Rapid children vision screening system and method based on Internet of Things

Abstract

The invention belongs to the technical field of vision screening, and discloses a rapid children vision screening system and method based on the Internet of Things. The system comprises a screening terminal device, an Internet of Things communication device and a cloud server. The method comprises the following steps: collecting real-time basic information of a child, and uploading the real-time basic information to a cloud server; generating a real-time vision screening strategy according to the real-time basic information; according to a real-time vision screening strategy, selecting visual chart elements from a visual chart element library, generating a real-time visual chart page, and sending the real-time visual chart page to screening terminal equipment; visualizing the real-time visual chart page, collecting real-time interaction data of the child, and uploading the real-time interaction data to a cloud server; and according to the real-time vision screening strategy, analyzing the real-time interaction data to obtain a real-time vision screening analysis result, and sending the real-time vision screening analysis result to the screening terminal equipment. According to the invention, the problems of high hardware cost, low efficiency, poor effect, dependence on professionals and poor real-time performance in the prior art are solved.

Description

Technical Field

[0001] This invention belongs to the field of vision screening technology, specifically relating to a rapid vision screening system and method for children based on the Internet of Things. Background Technology

[0002] Children's vision health is a crucial issue related to their healthy growth and future development. Early detection and early intervention are effective ways to prevent and control common vision problems in children (such as myopia, hyperopia, astigmatism, and amblyopia). Traditional methods of children's vision screening usually rely on standard eye charts (such as the Snellen eye chart and the E chart) and on-site operation by medical personnel. These methods have the following shortcomings: 1) High hardware costs: Existing technologies for vision screening rely on sophisticated instruments, including automated refractometers, which require additional hardware costs. Furthermore, in some temporary vision screening activities, the use of sophisticated instruments obviously complicates the process. 2) Low efficiency: When dealing with groups of children, vision screening needs to be conducted on each child individually, and the screening speed may be affected by factors such as children's cooperation and environmental interference. Especially when there are many people to be screened, it takes a long time, resulting in low efficiency of existing technology in large-scale children's scenarios. 3) Poor results: Existing technologies often use fixed eye charts, which may lead to children answering questions based on memory rather than their actual vision. This results in the screening being unable to meet the objectivity requirements, and the fixed eye charts are not attractive to children, leading to poor screening results. 4) Reliance on professional personnel: The accuracy and consistency of current vision screening analysis technologies largely depend on the experience and skill level of the operators, resulting in high labor costs; 5) Poor real-time performance: Existing technologies require a lot of manual intervention for data collection and analysis, and lack an automated vision screening system, resulting in the inability to meet real-time requirements. Summary of the Invention

[0003] To address the problems of high hardware costs, low efficiency, poor results, reliance on professional personnel, and poor real-time performance in existing technologies, the present invention aims to provide a rapid vision screening system and method for children based on the Internet of Things.

[0004] The technical solution adopted in this invention is as follows: A rapid vision screening system for children based on the Internet of Things (IoT) includes a screening terminal device, an IoT communication device, and a cloud server. The screening terminal device communicates and connects with the cloud server through the IoT communication device. Screening terminal equipment is used to collect children's real-time basic information and real-time interactive data; and to visualize the real-time vision chart page sent by the cloud server. The Internet of Things (IoT) communication device is used to upload real-time basic information and real-time interactive data collected by the screening terminal device to the cloud data center; and to send the real-time vision chart page and real-time vision screening analysis results generated by the cloud server to the screening terminal device. The cloud server is used to generate vision screening strategies based on real-time basic information, resulting in corresponding real-time vision screening strategies; based on the real-time vision screening strategies, vision chart elements are selected from the vision chart element library to generate a real-time vision chart page; based on the real-time vision screening strategies, the real-time vision chart page and the corresponding real-time interactive data are analyzed to obtain real-time vision screening analysis results.

[0005] Furthermore, the screening terminal device includes a child information input unit, a visualization unit, and an interactive data acquisition unit, all of which are connected to an Internet of Things (IoT) communication device.

[0006] Furthermore, the interactive data acquisition unit includes a touch screen, a voice data acquisition device, and a video data acquisition device, all of which are connected to IoT communication devices.

[0007] Furthermore, the cloud server is equipped with a vision screening strategy generation unit, a vision chart page generation unit, a data preprocessing unit, and a vision screening analysis unit. All of these units are connected to IoT communication devices, with the vision screening strategy generation unit connected to the vision chart page generation unit and the data preprocessing unit connected to the vision screening analysis unit.

[0008] Furthermore, the vision screening strategy generation unit is equipped with a vision screening strategy generation model, and the vision screening analysis unit is equipped with a vision screening analysis model.

[0009] A rapid vision screening method for children based on the Internet of Things (IoT), using a rapid vision screening system for children, including a screening terminal device, an IoT communication device, and a cloud server, includes the following steps: Using the child information input unit of the screening terminal device, real-time basic information of the child is collected and uploaded to the cloud server through the Internet of Things communication device; Based on real-time basic information, a vision screening strategy generation model is used on a cloud server to generate a vision screening strategy, resulting in a real-time vision screening strategy. Based on the real-time vision screening strategy, vision chart elements are selected from the vision chart element library on the cloud server to generate a real-time vision chart page, which is then sent to the screening terminal device via an IoT communication device. The real-time vision chart page is visualized using the visualization unit of the screening terminal device, and the real-time interactive data of the child is collected using the interactive data acquisition unit of the screening terminal device and uploaded to the cloud server through the Internet of Things communication device. Based on the real-time vision screening strategy, the vision screening analysis model of the cloud server is used to analyze the real-time vision chart page and the corresponding real-time interactive data to obtain the real-time vision screening analysis results, which are then sent to the screening terminal device via IoT communication devices.

[0010] Furthermore, the vision screening strategy generation model is built based on a reinforcement learning algorithm, and the vision screening strategy generation model is equipped with an agent and an experience replay pool; The vision screening and analysis model is built based on deep learning algorithms and includes an interactive data processing module, a vision chart comparison module, and a vision screening and analysis module connected in sequence.

[0011] Furthermore, the real-time vision screening strategy includes real-time site layout decisions, real-time vision chart element decisions, real-time visualization decisions, and real-time data preprocessing decisions.

[0012] Furthermore, based on real-time basic information, a vision screening strategy is generated using the vision screening strategy generation model on the cloud server, resulting in a real-time vision screening strategy, including the following steps: Based on real-time basic information, several historical experiences are extracted from the experience replay pool in the vision screening strategy generation model, and the action parameters of the agent's action space in the vision screening strategy generation model are updated based on these historical experiences to obtain the updated action space. Based on real-time basic information, update the state parameters of the state space of the agent in the vision screening strategy generation model to obtain the updated state space; Based on the updated action space and updated state space, an intelligent agent is used to generate a vision screening strategy, resulting in a real-time vision screening strategy.

[0013] Furthermore, based on the real-time vision screening strategy, the vision screening analysis model of the cloud server is used to analyze the real-time vision chart page and the corresponding real-time interactive data to obtain the real-time vision screening analysis results, which are then sent to the screening terminal device via IoT communication devices, including the following steps: Based on the real-time data preprocessing decision of the real-time vision screening strategy, the real-time interactive data is preprocessed to obtain preprocessed real-time interactive data. The interactive data processing module in the vision screening analysis model of the cloud server is used to process the preprocessed real-time interactive data to obtain real-time interactive information. The vision chart comparison module in the vision screening and analysis model of the cloud server is used to compare the real-time vision chart page with the corresponding real-time interactive information to obtain the real-time comparison results. The vision screening analysis module in the vision screening analysis model of the cloud server is used to analyze real-time basic information and real-time comparison results to obtain real-time vision screening analysis results. The real-time vision screening analysis results are sent to the screening terminal device via IoT communication devices, and the visualization unit is used to visualize the real-time vision screening analysis results.

[0014] The beneficial effects of this invention are as follows: This invention provides a rapid vision screening system and method for children based on the Internet of Things (IoT). Vision screening can be achieved through the cooperation of modular screening terminal devices, IoT communication devices, and cloud servers, avoiding reliance on precision instruments, reducing hardware costs, and facilitating deployment in various scenarios (including resource-limited areas). Through automated processes, intelligent strategy generation, and rapid interaction, the screening time for a single child is significantly shortened. The system can operate continuously, significantly improving the efficiency of large-scale screening. Utilizing dynamically generated vision chart pages and multiple interaction methods (touch, voice, video), interference from children's memory is effectively avoided, improving the objectivity and accuracy of the screening. Diverse interactive and visual stimuli better attract children's attention, increasing participation and screening effectiveness, and greatly improving the cooperation of young children or those who are not good at verbal expression. The system is easy to operate, with most analysis completed by cloud-based intelligent models, resulting in objective and consistent results. This reduces the need for specialized knowledge and experience among operators, thus lowering labor costs. The cloud server handles complex algorithm processing and result evaluation, while the screening terminal only needs to display the data and collect basic information. This allows non-professionals (such as school teachers, community workers, and even parents) to easily perform screening operations, significantly reducing labor costs and professional barriers. This "de-professionalization" of screening makes it easier to popularize in grassroots settings, schools, and communities. The IoT-based architecture integrates the generation of the vision chart, data collection, transmission, analysis, and result feedback into a closed-loop system, ensuring a smooth screening process. Combined with reinforcement learning and deep learning models, the system can dynamically adjust screening strategies based on the child's real-time status, achieving an intelligent and personalized screening experience.

[0015] Other beneficial effects of the present invention will be further explained in the specific embodiments. Attached Figure Description

[0016] Figure 1 This is a structural block diagram of the Internet of Things-based rapid vision screening system for children in this invention.

[0017] Figure 2This is a flowchart of the rapid vision screening method for children based on the Internet of Things in this invention. Detailed Implementation

[0018] The present invention will be further explained below with reference to the accompanying drawings and specific embodiments.

[0019] Example 1: like Figure 1 As shown, this embodiment provides a rapid vision screening system for children based on the Internet of Things (IoT), including a screening terminal device, an IoT communication device, and a cloud server. The screening terminal device communicates with the cloud server through the IoT communication device. Screening terminal equipment is used to collect children's real-time basic information and real-time interactive data; and to visualize the real-time vision chart page sent by the cloud server. The Internet of Things (IoT) communication device is used to upload real-time basic information and real-time interactive data collected by the screening terminal device to the cloud data center; and to send the real-time vision chart page and real-time vision screening analysis results generated by the cloud server to the screening terminal device. The cloud server is used to generate vision screening strategies based on real-time basic information, resulting in corresponding real-time vision screening strategies; based on the real-time vision screening strategies, vision chart elements are selected from the vision chart element library to generate a real-time vision chart page; based on the real-time vision screening strategies, the real-time vision chart page and the corresponding real-time interactive data are analyzed to obtain real-time vision screening analysis results.

[0020] As a preferred embodiment, the screening terminal device includes a child information input unit, a visualization unit, and an interactive data acquisition unit, all of which are connected to an Internet of Things (IoT) communication device.

[0021] The child information input unit is used to collect the child's real-time basic information and send the child's real-time basic information to the Internet of Things communication device within the communication range; A visualization unit is used to receive the real-time vision chart page and the real-time vision screening analysis results within the communication range, and to visualize the real-time vision chart page and the real-time vision screening analysis results; in this embodiment, the visualization unit is a display screen; The interactive data acquisition unit is used to collect children's real-time interactive data and send the children's real-time interactive data to IoT communication devices within the communication range.

[0022] Preferably, the interactive data acquisition unit includes a touch screen, a voice data acquisition device, and a video data acquisition device, all of which are connected to IoT communication devices. A touchscreen is used to collect real-time touch data of children touching various elements on the real-time vision chart page. A voice data collector used to collect real-time voice data of children's responses; Video data acquisition devices, including cameras, infrared sensors, eye-tracking devices, etc., are used to collect real-time video data in the form of children's body movements or facial expression changes.

[0023] Preferably, the cloud server is equipped with a vision screening strategy generation unit, a vision chart page generation unit, a data preprocessing unit, and a vision screening analysis unit. All of these units are connected to IoT communication devices, with the vision screening strategy generation unit connected to the vision chart page generation unit and the data preprocessing unit connected to the vision screening analysis unit.

[0024] The vision screening strategy generation unit is used to generate a vision screening strategy based on real-time basic information and the vision screening strategy generation model of the cloud server, so as to obtain a real-time vision screening strategy. The vision chart page generation unit is used to select vision chart elements from the vision chart element library on the cloud server according to the real-time vision screening strategy and generate a real-time vision chart page. The data preprocessing unit is used to preprocess the real-time interactive data according to the real-time vision screening strategy to obtain preprocessed real-time interactive data. The vision screening and analysis unit is used to analyze the real-time vision chart page and the corresponding real-time interactive data using the vision screening and analysis model of the cloud server, based on the real-time vision screening strategy, to obtain the real-time vision screening and analysis results.

[0025] Preferably, the vision screening strategy generation unit is equipped with a vision screening strategy generation model, and the vision screening analysis unit is equipped with a vision screening analysis model.

[0026] Example 2: like Figure 2 As shown, this embodiment provides a rapid vision screening method for children based on the Internet of Things (IoT). Based on a rapid vision screening system for children, the system includes a screening terminal device, an IoT communication device, and a cloud server, and includes the following steps: S1: Using the child information input unit of the screening terminal device, collect the child's real-time basic information and upload it to the cloud server through the Internet of Things communication device; Real-time basic information includes the child's name, age, gender, class, and historical vision screening analysis results; S2: Based on real-time basic information, use the vision screening strategy generation model of the cloud server to generate a vision screening strategy and obtain a real-time vision screening strategy; The vision screening strategy generation model is built based on a reinforcement learning algorithm, and the vision screening strategy generation model is equipped with an agent and an experience replay pool; Reinforcement learning algorithms include Q-Learning, Deep Q-Network (DQN), and Deep Deterministic Policy Gradient (DDPG); in this embodiment, the vision screening strategy generation model is constructed based on the DQN algorithm. Real-time vision screening strategies include real-time site layout decisions, real-time vision chart element decisions, real-time visualization decisions, and real-time data preprocessing decisions (including child interactive data analysis and comparison algorithms, vision assessment algorithms, etc.). Based on real-time basic information, a vision screening strategy generation model is used on a cloud server to generate a real-time vision screening strategy, including the following steps: S2-1: Based on real-time basic information, extract several historical experiences from the experience replay pool in the vision screening strategy generation model, and update the action parameters of the agent's action space in the vision screening strategy generation model based on these historical experiences to obtain the updated action space. In this embodiment, the updated action space includes: Site layout decision space: {Adjust lighting brightness, adjust the distance between visualization units, adjust the angle of visualization units, maintain the current layout...}; Vision chart element decision space: {Display the letter E of a specific size / shape in the next row, display the letter C in a random direction, display numbers in a specific pattern, etc.}; Visualizing the decision space: {Using a cartoon eye chart, using a standard eye chart, adjusting font color / background color...}; Data preprocessing decision space: {Applying filters to remove noise, performing motion compensation, directly using raw data...}; S2-2: Based on real-time basic information, current ambient light level, current child attention status score, and current screening progress, update the state parameters of the agent's state space in the vision screening strategy generation model to obtain the updated state space; S2-3: Based on the updated action space and the updated state space, use an intelligent agent to generate a vision screening strategy and obtain a real-time vision screening strategy; S3: Based on the real-time vision screening strategy, select vision chart elements from the vision chart element library on the cloud server, generate a real-time vision chart page, and send it to the screening terminal device through the Internet of Things communication device; The visual acuity chart elements include optotypes of different sizes and orientations, such as E-shaped, C-shaped, numbers, and graphics. Optotypes (visual acuity chart elements) are randomly selected from the visual acuity chart element library and combined to generate real-time visual acuity chart pages with different optotype sizes and different arrangement orders. The size, number of optotypes, and arrangement of the real-time visual acuity chart page can be randomly changed within preset rules to increase the fun of the screening and the ability to resist memory interference. S4: Use the visualization unit of the screening terminal device to visualize the real-time vision chart page, use the interactive data acquisition unit of the screening terminal device to collect the child's real-time interactive data, and upload it to the cloud server through the Internet of Things communication device; Real-time interactive data includes real-time touch data of children interacting with various elements of the real-time vision chart on the real-time vision chart page, real-time voice data of children's responses (such as reading out optotypes, describing directions, etc.), and real-time video data in the form of body movements (such as pointing, nodding, shaking head, etc.) or facial expression changes. S5: Based on the real-time vision screening strategy, the vision screening analysis model of the cloud server is used to analyze the real-time vision chart page and the corresponding real-time interactive data to obtain the real-time vision screening analysis results, which are then sent to the screening terminal device via IoT communication devices. The vision screening and analysis model is built based on deep learning algorithms and includes an interactive data processing module, a vision chart comparison module, and a vision screening and analysis module connected in sequence. Based on the real-time vision screening strategy, a vision screening analysis model on a cloud server is used to analyze the real-time vision chart page and corresponding real-time interactive data to obtain real-time vision screening analysis results, which are then sent to the screening terminal device via an IoT communication device. The process includes the following steps: S5-1: Based on the real-time data preprocessing decision of the real-time vision screening strategy (including applying filters to remove noise), preprocess the real-time interactive data to obtain preprocessed real-time interactive data, including the following steps: S5-1-1: Set precise timestamps for real-time data points (real-time touch events of real-time touch data, real-time voice frames of real-time voice data, and video frames of real-time video data) in real-time interactive data of different modalities, and use the timestamps to align real-time data points of different modalities on the time axis to ensure that multimodal information at the same moment can be associated during subsequent processing to obtain aligned real-time interactive data. S5-1-2: Based on the application filter to remove noise in the real-time data preprocessing decision according to the real-time vision screening strategy, the aligned real-time touch data is sequentially de-jittered (using mean filtering or Gaussian filtering to remove small jitters in the touch coordinates to obtain a smoother trajectory), outlier removal (identifying and removing touch points that deviate significantly from the normal range, such as suddenly jumping to the other side of the screen), and signal smoothing (low-pass filtering of touch pressure or duration signals to reduce high-frequency noise), to obtain the preprocessed real-time touch data; S5-1-3: Perform environmental noise suppression (using spectral subtraction, Wiener filtering, or a deep learning-based noise reduction model to remove background noise (such as air conditioner noise, other children's sounds)), echo cancellation (if a speaker is present, apply echo cancellation technology), and speech activity detection (identify and segment the actual speech segments, and remove silent segments and non-speech parts such as breathing sounds) on the aligned real-time speech data to obtain preprocessed real-time speech data; S5-1-4: Perform the following steps on the aligned real-time video data in sequence: denoise (apply spatial domain (such as Gaussian filtering) or frequency domain (such as wavelet denoising) filters to video frames to reduce image noise), motion blur reduction (if there is slight shaking of the camera or the child's head causing blurring, try using motion deblurring algorithms (possibly based on optical flow or deep learning)), and background subtraction (the background is relatively fixed, so the background can be subtracted to retain only the foreground (child) information and reduce background interference) to obtain the preprocessed real-time video data; S5-1-5: Integrate preprocessed real-time touch data, preprocessed real-time voice data, and preprocessed real-time video data to obtain preprocessed real-time interactive data; S5-2: Using the interactive data processing module in the vision screening analysis model of the cloud server, the preprocessed real-time interactive data is processed to obtain real-time interactive information, including the following steps: S5-2-1: Using the interactive data processing module in the vision screening analysis model of the cloud server, the preprocessed real-time interactive data is processed sequentially to perform target recognition (determine the specific visual chart element corresponding to the touch point (such as which direction the E-shape is, which size the optotype is)), reaction time calculation (calculate the time difference from the display of the optotype to the occurrence of the touch), and touch pattern analysis (analyze the accuracy of the touch (whether the center of the target is accurately clicked) and continuity (whether it is a single click or multiple adjustments)) to obtain the real-time touch data processing results. S5-2-2: The preprocessed real-time interactive data is processed sequentially as follows: speech recognition (using the Wav2Vec 2.0 algorithm to convert speech segments into text), content parsing (using the Transformers algorithm to identify which visual target or direction the child is referring to, or whether it indicates misunderstanding or a request for repetition), reaction time calculation (calculating the time difference from the display of the visual target to the end of the speech response), and speech feature extraction (using the Recurrent Neural Network (RNN) algorithm to extract features such as volume and speech rate that may reflect the child's attention or difficulty level), to obtain the real-time speech data processing results; S5-2-3: Perform sequential pose / action analysis (using a convolutional neural network (CNN) to identify key actions such as nodding (indicating confirmation), shaking the head (indicating negation), and pointing (pointing to the target direction or location) and calculate their occurrence time) and reaction time calculation (calculating the time difference from the display of the target to the completion of a specific action (such as pointing or nodding)) to obtain the real-time video data processing results; S5-2-4: Integrate real-time touch data processing results, real-time voice data processing results, and real-time video data processing results to obtain real-time interactive information; For example, real-time interactive information = { "Target ID": "E_32_Right", "Response": "Right", "Reaction Time": 1.2 seconds, "Modality": ["Touch", "Voice"], "Touch Accuracy": 0.95, "Action": "Point Right"}; S5-3: Using the vision chart comparison module in the vision screening analysis model of the cloud server, compare the real-time vision chart page with the corresponding real-time interactive information to obtain the real-time comparison results, including the following steps: S5-3-1: Input the real-time vision chart page information and real-time interactive information into the vision chart comparison module of the vision screening analysis model on the cloud server; S5-3-2: Extract the currently displayed optotype content (such as direction "right", "up", "left", "down" or specific letters / numbers) and optotype size (such as the letter E in 0.6 rows) from the real-time vision chart page information, and extract the child's real-time response content, the interaction modality used (touch, voice, action), reaction time, etc. from the real-time interactive information; S5-3-3: Compare the child's real-time response with the standard answer of the current visual target. For example, if the visual target is "E_Right" and the child answers "Right" (sound or gesture pointing to the right), it is judged as correct. If the child touches the area in the corresponding direction, it is also judged as correct. If the answer does not match, it is judged as wrong, and the first real-time comparison element is obtained. S5-3-4: Record which modality(s) the child mainly used to respond, record the reaction time at the comparison moment, obtain the second real-time comparison element, and combine it with the first real-time comparison element to obtain the real-time comparison result; For example, the real-time comparison result = { "Target ID": "E_32_Right", "Current Target Size": 0.6, "Response": "Right", "Correct or Incorrect": True, "Reaction Time": 1.2 seconds, "Main Interaction Modality": "Voice", "Confidence": 0.9}; S5-4: Use the vision screening analysis module in the vision screening analysis model of the cloud server to analyze the real-time basic information and real-time comparison results to obtain the real-time vision screening analysis results; Based on the input real-time basic information and real-time comparison results, a multilayer perceptron (MLP) is used for analysis to obtain real-time vision screening analysis results; For example, the real-time vision screening analysis result = { "Vision level": "Normal", "Estimated LogMAR value": 0.0,"Recommendation": "End screening"} or { "Vision level": "Suspicious", "Estimated LogMAR value": 0.2, "Recommendation": "Recommend further examination by an ophthalmologist"}; S5-5: The real-time vision screening analysis results are sent to the screening terminal device via IoT communication devices, and the visualization unit is used to visualize the real-time vision screening analysis results.

[0027] This invention provides a rapid vision screening system and method for children based on the Internet of Things (IoT). Vision screening can be achieved through the cooperation of modular screening terminal devices, IoT communication devices, and cloud servers, avoiding reliance on precision instruments, reducing hardware costs, and facilitating deployment in various scenarios (including resource-limited areas). Through automated processes, intelligent strategy generation, and rapid interaction, the screening time for a single child is significantly shortened. The system can operate continuously, significantly improving the efficiency of large-scale screening. Utilizing dynamically generated vision chart pages and multiple interaction methods (touch, voice, video), interference from children's memory is effectively avoided, improving the objectivity and accuracy of the screening. Diverse interactive and visual stimuli better attract children's attention, increasing participation and screening effectiveness, and greatly improving the cooperation of young children or those who are not good at verbal expression. The system is easy to operate, with most analysis completed by cloud-based intelligent models, resulting in objective and consistent results. This reduces the need for specialized knowledge and experience among operators, thus lowering labor costs. The cloud server handles complex algorithm processing and result evaluation, while the screening terminal only needs to display the data and collect basic information. This allows non-professionals (such as school teachers, community workers, and even parents) to easily perform screening operations, significantly reducing labor costs and professional barriers. This "de-professionalization" of screening makes it easier to popularize in grassroots settings, schools, and communities. The IoT-based architecture integrates the generation of the vision chart, data collection, transmission, analysis, and result feedback into a closed-loop system, ensuring a smooth screening process. Combined with reinforcement learning and deep learning models, the system can dynamically adjust screening strategies based on the child's real-time status, achieving an intelligent and personalized screening experience.

[0028] This invention is not limited to the optional embodiments described above, and anyone can derive other various forms of products based on the inspiration of this invention. The specific embodiments described above should not be construed as limiting the scope of protection of this invention; the scope of protection of this invention should be determined by the claims, and the specification can be used to interpret the claims.

Claims

1. An Internet of Things based rapid child vision screening system characterized in that: It includes screening terminal equipment, Internet of Things (IoT) communication equipment, and a cloud server, wherein the screening terminal equipment communicates with the cloud server through the IoT communication equipment. Screening terminal equipment is used to collect children's real-time basic information and real-time interactive data; and to visualize the real-time vision chart page sent by the cloud server. Internet of Things (IoT) communication devices are used to upload real-time basic information and real-time interactive data collected by screening terminal devices to cloud data centers. Send the real-time vision chart page and real-time vision screening analysis results generated by the cloud server to the screening terminal device; The cloud server is used to generate vision screening strategies based on real-time basic information, and obtain the corresponding real-time vision screening strategies. Based on the real-time vision screening strategy, vision chart elements are selected from the vision chart element library to generate a real-time vision chart page. Based on the real-time vision screening strategy, the real-time vision chart page and the corresponding real-time interactive data are analyzed to obtain the real-time vision screening analysis results.

2. The IoT based quick vision screening system for children as claimed in claim 1 wherein: The screening terminal device includes a child information input unit, a visualization unit, and an interactive data acquisition unit, all of which are connected to an Internet of Things (IoT) communication device.

3. The IoT based quick vision screening system for children as claimed in claim 2, wherein: The interactive data acquisition unit includes a touch screen, a voice data acquisition device, and a video data acquisition device, all of which are connected to IoT communication devices.

4. The fast children's vision screening system based on the Internet of Things according to claim 3, characterized in that: The cloud server is equipped with a vision screening strategy generation unit, a vision chart page generation unit, a data preprocessing unit, and a vision screening analysis unit. The vision screening strategy generation unit, the vision chart page generation unit, the data preprocessing unit, and the vision screening analysis unit are all connected to IoT communication devices. The vision screening strategy generation unit is connected to the vision chart page generation unit, and the data preprocessing unit is connected to the vision screening analysis unit.

5. The IoT based quick vision screening system for children as claimed in claim 4, wherein: The vision screening strategy generation unit is equipped with a vision screening strategy generation model, and the vision screening analysis unit is equipped with a vision screening analysis model.

6. A fast children's vision screening method based on the Internet of Things, based on the fast children's vision screening system of claim 5, the system comprising a screening terminal device, an Internet of Things communication device and a cloud server, characterized in that: Includes the following steps: Using the child information input unit of the screening terminal device, real-time basic information of the child is collected and uploaded to the cloud server through the Internet of Things communication device; Based on real-time basic information, a vision screening strategy generation model is used on a cloud server to generate a vision screening strategy, resulting in a real-time vision screening strategy. Based on the real-time vision screening strategy, vision chart elements are selected from the vision chart element library on the cloud server to generate a real-time vision chart page, which is then sent to the screening terminal device via an IoT communication device. The real-time vision chart page is visualized using the visualization unit of the screening terminal device, and the real-time interactive data of the child is collected using the interactive data acquisition unit of the screening terminal device and uploaded to the cloud server through the Internet of Things communication device. Based on the real-time vision screening strategy, the vision screening analysis model of the cloud server is used to analyze the real-time vision chart page and the corresponding real-time interactive data to obtain the real-time vision screening analysis results, which are then sent to the screening terminal device via IoT communication devices.

7. The method as claimed in claim 6, wherein the method is based on Internet of Things (IoT) for quick vision screening of children. The vision screening strategy generation model is constructed based on a reinforcement learning algorithm, and the vision screening strategy generation model is provided with an agent and an experience replay pool; The vision screening analysis model is constructed based on a deep learning algorithm, and the vision screening analysis model comprises an interactive data processing module, a vision chart comparison module and a vision screening analysis module connected in sequence.

8. The method of claim 7, wherein the method is based on Internet of Things. The real-time vision screening strategy includes real-time site arrangement decision, real-time vision chart element decision, real-time visualization decision and real-time data preprocessing decision.

9. The method of claim 8, wherein the method is based on Internet of Things. According to the real-time basic information, the vision screening strategy generation model of the cloud server is used to generate the vision screening strategy, and the real-time vision screening strategy is obtained, including the following steps: According to the real-time basic information, a plurality of historical experiences are extracted from the experience replay pool in the vision screening strategy generation model, and the action parameters of the action space of the agent in the vision screening strategy generation model are updated according to the plurality of historical experiences, to obtain an updated action space; According to the real-time basic information, the state parameters of the state space of the agent in the vision screening strategy generation model are updated, to obtain an updated state space; Based on the updated action space and the updated state space, the agent is used to generate the vision screening strategy, and the real-time vision screening strategy is obtained.

10. The method of claim 9, wherein: According to the real-time vision screening strategy, the real-time vision chart page and the corresponding real-time interactive data are analyzed by using the vision screening analysis model of the cloud server, to obtain a real-time vision screening analysis result, which is sent to the screening terminal device through the Internet of Things communication equipment, including the following steps: According to the real-time data preprocessing decision of the real-time vision screening strategy, the real-time interactive data is preprocessed to obtain preprocessed real-time interactive data; The interactive data processing module in the vision screening analysis model of the cloud server is used to process the preprocessed real-time interactive data to obtain real-time interactive information; The vision chart comparison module in the vision screening analysis model of the cloud server is used to compare the real-time vision chart page and the corresponding real-time interactive information to obtain a real-time comparison result; The vision screening analysis module in the vision screening analysis model of the cloud server is used to analyze the real-time basic information and the real-time comparison result to obtain a real-time vision screening analysis result; The real-time vision screening analysis result is sent to the screening terminal device through the Internet of Things communication equipment, and the real-time vision screening analysis result is visualized by using a visualization unit.

Images