Preschool education interaction system based on intelligent image recognition
Through intelligent image recognition technology combining ε-greedy search algorithm and deep learning algorithm in the preschool education system, the problems of slow recognition speed and low accuracy are solved, the generation and teaching of personalized interactive content are realized, and the targeted nature of personalized interactive content is improved. Learning interests and educational effects of preschool children.
Patent Information
- Application Number
- CN202410990588.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-23
- Publication Date
- 2024-11-08
AI Technical Summary
The existing preschool education system based on image recognition is slow in natural image recognition, and the recognition results are susceptible to external factors, resulting in the problem of low recognition accuracy.
Using an intelligent image recognition system combining ε-greedy search algorithm and deep learning algorithm, the image acquisition module, image recognition module, interactive module, teaching content module and feedback module can analyze the expressions and behaviors of preschool children in real time to generate personalized interactive content , and optimize teaching content through feedback information.
Improve recognition speed and accuracy, enhance interaction and stability, can better attract the attention of preschool children, improve learning interest and participation, provide personalized teaching, meet the unique needs of each child, and improve educational effectiveness .
Smart Images

Figure CN118918376A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of educational technology, and in particular is a preschool education interactive system based on intelligent image recognition. Background Art
[0002] In the field of preschool education, traditional education methods often lack pertinence and interest, and it is difficult to fully mobilize the learning enthusiasm and initiative of preschool children. With the continuous development of science and technology, intelligent image recognition technology has brought new opportunities for preschool education. Through intelligent image recognition technology, the expressions, movements and behaviors of preschool children can be analyzed in real time, so as to better understand their needs and interests, and provide a basis for providing personalized educational interactive content.
[0003] At present, existing preschool education products based on image recognition still have problems with recognition speed and stability in terms of recognition speed, stability, and interactivity: the existing system has a slow recognition speed for natural images, and the recognition results are easily affected by external factors, resulting in low recognition accuracy.
[0004] To this end, those skilled in the art have proposed a preschool education interactive system based on intelligent image recognition to solve the problems raised by the background technology. Summary of the invention
[0005] In order to solve the above technical problems, the present invention provides a preschool education interactive system based on intelligent image recognition to solve the problems in the prior art that the existing system has a slow recognition speed for natural images, and the recognition results are easily affected by factors such as lighting and angle, resulting in low recognition accuracy.
[0006] A preschool education interactive system based on intelligent image recognition, comprising:
[0007] An image acquisition module, used to collect image information of preschool children;
[0008] An image recognition module, connected to the image acquisition module, is used to intelligently recognize the acquired image information and analyze the expressions, movements and behaviors of preschool children;
[0009] An interactive module, connected to the image recognition module, generates corresponding interactive content according to the image recognition result, including games, stories, animations, etc.; the interactive module can generate interactive content in a personalized manner according to the interests and learning progress of the preschool children;
[0010] The teaching content module stores various teaching contents suitable for preschool children and is connected to the interactive module to provide corresponding teaching materials according to the interactive needs;
[0011] The feedback module is used to collect the feedback information of the preschool children on the interactive content and transmit it to the image recognition module and the interactive module to optimize the subsequent interactive content and teaching content.
[0012] Preferably, the image acquisition module uses an ε-greedy search algorithm to search, identify and acquire image information, and can accurately identify and acquire various expressions, movements and behaviors of preschool children. The ε-greedy search algorithm formula is as follows:
[0013]
[0014] Where x represents a random variable between [0,1]. When x < 1-ε, the exploit action or search action is executed. x∈[0,1] represents an adjustable parameter. At the beginning of model training, ε is set to 1. As the training step increases, ε gradually decreases to θ in a linear manner. min , as shown below:
[0015] ε=θ max +u(θ min -θ max );
[0016] in, represents the utilization factor, represents the current training step length, t η Represents the total step size of the decaying ε-greedy search algorithm, as the training step size As the value increases, u will gradually increase to 1 in a linear manner, θ max represents the maximum search factor, θ min Represents the minimum search factor.
[0017] Preferably, the image recognition module uses a deep learning algorithm for image recognition, which can accurately recognize a variety of expressions, movements and behaviors of preschool children. The operation steps of using the deep learning algorithm for image recognition are as follows:
[0018] S2.1. First, the image data transmitted from the image acquisition module to the image recognition module is preprocessed, including image cropping, scaling, and normalization processing, to improve the efficiency and accuracy of subsequent processing. The formula for the normalization processing is as follows:
[0019]
[0020] Among them, x i is the i-th output of the encoder; is the average value of the encoder output; z i is the i-th input of the decoder;
[0021] S2.2, then build a convolutional neural network, including convolutional layer, pooling layer, and fully connected layer; convolutional layer: used to extract local features in the image; each convolutional layer consists of multiple convolutional kernels, which slide on the input image and obtain feature maps through convolution operations; pooling layer: used to reduce the dimension of the feature map, reduce the amount of calculation, and retain important information. Common pooling operations include maximum pooling and average pooling; fully connected layer: at the back end of the convolutional neural network, it is used to integrate the previously extracted features for classification or regression operations;
[0022] S2.3, then use the preprocessed image data to train the model, divide the image data into a training set, a validation set and a test set, use the training set to train the model, the validation set is used to adjust the model parameters, and the test set is used to evaluate the model performance. At the same time, during the training process, the model parameters are optimized through the loss function, and the loss function formula is as follows:
[0023]
[0024] Where T is the number of samples in the training set. is the norm, i is the number of the estimated image information and the actual image information;
[0025] S2.4. Finally, use the test set to evaluate the recognition accuracy (A), precision (P), recall (R), and F1 score indicators of the model; and adjust the model structure or parameters according to the evaluation results to optimize the model;
[0026] The calculation formulas for the accuracy (A), precision (P), recall (R), and F1 score indicators are as follows:
[0027]
[0028] Among them, TP and FP represent true positive examples and false positive examples respectively; FN and TN represent false negative examples and true negative examples respectively.
[0029] When in use, the trained model is deployed to the preschool education interactive system; when the system obtains image information of preschool children through the image acquisition module, the deployed deep learning model is used for real-time recognition; based on the recognition results, the interactive module generates corresponding interactive content.
[0030] Preferably, the teaching content in the teaching content module is designed and classified according to the curriculum standards of preschool education and the laws of children's development.
[0031] Preferably, the feedback module can monitor the participation, attention and emotional state of preschool children in real time, and feed back the information to the image recognition module and the interaction module so as to adjust the interactive content and teaching content in time, and the feedback module includes a detection unit, which is used to evaluate the feedback information. The evaluation index uses the normalized mean square error (MMSE). The smaller the MMSE value, the higher the similarity between the predicted value and the true value, and the better the feedback and reconstruction performance of the algorithm. The normalized mean square error formula is as follows:
[0032]
[0033] Among them, f(j,k) is the true value, is the predicted value.
[0034] A processor is configured to execute the above-mentioned preschool education interactive system based on intelligent image recognition.
[0035] A computer-readable storage medium stores a computer program, and the computer program is executed by a processor to produce the above-mentioned preschool education interactive system based on intelligent image recognition.
[0036] Compared with the prior art, the present invention has the following beneficial effects:
[0037] The present invention analyzes the expressions and behaviors of preschool children through intelligent image recognition technology to generate personalized interactive content, which can better attract the attention of preschool children and improve their interest and participation in learning. At the same time, according to the interests and learning progress of preschool children, personalized teaching content and interactive activities are provided to meet the unique needs of each child, which helps to improve the educational effect. The system monitors the learning status and feedback information of preschool children in real time, and can adjust the teaching content and interactive methods in time to make teaching more targeted and improve the teaching quality. The system can assist teachers to better understand the learning situation of preschool children, reduce the workload of teachers, and improve teaching efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] Figure 1 This is a diagram of the interactive system for preschool education based on intelligent image recognition of the present invention. DETAILED DESCRIPTION
[0039] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.
[0040] like Figure 1 As shown:
[0041] Embodiment: The present invention provides a preschool education interactive system based on intelligent image recognition, including an image acquisition module, an image recognition module, an interactive module, a teaching content module and a feedback module, wherein the image acquisition module, the image recognition module, the interactive module, the teaching content module and the feedback module are electrically connected in sequence:
[0042] An image acquisition module, used to collect image information of preschool children;
[0043] An image recognition module is connected to the image acquisition module and is used to intelligently recognize the collected image information; by analyzing the expressions and behaviors of preschool children through intelligent image recognition technology, their learning status and interests can be more accurately understood, thereby generating more personalized and interesting interactive content. This interactive method can better attract the attention of preschool children, improve their learning interest and participation, and make the learning process more enjoyable and efficient;
[0044] An interactive module, connected to the image recognition module, generates corresponding interactive content according to the image recognition result; the interactive module can generate interactive content in a personalized manner according to the interests and learning progress of the preschool children;
[0045] The teaching content module stores various teaching contents suitable for preschool children, is connected to the interactive module, and provides corresponding teaching materials according to the interactive needs; the teaching contents in the teaching content module are designed and classified according to the curriculum standards of preschool education and the laws of children's development;
[0046] The feedback module is used to collect the feedback information of the preschool children on the interactive content and transmit it to the image recognition module and the interactive module to optimize the subsequent interactive content and teaching content; the feedback module can monitor the learning status and feedback information of the preschool children in real time and transmit the information to the image recognition module and the interactive module in time. Based on the feedback information, the system can quickly adjust the teaching content and interactive methods to make the teaching more targeted, avoid the one-size-fits-all teaching method, and improve the overall teaching quality;
[0047] As can be seen from the above, the system can flexibly adjust the interactive content and teaching materials according to the interests and learning progress of preschool children. This personalized education method can meet the unique needs of each child, ensure that they are fully developed in the areas that they are interested in and suitable for, thereby improving the educational effect; at the same time, the system can assist teachers to better understand the learning situation of preschool children, provide detailed data support and personalized teaching suggestions. This helps to reduce the workload of teachers, allowing them to focus more on teaching itself and improve teaching efficiency. Through rich and diverse interactive content and teaching materials, the system can cultivate preschool children's cognitive ability, language expression ability, social ability and creativity. This all-round education method helps to promote the all-round development of preschool children and lay a solid foundation for their future learning and life.
[0048] In one embodiment of the present invention, the image acquisition module uses an ε-greedy search algorithm to search, identify and acquire image information, and can accurately identify and acquire various expressions, movements and behaviors of preschool children. The ε-greedy search algorithm formula is as follows:
[0049]
[0050] Where x represents a random variable between [0,1]. When x < 1-ε, the exploit action or search action is executed. x∈[0,1] represents an adjustable parameter. At the beginning of model training, ε is set to 1. As the training step increases, ε gradually decreases to θ in a linear manner. min , as shown below:
[0051] ε=θ max +u(θ min -θ max );
[0052] in, represents the utilization factor, represents the current training step length, t η Represents the total step size of the decaying ε-greedy search algorithm, as the training step size As the value increases, u will gradually increase to 1 in a linear manner, θ max represents the maximum search factor, θ min Represents the minimum search factor.
[0053] From the above, it can be seen that by setting the ε-greedy search algorithm, the algorithm adjusts the search factor and the utilization factor through the formula, which can more accurately control the search process and ensure a higher accuracy rate when identifying various expressions, movements and behaviors of preschool children; it can significantly improve the performance of the image acquisition module and ensure that the system can quickly and accurately obtain children's expression and behavior information.
[0054] In one embodiment of the present invention, the image recognition module uses a deep learning algorithm to perform image recognition, which can accurately recognize a variety of expressions, movements and behaviors of preschool children. The operation steps of using the deep learning algorithm to perform image recognition are as follows:
[0055] S2.1. First, the image data transmitted from the image acquisition module to the image recognition module is preprocessed, including image cropping, scaling, and normalization processing, to improve the efficiency and accuracy of subsequent processing. The formula for the normalization processing is as follows:
[0056]
[0057] Among them, x i is the i-th output of the encoder; is the average value of the encoder output; z i is the i-th input of the decoder;
[0058] S2.2, then build a convolutional neural network, including convolutional layer, pooling layer, and fully connected layer;
[0059] S2.3, then use the preprocessed image data to train the model, divide the image data into a training set, a validation set and a test set, use the training set to train the model, the validation set is used to adjust the model parameters, and the test set is used to evaluate the model performance. At the same time, during the training process, the model parameters are optimized through the loss function, and the loss function formula is as follows:
[0060]
[0061] Where T is the number of samples in the training set. is the norm, i is the number of the estimated image information and the actual image information;
[0062] S2.4. Finally, use the test set to evaluate the recognition accuracy (A), precision (P), recall (R), and F1 score indicators of the model; and adjust the model structure or parameters according to the evaluation results to optimize the model;
[0063] The calculation formulas for the accuracy (A), precision (P), recall (R), and F1 score indicators are as follows:
[0064]
[0065] Among them, TP and FP represent true positive examples and false positive examples respectively; FN and TN represent false negative examples and true negative examples respectively.
[0066] From the above, we can see that by setting the normalization processing, the image data can be preprocessed, the dimensional differences between the data can be eliminated, and the efficiency and accuracy of subsequent processing can be improved. At the same time, the loss function is used to optimize the model parameters during the training process to ensure that the model can more accurately identify the information in the image, thereby significantly improving the accuracy of the image recognition module and providing more reliable data support for the subsequent interactive content generation. Furthermore, through the evaluation indicators, the performance of the model can be comprehensively evaluated, and the model structure and parameters can be adjusted according to the evaluation results, which helps to continuously optimize the performance of the image recognition model and improve the overall recognition ability and stability of the system.
[0067] In one embodiment of the present invention, the feedback module can monitor the participation, attention and emotional state of preschool children in real time, and feed back the information to the image recognition module and the interaction module so as to adjust the interactive content and teaching content in time, and the feedback module includes a detection unit, which is used to evaluate the feedback information. The evaluation index uses the normalized mean square error (MMSE). The smaller the MMSE value, the higher the similarity between the predicted value and the true value, and the better the feedback and reconstruction performance of the algorithm; the normalized mean square error formula is as follows:
[0068]
[0069] Among them, f(j,k) is the true value, is the predicted value.
[0070] From the above, we can see that by setting the normalized mean square error and evaluating the accuracy of feedback information through this formula, the participation, attention and emotional state of preschool children can be monitored in real time, and the information can be fed back to the image recognition module and the interactive module; it helps the system to adjust the interactive content and teaching content in time to meet the needs and interests of children, thereby improving user experience and learning effects.
[0071] The performance indicators of the existing method (i.e., the preschool education product based on image recognition on the market) and the preschool education interactive system based on intelligent image recognition provided by this embodiment are compared, and the comparison results are shown in Table 1:
[0072] Table 1- Performance index comparison results
[0073] Existing methods This embodiment Recognition speed 0.25s 0.06s Recognition accuracy 68.7% 93.8% Interaction Manual Operation No manual work required stability Poor Higher
[0074] From the above, it can be seen from Table 1 that compared with the existing methods, the preschool education interactive system based on intelligent image recognition provided by this embodiment has better effects in recognition speed, recognition accuracy, interactivity and stability. This method can provide more efficient and accurate personalized educational interactive content for preschool education.
[0075] The embodiment of the present application provides an electronic device, which is applicable to the above-mentioned preschool education interactive system based on intelligent image recognition, including:
[0076] Memory, used to protect computer programs and data;
[0077] Processor, used to run system programs.
[0078] The embodiment of the present application provides a computer storage medium, which is applicable to the above-mentioned preschool education interactive system based on intelligent image recognition, and performs hierarchical confidentiality management on the above-mentioned system and data in accordance with confidentiality management requirements.
[0079] Those skilled in the art will appreciate that the embodiments of the present application may be provided as a system or a computer program product. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.
[0080] The present application is described with reference to the flowcharts and / or block diagrams of the devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0081] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0082] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0083] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0084] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.
[0085] Computer readable media include permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.
[0086] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, commodity or device including the elements.
[0087] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.
Claims
1. A preschool education interactive system based on intelligent image recognition, characterized by: include: An image acquisition module, used to collect image information of preschool children; An image recognition module, connected to the image acquisition module, for intelligently recognizing the acquired image information; An interactive module, connected to the image recognition module, generates corresponding interactive content according to the image recognition result; the interactive module can generate interactive content in a personalized manner according to the interests and learning progress of the preschool children; The teaching content module stores various teaching contents suitable for preschool children and is connected to the interactive module to provide corresponding teaching materials according to the interactive needs; The feedback module is used to collect the feedback information of the preschool children on the interactive content and transmit it to the image recognition module and the interactive module to optimize the subsequent interactive content and teaching content.
2. A preschool education interactive system based on intelligent image recognition as claimed in claim 1, characterized in that: The image acquisition module uses the ε-greedy search algorithm to search, identify and acquire image information, and can accurately identify and acquire various expressions, movements and behaviors of preschool children. The formula of the ε-greedy search algorithm is as follows: Where x represents a random variable between [0,1]. When x < 1-ε, the exploit action or search action is executed. x∈[0,1] represents an adjustable parameter. At the beginning of model training, ε is set to 1. As the training step increases, ε gradually decreases to θ in a linear manner. min , as shown below: e=θ max +u(θ min -θ max ); in, represents the utilization factor, Indicates the current training step length, t η Represents the total step size of the decaying ε-greedy search algorithm, as the training step size As the value increases, u will gradually increase to 1 in a linear manner, θ max represents the maximum search factor, θ min Represents the minimum search factor.
3. The preschool education interactive system based on intelligent image recognition as claimed in claim 1, characterized in that: The image recognition module uses a deep learning algorithm to perform image recognition, and can accurately recognize various expressions, movements and behaviors of preschool children. The operation steps of using the deep learning algorithm to perform image recognition are as follows: S2.
1. First, the image data transmitted from the image acquisition module to the image recognition module is preprocessed, including image cropping, scaling, and normalization processing, to improve the efficiency and accuracy of subsequent processing. The formula for the normalization processing is as follows: Among them, x i is the i-th output of the encoder; is the average value of the encoder output; z i is the i-th input of the decoder; S2.2, then build a convolutional neural network, including convolutional layer, pooling layer, and fully connected layer; S2.3, then use the preprocessed image data to train the model, divide the image data into a training set, a validation set and a test set, use the training set to train the model, the validation set is used to adjust the model parameters, and the test set is used to evaluate the model performance. At the same time, during the training process, the model parameters are optimized through the loss function, and the loss function formula is as follows: Where T is the number of samples in the training set. is the norm, i is the number of the estimated image information and the actual image information; S2.
4. Finally, use the test set to evaluate the recognition accuracy (A), precision (P), recall (R), and F1 score indicators of the model; and adjust the model structure or parameters according to the evaluation results to optimize the model; The calculation formulas for the accuracy (A), precision (P), recall (R), and F1 score indicators are as follows: Among them, TP and FP represent true positive examples and false positive examples respectively; FN and TN represent false negative examples and true negative examples respectively.
4. The preschool education interactive system based on intelligent image recognition as claimed in claim 1, characterized in that: The teaching contents in the teaching content module are designed and classified according to the curriculum standards of preschool education and the laws of children's development.
5. The preschool education interactive system based on intelligent image recognition as claimed in claim 1, characterized in that: The feedback module can monitor the participation, attention and emotional state of preschool children in real time, and feed back the information to the image recognition module and the interactive module so as to adjust the interactive content and teaching content in time. The feedback module includes a detection unit, which is used to evaluate the feedback information. The evaluation index uses the normalized mean square error (MMSE). The smaller the MMSE value, the higher the similarity between the predicted value and the true value, and the better the feedback and reconstruction performance of the algorithm. The normalized mean square error formula is as follows: Among them, f(j,k) is the true value, is the predicted value.
6. A processor, characterized in that: The system is configured to execute the preschool education interactive system based on intelligent image recognition according to any one of claims 1 to 5.
7. A computer-readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to provide a preschool education interactive system based on intelligent image recognition as described in any one of claims 1 to 5.
Citation Information
Patent Citations
Target detection method for SAR (Synthetic Aperture Radar) image based on deep reinforcement learning
CN108764006A
Preschool education interaction system based on intelligent image recognition
CN114495605A
Intelligent classroom interaction system and method
CN117037552A
Advertisement creative design method and system based on artificial intelligence and big data analysis technology
CN118014652A
Multifunctional education robot
CN118079182A
