Touch sensing layer optimization system for ultra-thin dustproof liquid crystal display

Abstract

The application discloses a touch sensing layer optimization system of an ultrathin dustproof liquid crystal display screen and belongs to the technical field of display touch control. The application solves the problem that the prior art cannot effectively utilize and process abnormal data, leading to temporary failure or degradation of touch optimization function, and through correlation analysis on historical data, dust distribution threshold and relative humidity threshold are preset and mapped into an original data table; real-time data exceeding the threshold are marked as low-confidence data, and corresponding original high-confidence data thereof is extracted from the original data table as substitute data to form a comprehensive environmental state index; through the substitute mechanism, the continuity and reliability of the data are ensured, the robustness, fault tolerance and reliability of the touch optimization system in the face of real problems such as temporary failure of the sensor and abnormal reading are improved, the continuity of touch performance diagnosis and optimization strategy is ensured, and consistent and smooth touch experience is provided for users.

Description

Technical Field

[0001] This invention relates to the field of display touch technology, specifically to an optimized touch sensing layer system for ultra-thin dustproof liquid crystal displays. Background Technology

[0002] LCD screens are being used more and more widely in various electronic devices, and ultra-thin dustproof LCD screens are favored by the market due to their thinness, lightness, and dustproof characteristics.

[0003] However, existing ultra-thin dustproof LCD displays still have some problems with touch sensing; for example, existing optimization systems rely too much on real-time data collected by sensors. When key data is missing or obviously erroneous, the system often has no choice but to discard the data, resulting in incomplete decision information or being forced to adopt a default safety strategy that may lead to suboptimal results, causing the touch optimization function to temporarily fail or degrade.

[0004] Therefore, to address the shortcomings of existing systems, an optimized touch-sensing layer system for an ultra-thin, dustproof LCD display is proposed. Summary of the Invention

[0005] The purpose of this invention is to provide an optimization system for the touch sensing layer of an ultra-thin dustproof LCD display. By performing correlation analysis on historical data, a dust distribution threshold and a relative humidity threshold are preset and mapped to a raw data table. Real-time data exceeding the thresholds are marked as low-confidence data, and the corresponding raw high-confidence data is extracted from the raw data table as replacement data to form a comprehensive environmental status index. This replacement mechanism ensures the continuity and reliability of the data, improving the robustness, fault tolerance, and reliability of the touch optimization system when facing real-world problems such as temporary sensor failures and abnormal readings. It also ensures the continuity of touch performance diagnosis and optimization strategies, providing users with a consistent and smooth touch experience and solving the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: The touch sensing layer optimization system for ultra-thin dustproof LCD displays includes: The touch sensing layer is configured to be composed of an ultra-thin transparent nano-silver wire mesh and integrated into the inner side of the polarizer of the liquid crystal display or the inner surface of the encapsulation glass. The environmental sensing module is configured to detect dust distribution data on the display screen surface and ambient temperature and humidity data in real time through micro-particle sensors, temperature sensors, and humidity sensors; and to fuse the dust distribution data, temperature data, and humidity data to form a comprehensive environmental status index. The signal processing module is configured to calculate the signal-to-noise ratio and baseline capacitance drift of the original signal of the touch sensing layer, and use the calculation results as a comprehensive touch index; input the comprehensive environmental state index and the comprehensive touch index into a lightweight neural network model to diagnose the main causes of the current touch performance degradation and predict the trend of touch performance degradation; based on the prediction results, dynamically generate touch parameter adjustment instructions and dustproof drive instructions. The optimized control module is configured to select the optimal historical collaborative optimization strategy from the strategy library based on environmental monitoring results and generate specific control commands to dynamically adjust the touch sensing layer and signal processing module.

[0007] Furthermore, the environment perception module includes: The confidence analysis module is configured to acquire dust distribution data, temperature data, and humidity data from historical data when the display screen is in normal operating condition. A correlation analysis was performed between historical dust distribution data and historical temperature and humidity data. Based on the analysis results, dust distribution thresholds and relative humidity thresholds were preset to measure the confidence level of dust distribution data in relation to temperature and humidity data. Acquire real-time dust distribution data, temperature data, and humidity data, and compare them with dust distribution thresholds and relative humidity thresholds. If the dust distribution data exceeds the dust distribution thresholds, mark this set of dust distribution data as low-confidence data. A correlation analysis was performed between temperature and humidity data. If the humidity data exceeded the relative humidity threshold, the set of temperature and humidity data was marked as low-confidence data.

[0008] Furthermore, after the confidence analysis module identifies low-confidence data, it includes: Preprocessing is performed on historical data of dust distribution, temperature, and humidity when the display screen is in normal operation. Preprocessing includes: data cleaning, normalization, and time alignment. The processed dust distribution data, temperature data, and humidity data are matched according to time windows and mapped to form the original data table; When the analysis results show that there is low-confidence data, extract the data currently marked as low-confidence. The original high-confidence data corresponding to the low-confidence data is mapped from the original data table and used as the substitute data for the current low-confidence data, and marked as the substitute data. The substitute data and normal data are combined to form a comprehensive environmental status index, with the substitute data given a lower weight and the normal data given a higher weight.

[0009] Furthermore, the signal processing module includes: The model building module is configured to collect comprehensive touch indicators, comprehensive environmental state indicators, and corresponding causes and trends of touch performance degradation from historical data. It uses historical data to train a lightweight neural network model, extracts spatial features from the comprehensive touch indicators, learns the spatial correlation of signal attenuation in different regions, and analyzes the time series relationship between the comprehensive environmental state indicators and the comprehensive touch indicators to capture their dynamic evolution law and obtain the relationship between the comprehensive touch indicators, the comprehensive environmental state indicators, and the causes of touch performance degradation. The real-time diagnostic module is configured to input the comprehensive touch performance index and comprehensive environmental status index calculated in real time into a trained lightweight neural network model to diagnose the main reasons for the current decline in touch performance in real time; and predict the decline trend of touch performance, including possible changes in touch performance in the future.

[0010] Furthermore, the signal processing module further includes: The multi-dimensional analysis module is configured to generate verifiable physical correlation assumptions based on diagnostic and prediction results. Based on the generated hypotheses, the touch signal is re-analyzed, the statistical distribution of noise is calculated, and the impulse noise is verified; the harmonic distortion is calculated, and the signal leakage is verified; and the current dust distribution data, temperature data, and humidity data are examined to see if they meet the assumed environmental conditions. Based on the analysis results, verify the physical correlation assumptions; if the verification passes, execute the generated original optimization instructions; if the verification fails, initiate the general degradation strategy.

[0011] Furthermore, the signal processing module further includes: The data processing module is configured to acquire raw touch signals from the touch sensing layer, estimate the noise level in the signal through signal processing technology, calculate the ratio of signal power to noise power, and obtain the signal-to-noise ratio. When there is no touch operation, the baseline capacitance value of the touch sensing layer is measured; the difference between the current measured value and the initial baseline capacitance value is calculated to obtain the baseline capacitance drift; the signal-to-noise ratio and the baseline capacitance drift are normalized to bring them into the same dimension; according to the preset weights, the signal-to-noise ratio and the baseline capacitance drift are weighted and averaged to generate a comprehensive touch index; the comprehensive environmental state index and the comprehensive touch index are time-synchronized and aligned to form the input samples of the lightweight neural network model. The strategy formulation module is configured to formulate corresponding touch parameter adjustment strategies and dust prevention measures based on the diagnostic results of historical data, forming a strategy library.

[0012] Furthermore, the preset weights corresponding to the signal-to-noise ratio and baseline capacitance drift are set in the following manner: The touch frequency is collected in real time per unit time, wherein the value of the unit time ranges from 5 min to 8 min; The current touch activity coefficient Kt is obtained by using the touch frequency acquired per unit of time. Retrieve the noise spectrum corresponding to each original touch signal collected; The proportion of each frequency component in the noise spectrum is obtained based on the noise spectrum corresponding to the original touch signal. The standard deviation of the proportion of each frequency component is obtained based on the proportion value of each original touch signal collected each time. The standard deviation of the proportion corresponding to each frequency component is compared with a preset standard deviation threshold. The weights corresponding to the signal-to-noise ratio and baseline capacitance drift are set based on the comparison results between the standard deviation of the proportion corresponding to each frequency component and the preset standard deviation threshold.

[0013] Furthermore, the weights corresponding to the signal-to-noise ratio and baseline capacitance drift are set based on the comparison results between the standard deviation of the proportion corresponding to each frequency component and a preset standard deviation threshold, including: Based on the comparison results between the percentage standard deviation of each frequency component and the preset standard deviation threshold, it is determined whether there are frequency components whose percentage standard deviation is not lower than the preset standard deviation threshold. When there are no frequency components whose percentage standard deviation is not lower than the preset standard deviation threshold, the weights corresponding to the signal-to-noise ratio and the baseline capacitance drift are obtained by combining the percentage standard deviation of each frequency component with the current touch activity coefficient. When there are frequency components whose percentage standard deviation is not lower than the preset standard deviation threshold, the frequency components whose percentage standard deviation is not lower than the preset standard deviation threshold are selected as the target frequency components. The signal-to-noise ratio and the weights corresponding to the baseline capacitance drift are obtained by combining the standard deviation and percentage value of the target frequency component with the current touch activity coefficient.

[0014] Furthermore, the optimization control module selects the optimal historical collaborative optimization strategy from the strategy library, including: Obtain historical touch failure factors corresponding to historical collaborative optimization strategies, and match historical collaborative optimization strategies with historical touch failure factors; Based on historical touch failure factors, historical collaborative optimization strategies are categorized into levels, including: Mild interference - dust type and strategy A: If it is only local dust, then only dynamically adjust the touch parameters; Moderate Interference - Static Type and Strategy B: If diagnosed as being caused by dry static electricity, then simultaneously perform parameter adjustments and micro-electrostatic field dust removal; Severe Interference - Adhesion Type and Strategy C: If diagnosed as physically adhering dust particles, and predicted to severely affect touch control, then perform parameter adjustments, piezoelectric vibration, and environmental conditioning simultaneously.

[0015] Furthermore, the optimization control module, after dynamically adjusting the touch sensing layer and the signal processing module, includes: After implementing the optimization plan, record the touch performance data before and after the optimization. Re-collect environmental data and touch signals to evaluate the effectiveness of this optimization plan; If the evaluation shows that the optimization plan is effective, the optimization plan and the effect mapping are saved to the strategy library; If the evaluation results do not meet expectations, environmental data and touch signals will be collected again, and touch status diagnosis and prediction will be performed again. Touch optimization is performed based on the new predictions until touch performance is restored to a satisfactory level or the maximum number of optimizations is reached.

[0016] Compared with the prior art, the beneficial effects of the present invention are: This invention performs correlation analysis on historical data, presets dust distribution thresholds and relative humidity thresholds, and maps them to a raw data table. Real-time data exceeding the thresholds is marked as low-confidence data, and the corresponding original high-confidence data is extracted from the raw data table as replacement data to form a comprehensive environmental status index. The replacement mechanism ensures the continuity and reliability of the data, improving the robustness, fault tolerance, and reliability of the touch optimization system when facing real-world problems such as temporary sensor failures and abnormal readings. It also ensures the continuity of touch performance diagnosis and optimization strategies, providing users with a consistent and smooth touch experience. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the implementation of the optimized touch sensing layer system for the ultra-thin dustproof liquid crystal display screen of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] To address the technical issue that existing optimization systems overly rely on real-time data collected by sensors, often discarding crucial data when it is missing or obviously erroneous, leading to incomplete decision-making information or being forced to adopt default safety policies that may result in suboptimal outcomes, causing touch optimization functions to temporarily fail or degrade, please refer to [link to relevant documentation]. Figure 1 This embodiment provides the following technical solution: The touch sensing layer optimization system for ultra-thin dustproof LCD displays includes: The touch sensing layer is configured to be composed of an ultra-thin transparent nano-silver wire mesh and integrated into the inner side of the polarizer of the liquid crystal display or the inner surface of the encapsulation glass.

[0020] The environmental sensing module is configured to detect dust distribution data on the display screen surface and ambient temperature and humidity data in real time using a micro-particle sensor, a temperature sensor, and a humidity sensor. It then fuses the dust distribution data, temperature data, and humidity data to form a comprehensive environmental status indicator. Specifically, the temperature sensor is a thermistor installed at the edge of the display screen; the humidity sensor is a capacitive humidity sensor installed on the back of the display screen; and the dust sensor is an optical dust sensor installed on the side of the display screen. The environmental sensing module includes: The confidence analysis module is configured to acquire dust distribution data, temperature data, and humidity data of the display screen when it is in normal operating condition from historical data. It preprocesses the dust distribution data, temperature data, and humidity data of the display screen when it is in normal operating condition from historical data. The preprocessing includes: data cleaning, normalization processing, and time alignment processing. The processed dust distribution data, temperature data, and humidity data are matched according to time windows and mapped to form the original data table.

[0021] A correlation analysis is performed between historical dust distribution data and historical temperature and humidity data. For example, based on the principle that rising temperature leads to a decrease in relative humidity, the correlation between the two is assessed to determine whether it conforms to normal patterns. Based on the analysis results, dust distribution thresholds and relative humidity thresholds are preset to measure the confidence level of dust distribution data in relation to temperature and humidity data. For example, in a dry environment (humidity below 30% RH), the normal range for dust concentration is set to a higher value, with no more than 1000 dust particles per cubic meter of air, i.e., the dust distribution threshold for a dry environment is set to ≤1000. In a humid environment (humidity above 85% RH), the normal range for dust concentration is set to a lower value, with no more than 200 dust particles per cubic meter of air, i.e., the dust distribution threshold for a dry environment is set to ≤200. Real-time dust distribution data, temperature data, and humidity data are acquired and compared with dust distribution thresholds and relative humidity thresholds. If the dust distribution data exceeds the dust distribution threshold, this set of dust distribution data is marked as low-confidence data. For example, in a dry environment, dust is more likely to remain suspended in the air, resulting in a relatively high dust concentration; while in a humid environment, dust is more likely to adhere to the surface of objects, resulting in a relatively low dust concentration. When the humidity sensor reading shows that the environment is extremely humid, such as above 85% RH, the electrostatic adsorption effect should be significantly weakened. If the dust sensor reading still shows a sharp increase in dust concentration at this time, there is a physical and logical conflict between the two data sources, and this set of dust distribution data is determined to be abnormal, i.e., marked as low-confidence. A correlation analysis is performed between temperature and humidity data. If the humidity data exceeds the relative humidity threshold, the set of temperature and humidity data is marked as low confidence data. For example, if the internal temperature of the device rises sharply in a short period of time, and the humidity value does not decrease accordingly due to the high load on the processor, it violates the physical law that warm air can hold more water vapor. Therefore, it is determined that the humidity sensor data may be inaccurate or lagging, and it is marked as low confidence.

[0022] When the analysis results show low-confidence data, the data currently marked as low-confidence data is extracted; the original high-confidence data corresponding to the low-confidence data is mapped through the original data table and used as substitute data for the current low-confidence data, and marked as substitute data; the substitute data and normal data are combined to form a comprehensive environmental status index, with the substitute data given a lower weight and the normal data given a higher weight.

[0023] The signal processing module is configured to calculate the signal-to-noise ratio and baseline capacitance drift of the original signal from the touch sensing layer, and use the calculation results as a comprehensive touch performance index; input the comprehensive environmental state index and the comprehensive touch performance index into a lightweight neural network model to diagnose the main causes of the current touch performance degradation, such as whether it is caused by dry static electricity or by signal attenuation due to moisture and dust mixing, and predict the downward trend of touch performance; based on the prediction results, dynamically generate touch parameter adjustment instructions and dustproof drive instructions; the signal processing module includes: The data processing module is configured to acquire raw touch signals from the touch sensing layer, estimate the noise level in the signal using signal processing techniques, calculate the ratio of signal power to noise power to obtain the signal-to-noise ratio (SNR); measure the baseline capacitance value of the touch sensing layer when there is no touch operation; calculate the difference between the current measurement value and the initial baseline capacitance value to obtain the baseline capacitance drift; normalize the SNR and baseline capacitance drift to bring them into the same dimension for easier subsequent analysis; weight the SNR and baseline capacitance drift according to preset weights to generate a comprehensive touch index; and synchronize the comprehensive environmental state index with the comprehensive touch index in time to form the input samples for a lightweight neural network model.

[0024] The model building module is configured to collect comprehensive touch indicators, comprehensive environmental state indicators, and corresponding causes and trends of touch performance degradation from historical data. It uses historical data to train a lightweight neural network model, extracts spatial features from the comprehensive touch indicators, learns the spatial correlation of signal attenuation in different regions, and analyzes the time series relationship between the comprehensive environmental state indicators and the comprehensive touch indicators to capture their dynamic evolution law and obtain the relationship between the comprehensive touch indicators, the comprehensive environmental state indicators, and the causes of touch performance degradation.

[0025] The real-time diagnostic module is configured to input the comprehensive touch performance index and comprehensive environmental status index calculated in real time into a trained lightweight neural network model to diagnose the main causes of the current touch performance degradation, such as signal attenuation caused by dry static electricity, signal attenuation caused by moisture and dust mixing, etc.; and predict the downward trend of touch performance, including possible changes in touch performance in the future; for example: diagnosing 85% dry static electricity, 10% moisture and dust mixing, and 5% other factors; predicting that the overall signal-to-noise ratio will decrease linearly at a rate of 0.5 dB / s within the next 30 seconds.

[0026] The multi-dimensional analysis module is configured to generate verifiable physical correlation assumptions based on diagnostic and prediction results. For example, if the diagnosis is dominated by dry static electricity, the following verifiable assumptions exist: Assumption 1 (Signal Characteristics): A large amount of non-Gaussian, impulse noise should be detectable in the original touch signal. Assumption 2 (Environmental Correlation): Environmental data should meet the requirements of low humidity (H < 40% RH) and moderate temperature. Assumption 3 (Spatial Correlation): The decline in touch performance should be global or random, rather than limited to a specific area of ​​the screen, because static electricity has a wide range of effects. If the diagnosis is dominated by a mixture of moisture and dust, the following verifiable assumptions exist: Assumption 1 (Signal Characteristics): The touch signal should show monotonic drift of the baseline capacitance and a significant increase in harmonic distortion. Assumption 2 (Environmental Correlation): Environmental data should meet the requirements of high humidity (H > 75% RH) and high-density areas in the dust distribution matrix. Assumption 3 (Spatial Correlation): The decline in touch performance, such as low signal-to-noise ratio, should spatially overlap with high-density areas in the signal-to-noise ratio. Based on the generated hypotheses, the touch signal is re-analyzed, the statistical distribution of noise is calculated, and impulse noise is verified; harmonic distortion is calculated, and signal leakage is verified; and the current dust distribution data, temperature data, and humidity data are checked to see if they meet the assumed environmental conditions; based on the analysis results, the physical correlation assumptions are verified; if the verification passes, the original optimized instructions are executed; if the verification fails, a general degradation strategy is initiated; for example, only low-intensity, periodic physical cleaning, such as vibration, is initiated to avoid ineffective or even counterproductive optimizations due to misjudgment.

[0027] The strategy formulation module is configured to formulate corresponding touch parameter adjustment strategies and dust prevention measures based on the diagnostic results of historical data, forming a strategy library. For example, if the diagnostic results show signal attenuation caused by dry static electricity, the system may increase the electrostatic shielding strength of the touch sensing layer. If the diagnostic results show signal attenuation caused by moisture and dust mixing, the system may activate the dust prevention device to remove dust from the surface of the touch sensing layer.

[0028] The optimized control module is configured to select the optimal historical collaborative optimization strategy from the strategy library based on environmental monitoring results and generate specific control commands to dynamically adjust the touch sensing layer and signal processing module to optimize touch sensing performance; including: Obtain historical touch failure factors corresponding to historical collaborative optimization strategies, and match historical collaborative optimization strategies with historical touch failure factors; based on historical touch failure factors, classify historical collaborative optimization strategies into levels, including: Mild Interference - Microdust Type and Strategy A: If it is only local microdust, then only dynamically adjust the touch parameters; for example: temporarily increase the reporting rate of touch reports and the weight of the trajectory prediction algorithm in the dust-covered area to compensate for possible signal loss, and maintain standard parameters in the clean area to reduce power consumption.

[0029] Moderate Interference - Static Electricity Type and Strategy B: If diagnosed as being caused by dry static electricity, parameter adjustments and micro-static field dust removal are performed simultaneously; for example, by moderately increasing the global touch drive voltage to enhance signal penetration; at the same time, the dust removal drive module activates the transparent electrode array to generate a low-frequency alternating static field, neutralizing the charge and pushing away charged dust.

[0030] Severe Interference - Adhesion Type and Strategy C: If diagnosed as physically adhering dust particles, and predicted to severely affect touch control, then simultaneously perform parameter adjustments, piezoelectric vibration, and environmental conditioning; for example: while increasing the touch scanning frequency, start the micro piezoelectric vibrator to perform short pulse vibration of a specific spectrum; and based on the ambient humidity data, start the internal micro fan to provide weak airflow assistance.

[0031] After implementing the optimization plan, record the touch performance data before and after optimization; re-collect environmental data and touch signals to evaluate the effectiveness of the optimization plan; if the evaluation shows that the optimization plan is effective, save the optimization plan and its effect mapping to the policy library; for example, if the optimization plan is determined to be effective, i.e., the touch SNR is significantly improved and user operation is smooth again, then the optimization plan and its effect are mapped to the policy library to strengthen and correct the lightweight neural network model. If the evaluation shows that the effect does not meet expectations, re-collect environmental data and touch signals, and perform touch state diagnosis and prediction again; optimize the touch based on the new prediction results until the touch performance is restored to a satisfactory level or the maximum number of optimizations is reached.

[0032] The beneficial effects achieved by the above are as follows: By performing correlation analysis on historical data, dust distribution thresholds and relative humidity thresholds are preset and mapped to a raw data table; real-time data exceeding the thresholds are marked as low-confidence data, and the corresponding original high-confidence data is extracted from the raw data table as replacement data to form a comprehensive environmental status index; the replacement mechanism ensures the continuity and reliability of data, improves the robustness, fault tolerance, and reliability of the touch optimization system when facing real-world problems such as temporary sensor failures and abnormal readings, ensures the continuity of touch performance diagnosis and optimization strategies, and provides users with a consistent and smooth touch experience.

[0033] Working principle: By collecting and fusing temperature, humidity, and dust distribution data in real time, a comprehensive environmental status index is formed, and the signal-to-noise ratio and baseline drift of the touch signal are calculated as a comprehensive touch index. This is then input into a lightweight neural network model to diagnose the root cause of touch performance degradation and predict its trend. Based on the diagnostic results, the optimal collaborative optimization instruction is matched and executed from a pre-set strategy library. The optimization effect is continuously evaluated, thereby achieving dynamic maintenance and improvement of touch performance through continuous iteration, ensuring that the display can provide a smooth and reliable touch experience in various complex environments.

[0034] Specifically, the preset weights corresponding to the signal-to-noise ratio and baseline capacitance drift are set in the following ways: The touch frequency is collected in real time per unit time, wherein the value of the unit time ranges from 5 min to 8 min; The current touch activity coefficient Kt = 0.6*(f) is obtained by using the touch frequency acquired per unit time. t / f max +0.4p t ; where f t This indicates the touch frequency corresponding to the current unit of time; f max Indicates the maximum allowed touch frequency of the system; p t This represents the average touch pressure value after normalization. Retrieve the noise spectrum corresponding to each original touch signal collected; The proportion of each frequency component in the noise spectrum is obtained based on the noise spectrum corresponding to the original touch signal. The standard deviation of the proportion of each frequency component is obtained based on the proportion value of each original touch signal collected each time. The standard deviation of the proportion corresponding to each frequency component is compared with a preset standard deviation threshold. The weights corresponding to the signal-to-noise ratio and baseline capacitance drift are set based on the comparison results between the standard deviation of the proportion corresponding to each frequency component and the preset standard deviation threshold.

[0035] This embodiment combines real-time touch activity with the fine characteristics of the noise spectrum to achieve dynamic adaptation and precise control of the weights corresponding to the signal-to-noise ratio and baseline capacitance drift. The touch activity coefficient is calculated based on the touch frequency and average touch pressure per unit time, reflecting the intensity of user operations in real time, allowing the weight settings to adapt to system requirements under different touch frequency scenarios. Simultaneously, by extracting the proportion and standard deviation of each frequency component in the original touch signal noise spectrum and comparing it with a preset threshold, the distribution characteristics and fluctuation stability of noise in the frequency dimension can be accurately captured, enabling weight adjustments to respond specifically to dynamic noise changes. This weight setting method, combining operational activity and fine noise characteristics, improves the fit of comprehensive touch indicators to actual touch scenarios and environmental interference, enhances the system's adaptability to different touch intensities and noise fluctuation states, thereby optimizing the signal processing accuracy and anti-interference capability of the touch sensing layer, and improving the targeting and effectiveness of touch parameter adjustment strategies.

[0036] Specifically, the weights corresponding to the signal-to-noise ratio and baseline capacitance drift are set based on the comparison results between the standard deviation of the proportion corresponding to each frequency component and a preset standard deviation threshold, including: Based on the comparison results between the percentage standard deviation of each frequency component and the preset standard deviation threshold, it is determined whether there are frequency components whose percentage standard deviation is not lower than the preset standard deviation threshold. When there are no frequency components whose percentage standard deviation is not lower than the preset standard deviation threshold, the weights corresponding to the signal-to-noise ratio and the baseline capacitance drift are obtained by combining the percentage standard deviation of each frequency component with the current touch activity coefficient. The weights corresponding to the signal-to-noise ratio and baseline capacitance drift are obtained using the following formula: Among them, w 01 and w 02 These represent the weights corresponding to the signal-to-noise ratio and baseline capacitance drift, respectively; σ max and σ min σ represents the maximum and minimum values ​​of the percentage standard deviations for all frequency components, respectively; p Kt represents the average standard deviation of the proportions of all frequency components; Kt represents the current touch activity coefficient. When there are frequency components whose percentage standard deviation is not lower than the preset standard deviation threshold, the frequency components whose percentage standard deviation is not lower than the preset standard deviation threshold are selected as the target frequency components. The signal-to-noise ratio and the weights corresponding to the baseline capacitance drift are obtained by combining the standard deviation and percentage value of the target frequency component with the current touch activity coefficient.

[0037] The weights corresponding to the signal-to-noise ratio and baseline capacitance drift are obtained using the following formula: Among them, w 01 and w 02 These represent the weights corresponding to the signal-to-noise ratio and baseline capacitance drift, respectively; σ mp σ represents the average standard deviation of the proportions of all target frequency components; y Kt represents the preset standard deviation threshold; B represents the current touch activity coefficient; c B represents the difference between the maximum and minimum percentage values ​​of the target frequency component; p This represents the average percentage value corresponding to the target frequency component.

[0038] This embodiment uses a comparison between the standard deviation of the frequency component proportions in the noise spectrum and a preset threshold to switch the weight setting logic across different scenarios, ensuring that weight adjustments accurately match the fluctuation state of noise. When there are no frequency components exceeding the threshold, the weight is calculated based on the statistical characteristics of the standard deviation of all frequency components (maximum, minimum, and average values) combined with the touch activity coefficient. This comprehensively reflects the stability of the overall noise environment, ensuring that the weight adapts to the system requirements in a stable noise scenario. When there are frequency components exceeding the threshold, the weight calculation focuses on the core statistical parameters of the target frequency component (standard deviation, average value, proportion difference, and proportion average value). This allows for targeted responses to interference from localized strong fluctuation noise, preventing localized noise anomalies from undermining the overall rationality of the weight. On the one hand, by deeply mining the fine characteristics of the noise spectrum (frequency component proportion and standard deviation), it breaks through the limitations of traditional methods that judge interference intensity solely based on overall noise power, achieving precise capture of noise fluctuation patterns. This allows weight adjustments to more accurately offset the impact of noise on touch signals. On the other hand, by integrating the touch activity coefficient into the weight calculation process for both scenarios, the weight settings simultaneously consider the dynamic changes in user operation intensity. This ensures that under different touch frequencies and touch pressure scenarios, the comprehensive touch index can always accurately reflect the actual touch state, avoiding index distortion caused by changes in operation intensity. Through refined weight settings for different scenarios, the contribution of signal-to-noise ratio and baseline capacitance drift in the comprehensive touch index can be dynamically optimized according to noise conditions and operational behavior. This effectively reduces the interference of noise fluctuations and capacitance drift on touch signal recognition, reducing the probability of accidental touches and touch delays. Meanwhile, precise weight allocation can provide higher quality input samples for subsequent lightweight neural network models, improve the accuracy of the model in diagnosing touch status, and thus enhance the pertinence and effectiveness of touch parameter adjustment strategies and dust prevention measures, comprehensively optimizing the working performance and stability of the ultra-thin dustproof LCD touch sensing layer.

[0039] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0040] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An optimized touch sensing layer system for an ultra-thin dustproof LCD display, characterized in that, include: The touch sensing layer is configured to be composed of an ultra-thin transparent nano-silver wire mesh and integrated into the inner side of the polarizer of the liquid crystal display or the inner surface of the encapsulation glass. The environmental sensing module is configured to detect dust distribution data on the display screen surface and ambient temperature and humidity data in real time through micro-particle sensors, temperature sensors, and humidity sensors; and to fuse the dust distribution data, temperature data, and humidity data to form a comprehensive environmental status index. The signal processing module is configured to calculate the signal-to-noise ratio and baseline capacitance drift of the original signal of the touch sensing layer, and use the calculation results as a comprehensive touch index; input the comprehensive environmental state index and the comprehensive touch index into a lightweight neural network model to diagnose the main causes of the current touch performance degradation and predict the trend of touch performance degradation; based on the prediction results, dynamically generate touch parameter adjustment instructions and dustproof drive instructions. The optimized control module is configured to select the optimal historical collaborative optimization strategy from the strategy library based on environmental monitoring results and generate specific control commands to dynamically adjust the touch sensing layer and signal processing module. The environmental perception module includes: The confidence analysis module is configured to acquire dust distribution data, temperature data, and humidity data from historical data when the display screen is in normal operating condition. A correlation analysis was performed between historical dust distribution data and historical temperature and humidity data. Based on the analysis results, dust distribution thresholds and relative humidity thresholds were preset to measure the confidence level of dust distribution data in relation to temperature and humidity data. Acquire real-time dust distribution data, temperature data, and humidity data, and compare them with dust distribution thresholds and relative humidity thresholds. If the dust distribution data exceeds the dust distribution thresholds, mark this set of dust distribution data as low-confidence data. A correlation analysis was performed between temperature and humidity data. If the humidity data exceeded the relative humidity threshold, the set of temperature and humidity data was marked as low confidence data. After the confidence analysis module marks the low-confidence data, it includes: Preprocessing is performed on historical data of dust distribution, temperature, and humidity when the display screen is in normal operation. Preprocessing includes: data cleaning, normalization, and time alignment. The processed dust distribution data, temperature data, and humidity data are matched according to time windows and mapped to form the original data table; When the analysis results show that there is low-confidence data, extract the data currently marked as low-confidence. The original high-confidence data corresponding to the low-confidence data is mapped from the original data table and used as the substitute data for the current low-confidence data, and marked as the substitute data. The substitute data and normal data are combined to form a comprehensive environmental status index, with the substitute data given a lower weight and the normal data given a higher weight. The optimization control module selects the optimal historical collaborative optimization strategy from the strategy library, including: Obtain historical touch failure factors corresponding to historical collaborative optimization strategies, and match historical collaborative optimization strategies with historical touch failure factors; Based on historical touch failure factors, historical collaborative optimization strategies are categorized into levels, including: Mild interference - dust type and strategy A: If it is only local dust, then only dynamically adjust the touch parameters; Moderate Interference - Static Type and Strategy B: If diagnosed as being caused by dry static electricity, then simultaneously perform parameter adjustments and micro-electrostatic field dust removal; Severe Interference - Adhesion Type and Strategy C: If diagnosed as physically adhering dust particles, and predicted to severely affect touch control, then perform parameter adjustments, piezoelectric vibration, and environmental conditioning simultaneously.

2. The touch sensing layer optimization system for an ultra-thin dustproof liquid crystal display screen according to claim 1, characterized in that, The signal processing module includes: The model building module is configured to collect comprehensive touch indicators, comprehensive environmental state indicators, and corresponding causes and trends of touch performance degradation from historical data. It uses historical data to train a lightweight neural network model, extracts spatial features from the comprehensive touch indicators, learns the spatial correlation of signal attenuation in different regions, and analyzes the time series relationship between the comprehensive environmental state indicators and the comprehensive touch indicators to capture their dynamic evolution law and obtain the relationship between the comprehensive touch indicators, the comprehensive environmental state indicators, and the causes of touch performance degradation. The real-time diagnostic module is configured to input the comprehensive touch performance index and comprehensive environmental status index calculated in real time into a trained lightweight neural network model to diagnose the main reasons for the current decline in touch performance in real time; and predict the decline trend of touch performance, including possible changes in touch performance in the future.

3. The touch sensing layer optimization system for an ultra-thin dustproof liquid crystal display screen according to claim 2, characterized in that, The signal processing module further includes: The multi-dimensional analysis module is configured to generate verifiable physical correlation assumptions based on diagnostic and prediction results. Based on the generated hypotheses, the touch signal is re-analyzed, the statistical distribution of noise is calculated, and the impulse noise is verified; the harmonic distortion is calculated, and the signal leakage is verified; and the current dust distribution data, temperature data, and humidity data are examined to see if they meet the assumed environmental conditions. Based on the analysis results, verify the physical correlation assumptions; if the verification passes, execute the generated original optimization instructions; if the verification fails, initiate the general degradation strategy.

4. The touch sensing layer optimization system for an ultra-thin dustproof liquid crystal display screen according to claim 3, characterized in that, The signal processing module further includes: The data processing module is configured to acquire raw touch signals from the touch sensing layer, estimate the noise level in the signal through signal processing technology, calculate the ratio of signal power to noise power, and obtain the signal-to-noise ratio. When there is no touch operation, the baseline capacitance value of the touch sensing layer is measured; the difference between the current measured value and the initial baseline capacitance value is calculated to obtain the baseline capacitance drift; the signal-to-noise ratio and the baseline capacitance drift are normalized to bring them into the same dimension; according to the preset weights, the signal-to-noise ratio and the baseline capacitance drift are weighted and averaged to generate a comprehensive touch index; the comprehensive environmental state index and the comprehensive touch index are time-synchronized and aligned to form the input samples of the lightweight neural network model. The strategy formulation module is configured to formulate corresponding touch parameter adjustment strategies and dust prevention measures based on the diagnostic results of historical data, forming a strategy library.

5. The touch sensing layer optimization system for an ultra-thin dustproof liquid crystal display screen according to claim 4, characterized in that, The preset weights corresponding to the signal-to-noise ratio and baseline capacitance drift are set in the following ways: The touch frequency is collected in real time per unit time, wherein the value of the unit time ranges from 5 min to 8 min; The current touch activity coefficient Kt is obtained by using the touch frequency acquired per unit of time. Retrieve the noise spectrum corresponding to each original touch signal collected; The proportion of each frequency component in the noise spectrum is obtained based on the noise spectrum corresponding to the original touch signal. The standard deviation of the proportion of each frequency component is obtained based on the proportion value of each original touch signal collected each time. The standard deviation of the proportion corresponding to each frequency component is compared with a preset standard deviation threshold. The weights corresponding to the signal-to-noise ratio and baseline capacitance drift are set based on the comparison results between the standard deviation of the proportion corresponding to each frequency component and the preset standard deviation threshold.

6. The touch sensing layer optimization system for an ultra-thin dustproof liquid crystal display screen according to claim 5, characterized in that, The weights corresponding to the signal-to-noise ratio and baseline capacitance drift are set based on the comparison results between the standard deviation of the proportion corresponding to each frequency component and a preset standard deviation threshold, including: Based on the comparison results between the percentage standard deviation of each frequency component and the preset standard deviation threshold, it is determined whether there are frequency components whose percentage standard deviation is not lower than the preset standard deviation threshold. When there are no frequency components whose percentage standard deviation is not lower than the preset standard deviation threshold, the weights corresponding to the signal-to-noise ratio and the baseline capacitance drift are obtained by combining the percentage standard deviation of each frequency component with the current touch activity coefficient. When there are frequency components whose percentage standard deviation is not lower than the preset standard deviation threshold, the frequency components whose percentage standard deviation is not lower than the preset standard deviation threshold are selected as the target frequency components. The signal-to-noise ratio and the weights corresponding to the baseline capacitance drift are obtained by combining the standard deviation and percentage value of the target frequency component with the current touch activity coefficient.

7. The touch sensing layer optimization system for an ultra-thin dustproof liquid crystal display screen according to claim 1, characterized in that, The optimized control module, after dynamically adjusting the touch sensing layer and the signal processing module, includes: After implementing the optimization plan, record the touch performance data before and after the optimization. Re-collect environmental data and touch signals to evaluate the effectiveness of this optimization plan; If the evaluation shows that the optimization plan is effective, the optimization plan and the effect mapping are saved to the strategy library; If the evaluation results do not meet expectations, environmental data and touch signals will be collected again, and touch status diagnosis and prediction will be performed again. Touch optimization is performed based on the new predictions until touch performance is restored to a satisfactory level or the maximum number of optimizations is reached.

Images