching management and decision-making;

[0066] System management and configuration module, which is responsible for managing the users of the system and allowing system administrators to configure the parameters of the system;

[0067] It should be noted that user management includes user registration and login, permission management and user information maintenance. User registration and login are responsible for handling the registration process of system users, including collecting users' basic information, assigning account passwords, etc., and at the same time managing users' login operations and verifying users' identity information to ensure that only authorized users can access the system;

[0068] Permission management is used to assign different operation permissions according to the roles of users (such as administrators, teachers, students, etc.);

[0069] User information maintenance is used for administrators or the users themselves to modify and update user information;

[0070] System parameter configuration includes evaluation index setting, evaluation period setting, and data collection frequency configuration. Evaluation index setting: Administrators can define and adjust various indicators and their weights for teaching evaluation in this module;

[0071] Evaluation period setting: The evaluation period for teaching evaluation can be set according to the school's teaching arrangements and management requirements;

[0072] Data collection frequency configuration: For the data collection function of the sensing device, set the data collection frequency. Depending on different teaching scenarios and data requirements, determine whether to collect data in real time, at regular intervals, or trigger collection based on specific events, so as to ensure that sufficient teaching data can be obtained without causing data redundancy and waste of system resources.

[0073] The circuits and controls involved in the present invention are all prior arts and will not be elaborated here.

[0074] The above are only the embodiments of the present invention, and thus do not limit the patent scope of the present invention. Any equivalent structural or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A teaching evaluation system based on a sensing device, characterized in that: It includes a sensory data acquisition module, which is used to capture the sensory data of students’ facial expressions, classroom performance and interactive behaviors; The data processing module is responsible for preprocessing the original perception data; Feature extraction and analysis module, used to extract various behavioral features of teachers and students from perception data; A teaching evaluation module is used to evaluate the teacher's teaching process, the student's learning status and effect, and the teaching effect and atmosphere of the entire class based on the extracted various features and data. The teaching evaluation module includes a teacher teaching evaluation unit, a student learning evaluation unit, and a class overall evaluation unit. The teacher teaching evaluation unit is used to evaluate the teacher's teaching ability through a BP neural network. The teacher teaching evaluation unit organizes and quantifies the data on the teacher's teaching behavior in the collected perception data, converts it into input data suitable for BP neural network processing, and performs model training on the BP neural network based on similar teaching scenarios. The student learning evaluation unit is used to evaluate the student's learning status and effect, and the class overall evaluation unit is used to evaluate the teaching effect and atmosphere of the entire class; The result display and feedback module is responsible for presenting the analyzed and processed data to teachers, students and managers in the form of charts, graphs and reports; The system management and configuration module is responsible for managing system users and allowing system administrators to configure various system parameters.

2. The teaching evaluation system based on the sensing device according to claim 1 is characterized in that: When evaluating the teaching ability of a teacher through a BP neural network, the teacher teaching evaluation unit first constructs a BP neural network, selects a suitable theoretical model, establishes a mapping relationship between each indicator and data, collects corresponding original data, analyzes and preprocesses the data, and then trains the BP neural network, normalizes and groups the data according to the needs of neural network training, sets BP neural network parameters for training, testing and verification, and finally finds an evaluation model with the best evaluation effect by finding adjustments to the algorithm.

3. The teaching evaluation system based on the sensing device according to claim 2 is characterized in that: The student learning assessment unit is used to assess the student's learning status and effect, and provide personalized learning suggestions and feedback to each student through comprehensive analysis of the student's behavior, expression and learning results.

4. The teaching evaluation system based on the sensing device according to claim 3 is characterized in that: The overall classroom evaluation unit is used to evaluate the teaching effect and atmosphere of the entire class. The evaluation scope includes the quality of teacher-student interaction, classroom order, and learning atmosphere. By comprehensively analyzing various indicators, an overall evaluation and suggestions for the class are given.

5. The teaching evaluation system based on the sensing device according to claim 4 is characterized in that: The structure display and feedback module includes a visual display unit, a real-time feedback unit and a report production unit. The visual display unit is used to display the results of the teaching evaluation to teachers, students and education managers in the form of charts, graphs and reports. The real-time feedback unit is used to provide feedback information to teachers and students in real time. The report production unit is used to generate a detailed evaluation report based on the results of the teaching evaluation.

6. The teaching evaluation system based on the sensing device according to claim 5 is characterized in that: The visualization display unit is responsible for converting complex analytical data into charts, graphs, and reports to help teachers, students, and managers quickly obtain valuable information.

7. The teaching evaluation system based on the sensing device according to claim 6 is characterized in that: The real-time feedback unit is responsible for providing feedback information to teachers and students in real time during classroom teaching, so that teachers can adjust teaching strategies in a timely manner and students can understand their own learning situation and make adjustments in a timely manner.

8. The teaching evaluation system based on the sensing device according to claim 7 is characterized in that: The report production unit is responsible for generating a detailed evaluation report based on the results of the teaching evaluation. The evaluation report includes the advantages and disadvantages of classroom teaching, analysis of students' learning situation, and teaching suggestions. It provides a basis for teachers' teaching reflection and professional development, and also provides a reference for educational administrators' teaching management and decision-making.

9. The teaching evaluation system based on the sensing device according to claim 8, characterized in that: The perception data acquisition module includes a smart camera, a microphone array and an environmental sensor. The smart camera is installed in the front and back corners or on the ceiling of the classroom to clearly record the facial expressions and body movements of teachers and students. The microphone array is distributed in different locations in the classroom.

Images