An AI technology-based performance detection method and system for a visual interactive touch screen

Abstract

The application is suitable for the technical field of touch screen performance detection, and provides a performance detection method and system of visual interactive touch screen based on AI technology, which comprises the following steps: after multi-point touch detection is performed on a display unit to be detected, if the measurement data of the capacitive sensor obtained is abnormal, the current touch working mode is determined, and the initial gain amplitude parameter prepared for the capacitive sensor and the historical processing record of the noise conditioning device used for processing the measurement data of the capacitive sensor are obtained. The initial gain amplitude parameter is dynamically corrected by combining the first correction factor and the second correction factor, so that the adaptive optimization of the processing capacity of the noise conditioning device is realized. The first correction factor is used for evaluating the noise baseline drift condition and reflecting the long-term noise change trend, and the second correction factor is used for analyzing the change condition of the calibration error and evaluating the real-time correction effect of the noise conditioning device.

Description

Technical Field

[0001] This invention belongs to the field of touch screen performance testing technology, and in particular relates to a performance testing method and system for a visual interactive touch screen based on AI technology. Background Technology

[0002] Visual interactive touchscreens are display devices that enable user input through touch, and are widely used in smart terminals, industrial control panels, medical equipment, and automotive systems. These touchscreens typically utilize capacitive sensor technology, determining the user's input position by detecting changes in capacitance at the touch point, and enabling complex interactive operations based on multi-touch technology. The performance of the touchscreen directly impacts the user experience, with core testing parameters including touch sensitivity, response time, anti-interference capability, and signal stability. Among these, the measurement data from the capacitive sensor is a key indicator, directly determining the accuracy and stability of touch input. However, in practical applications, the operating environment of touchscreens is complex and variable. Factors such as temperature changes, electromagnetic interference, and equipment aging due to prolonged operation can all affect the touch signal. Therefore, accurate testing of the performance of visual interactive touchscreens is crucial.

[0003] In existing technologies, touchscreen performance testing typically relies on the analysis of capacitive sensor measurement data and the optimization of touch signal quality through gain amplitude compensation to reduce interference from environmental factors. Furthermore, to improve the touchscreen's noise immunity, existing technologies commonly employ noise conditioning devices to process sensor signals, such as filtering, correction, and noise suppression, to ensure the stability of touch data. However, these traditional testing and optimization methods have certain limitations. First, the gain amplitude compensation setting is usually fixed, failing to adaptively adjust to noise changes in different environments, resulting in limited touch accuracy. Second, the noise suppression capability of noise conditioning devices may gradually decrease during long-term operation, but existing technologies lack dynamic assessment of this change. This leads to continued compensation using the original strategy even when noise baseline drift or calibration errors increase, potentially amplifying noise interference and affecting touch accuracy. Moreover, existing technologies fail to effectively identify the relationship between abnormal capacitive sensor measurement data and noise changes. Even when noise is significant, a fixed gain strategy is still used, ignoring the impact of noise amplification on measurement data deviation. Summary of the Invention

[0004] The purpose of this invention is to provide a performance testing method and system for a visual interactive touchscreen based on AI technology, aiming to solve the problems mentioned in the background art.

[0005] This invention is implemented as follows: a performance testing method for a visual interactive touchscreen based on AI technology, the method comprising:

[0006] After performing multi-touch detection on the display unit under test, if the acquired capacitive sensor measurement data is abnormal, the current touch working mode is determined, and the initial gain amplitude parameters prepared for the capacitive sensor and the historical processing records of the noise conditioning device used to process the capacitive sensor measurement data are acquired.

[0007] Based on AI technology, historical processing records are analyzed to select a predetermined number of logs for specific time periods where the processed display unit is of the same type as the tested display unit, the capacitive sensor measurement data is normal, the time interval is consistent, and there is a calibration event.

[0008] Based on the preset reference model, the current test reference noise mean value of the noise conditioning device processing the capacitive sensor measurement data of the display unit under test in the current touch working mode is determined, the historical noise mean value of the log for each specific time period is determined, and the first correction factor is generated according to the changing trend of the deviation of each historical noise mean value relative to the current test reference noise mean value.

[0009] The logs for a specific time period are analyzed, and the average noise values ​​within a predetermined time period before and after the calibration time are compared to obtain the calibration error value. A second correction factor is generated based on the changing trends of several calibration error values.

[0010] The initial gain amplitude parameter is adjusted by combining the first correction factor and the second correction factor.

[0011] As a further limitation of the technical solution of the present invention, the preset reference model refers to the average reference noise value obtained in each touch working mode when the noise conditioning device processes the capacitive sensor data measured by different types of display units under standard working conditions.

[0012] As a further limitation of the technical solution of this invention, the steps of determining the current test reference noise mean value of the noise conditioning device processing the capacitive sensor measurement data of the display unit under test in the current touch working mode based on a preset reference model, determining the historical noise mean value of the log for each specific time period, and generating a first correction factor based on the changing trend of the deviation amplitude of each historical noise mean value relative to the current test reference noise mean value include:

[0013] Retrieve the preset reference model, determine the specific type of the display unit under test, and use the specific type and the current touch working mode as input to determine the corresponding current test benchmark noise mean;

[0014] Analyze the logs for the selected specific time period and calculate the historical noise mean for each specific time period.

[0015] Quantify the deviation of each historical noise mean relative to the current test baseline noise mean, plot these deviation values ​​in chronological order as a curve reflecting the trend of noise baseline drift, calculate the average slope of the curve, and use it as the first correction factor.

[0016] As a further limitation of the technical solution of this embodiment of the invention, the steps of parsing logs for a specific time period, comparing the average noise values ​​within a predetermined time period before and after the calibration time to obtain calibration error values, and generating a second correction factor based on the changing trends of several calibration error values ​​include:

[0017] Analyze logs for a specific time period and calculate the mean noise before and after calibration within the predetermined time period.

[0018] The calibration error value is calculated based on the mean noise before calibration and the mean noise after calibration;

[0019] The calibration error values ​​from logs for several specific time periods are plotted in chronological order to create a curve reflecting the trend of calibration error changes. The average slope of this curve is calculated and used as the second correction factor.

[0020] As a further limitation of the technical solution of this embodiment of the invention, the step of adjusting the initial gain amplitude parameter by combining the first correction factor and the second correction factor includes:

[0021] The preset gain amplitude parameter adjustment formula is invoked, and the initial gain amplitude parameter is dynamically corrected by combining the first correction factor and the second correction factor to obtain the optimized gain amplitude parameter.

[0022] The optimized gain amplitude parameter is applied to the sensitivity adjustment unit of the capacitive sensor of the display unit under test.

[0023] As a further limitation of the technical solution of this embodiment of the invention, the formula for adjusting the gain amplitude parameter is as follows: ,in This refers to the optimized gain amplitude parameter. This refers to the initial gain amplitude parameter. This refers to the first correction factor, which is the average slope of the curve reflecting the trend of noise baseline drift. This refers to the adjustment coefficient corresponding to the first correction factor. This refers to the second correction factor, which is the average slope of the curve reflecting the trend of calibration error changes. This refers to the adjustment coefficient corresponding to the second correction factor;

[0024] In the formula for adjusting the gain amplitude parameter ,in This refers to the total number of logs within a specific time period. It refers to the first The midpoint of a specific time period log. This refers to the average of the median times of all logs for a specific time period. Indicates the first The magnitude of the deviation between the historical noise mean and the current test baseline noise mean in the log for a specific time period. This refers to the average deviation magnitude values ​​corresponding to all logs for a specific time period;

[0025] ,in It refers to the first Historical noise mean of logs for a specific time period This refers to the current benchmark noise mean;

[0026] ,in It refers to the first The time point of calibration events in a specific time period log. This refers to the average time point of calibration events across all logs for a specific time period. It refers to the first The calibration error value between the mean noise before calibration and the mean noise after calibration in the log for a specific time period. This refers to the average of the calibration error values ​​corresponding to all logs for a specific time period;

[0027] ,in It refers to the first Mean noise before calibration in logs for a specific time period. It refers to the first The mean noise after calibration in the logs for a specific time period.

[0028] A performance testing system for a visual interactive touchscreen based on AI technology, the system comprising: a data acquisition module, a historical record filtering module, a first correction factor determination module, a second correction factor determination module, and a gain amplitude parameter optimization module, wherein:

[0029] The data acquisition module is used to determine the current touch working mode if the acquired capacitive sensor measurement data is abnormal after multi-touch detection of the display unit under test, and to acquire the initial gain amplitude parameters prepared for the capacitive sensor and the historical processing records of the noise conditioning device used to process the capacitive sensor measurement data.

[0030] The historical record filtering module is used to parse historical processing records and filter out a predetermined number of logs for specific time periods where the processed display unit is of the same type as the tested display unit, the capacitive sensor measurement data is normal, the time interval is consistent, and there is a calibration event.

[0031] The first correction factor determination module is used to determine the current test reference noise mean value of the noise conditioning device processing the capacitive sensor measurement data of the display unit under test in the current touch working mode based on the preset reference model, determine the historical noise mean value of the log for each specific time period, and generate the first correction factor according to the changing trend of the deviation of each historical noise mean value relative to the current test reference noise mean value.

[0032] The preset reference model refers to the average reference noise value obtained in each touch operation mode when the noise conditioning device processes the capacitive sensor data measured by different types of display units under standard operating conditions.

[0033] The second correction factor determination module is used to parse the logs for a specific time period, compare the noise mean within a predetermined time period before and after the calibration time to obtain the calibration error value, and generate a second correction factor based on the changing trend of several calibration error values.

[0034] The gain amplitude parameter optimization module is used to adjust the initial gain amplitude parameter by combining the first correction factor and the second correction factor.

[0035] As a further limitation of the technical solution of this embodiment of the invention, the first correction factor determination module specifically includes:

[0036] The reference noise mean determination unit is used to retrieve a preset reference model, determine the specific type of the display unit under test, and use the specific type and the current touch working mode as input to determine the corresponding current test reference noise mean.

[0037] The historical noise mean calculation unit is used to parse the logs for a selected specific time period and calculate the historical noise mean corresponding to each specific time period log.

[0038] The first correction factor calculation unit is used to quantify the deviation of each historical noise mean relative to the current test reference noise mean, plot these deviation values ​​in chronological order as a curve reflecting the trend of noise baseline drift, calculate the average slope of the curve, and use it as the first correction factor.

[0039] As a further limitation of the technical solution of this embodiment of the invention, the second correction factor determination module specifically includes:

[0040] The noise mean calculation unit before and after calibration is used to parse logs for a specific time period and calculate the noise mean before calibration and the noise mean after calibration within a predetermined time period.

[0041] The calibration error value determination unit is used to calculate the calibration error value based on the mean noise before calibration and the mean noise after calibration.

[0042] The second correction factor calculation unit is used to plot the corresponding calibration error values ​​in the logs of several specific time periods into a curve reflecting the trend of calibration error change in chronological order, calculate the average slope of the curve, and use it as the second correction factor.

[0043] As a further limitation of the technical solution of this embodiment of the invention, the gain amplitude parameter optimization module specifically includes:

[0044] The initial gain amplitude parameter optimization unit is used to call the preset gain amplitude parameter adjustment formula, and dynamically correct the initial gain amplitude parameter by combining the first correction factor and the second correction factor to obtain the optimized gain amplitude parameter.

[0045] An optimized gain amplitude parameter application unit is used to apply the optimized gain amplitude parameter to the sensitivity adjustment unit of the capacitive sensor of the display unit under test.

[0046] The formula for adjusting the gain amplitude parameter is: ,in This refers to the optimized gain amplitude parameter. This refers to the initial gain amplitude parameter. This refers to the first correction factor, which is the average slope of the curve reflecting the trend of noise baseline drift. This refers to the adjustment coefficient corresponding to the first correction factor. This refers to the second correction factor, which is the average slope of the curve reflecting the trend of calibration error changes. This refers to the adjustment coefficient corresponding to the second correction factor;

[0047] In the formula for adjusting the gain amplitude parameter ,in This refers to the total number of logs within a specific time period. It refers to the first The midpoint of a specific time period log. This refers to the average of the median times of all logs for a specific time period. Indicates the first The magnitude of the deviation between the historical noise mean and the current test baseline noise mean in the log for a specific time period. This refers to the average deviation magnitude values ​​corresponding to all logs for a specific time period;

[0048] ,in It refers to the first Historical noise mean of logs for a specific time period This refers to the current benchmark noise mean;

[0049] ,in It refers to the first The time point of calibration events in a specific time period log. This refers to the average time point of calibration events across all logs for a specific time period. It refers to the first The calibration error value between the mean noise before calibration and the mean noise after calibration in the log for a specific time period. This refers to the average of the calibration error values ​​corresponding to all logs for a specific time period;

[0050] ,in It refers to the first Mean noise before calibration in logs for a specific time period. It refers to the first The mean noise after calibration in the logs for a specific time period.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] This invention achieves adaptive optimization of the noise conditioning device's processing capability by combining a first correction factor and a second correction factor to dynamically correct the initial gain amplitude parameter. The first correction factor is used to assess the noise baseline drift and reflect long-term noise variation trends, while the second correction factor is used to analyze the changes in calibration error and evaluate the real-time correction effect of the noise conditioning device.

[0053] By combining these two approaches, this invention can accurately identify the correlation between abnormal capacitive sensor measurement data and noise effects, and reduce the gain amplitude parameter when noise is significant. This avoids the bias problem of relying solely on gain amplitude compensation while ignoring the impact of noise amplification, as seen in existing technologies. Compared to existing technologies, this invention effectively reduces measurement errors caused by noise interference, improves the stability and accuracy of touch signals, and makes gain adjustment more intelligent and adaptive. Attached Figure Description

[0054] Figure 1 A flowchart of the method provided in the embodiments of the present invention;

[0055] Figure 2 This is a flowchart illustrating the determination of the first correction factor in the method provided in this embodiment of the invention;

[0056] Figure 3 This is a flowchart illustrating the determination of the second correction factor in the method provided in this embodiment of the invention;

[0057] Figure 4 This is a flowchart illustrating the optimization of the initial gain amplitude parameter in the method provided in this embodiment of the invention;

[0058] Figure 5Application architecture diagram of the system provided in the embodiments of the present invention;

[0059] Figure 6 This is a structural block diagram of the first correction factor determination module in the system provided in the embodiments of the present invention;

[0060] Figure 7 This is a structural block diagram of the second correction factor determination module in the system provided in the embodiments of the present invention;

[0061] Figure 8 This is a structural block diagram of the gain amplitude parameter optimization module in the system provided in the embodiment of the present invention. Detailed Implementation

[0062] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0063] Figure 1 A flowchart of the method provided by an embodiment of the present invention is shown.

[0064] Specifically, a performance testing method for a visual interactive touchscreen based on AI technology includes the following steps:

[0065] Step S100: After performing multi-touch detection on the display unit under test, if the acquired capacitive sensor measurement data is abnormal, the current touch working mode is determined, and the initial gain amplitude parameters prepared for the capacitive sensor and the historical processing records of the noise conditioning device used to process the capacitive sensor measurement data are acquired.

[0066] In this embodiment of the invention, the display unit under test refers to a display device that undergoes multi-touch detection, and this invention is particularly aimed at visual interactive touchscreens. The display unit under test includes a capacitive touch sensor for sensing touch operations and converting them into electrical signals for subsequent processing and analysis. Multi-touch detection refers to the simultaneous detection of multiple touch points on the display unit, and is commonly used in smart devices, industrial control panels, medical touchscreens, and other scenarios to detect the accuracy of touch response, gesture recognition capabilities, and device stability.

[0067] Capacitive sensor measurement data refers to the electrical signal output by the sensor when detecting touch events, including capacitance value, noise level, response time, and other data. When the measured data deviates from the expected range, such as abnormal fluctuations in capacitance value, loss of touch signal, or accidental touch, it indicates that the touch sensor may be malfunctioning. Specific anomaly determination can be based on historical data comparison, threshold detection, or AI anomaly recognition methods.

[0068] The current touch operation mode refers to the working state of the touch sensor under different application scenarios or user operating habits, including touch sensitivity settings, signal sampling mode, and input signal filtering method. Different touch modes may affect the way touch data is collected. Therefore, when detecting anomalies, it is necessary to determine the current touch operation mode so that subsequent corrective measures can be more targeted.

[0069] The initial gain amplitude parameter for this capacitive sensor refers to the signal gain adjustment parameter set for the sensor under default or preset conditions, used to optimize the acquisition accuracy and noise suppression capability of the touch signal. This parameter can be obtained through experimental data, sensor calibration, or automatic system adjustment. In the prior art, there are already methods based on automatic gain control (AGC) to dynamically adjust the sensor gain.

[0070] Noise conditioning devices used for processing capacitive sensor measurement data refer to hardware or software modules that process touch signals to reduce noise interference and improve signal accuracy. Their historical processing records refer to the conditioning processes and results performed on the capacitive sensor measurement data by the device at different times and under different test conditions. These records can be obtained from device logs, noise filtering algorithm logs, or the data storage of the sensor signal processing unit.

[0071] Historical processing records should include at least the following: First, each record should indicate the type of the display unit under test to ensure that the comparison is made with the same or similar types of devices during subsequent analysis; second, it should include the processing results of the noise conditioning device in different touch operation modes to compare the noise characteristics under different modes; in addition, it should record the specific time of the calibration event and store the noise data before and after calibration to analyze the impact of calibration on noise; finally, it should record the specific time of each measurement to facilitate subsequent trend analysis in chronological order.

[0072] Furthermore, the performance testing method for the AI-based visual interactive touchscreen also includes the following steps:

[0073] Step S200: Based on AI technology, analyze historical processing records and select a predetermined number of logs for specific time periods that are consistent with the type of the processed display unit and the tested display unit, have no abnormalities in the capacitive sensor measurement data, have consistent time intervals, and have a calibration event.

[0074] In this embodiment of the invention, selecting logs from specific time periods is significant in ensuring that the historical data used to calculate the correction factor is of high quality and comparable. By filtering logs that meet specific conditions, interference from irrelevant factors can be reduced, the impact of abnormal data can be minimized, and the accuracy of the correction factor calculation can be ensured. This guarantees that when adjusting the gain amplitude parameter, the data used can truly reflect the changing trends of noise baseline drift and calibration error, making the final optimized gain amplitude parameter more accurate and reliable.

[0075] Several factors need to be considered when filtering logs, primarily to ensure data stability and representativeness. First, logs consistent with the type of display unit being tested should be selected to avoid data mismatches caused by differences in device characteristics. Second, capacitive sensor measurements should be anomalies-free to eliminate data deviations caused by equipment malfunctions or environmental interference. Third, consistent time intervals should be ensured to guarantee uniform data distribution and prevent trend analysis errors due to uneven data density. Finally, each log should include one calibration event. This is to calculate calibration error and generate a second correction factor, ensuring that the optimized gain amplitude parameter adapts to performance changes in different calibration states. Combining these filtering criteria ensures high consistency of the data used for analysis, making the calculated correction factor more valuable.

[0076] Furthermore, the performance testing method for the AI-based visual interactive touchscreen also includes the following steps:

[0077] Step S300: Based on the preset reference model, determine the current test reference noise mean value of the noise conditioning device processing the capacitive sensor measurement data of the display unit under test in the current touch working mode, determine the historical noise mean value of the log for each specific time period, and generate a first correction factor based on the changing trend of the deviation amplitude of each historical noise mean value relative to the current test reference noise mean value.

[0078] The preset reference model refers to the average baseline noise value obtained under standard operating conditions when the noise conditioning device processes capacitive sensor data measured by different types of display units in each touch operation mode.

[0079] Specifically, Figure 2 A flowchart for determining the first correction factor is shown.

[0080] The process of determining the current test baseline noise mean of the noise conditioning device processing capacitive sensor measurement data of the display unit under test in the current touch operation mode based on a preset reference model, determining the historical noise mean of the log for each specific time period, and generating a first correction factor based on the changing trend of the deviation of each historical noise mean relative to the current test baseline noise mean specifically includes the following steps:

[0081] Step S301: Retrieve the preset reference model, determine the specific type of the display unit under test, and use the specific type and the current touch working mode as input to determine the corresponding current test benchmark noise mean value;

[0082] Step S302: parse the logs for the selected specific time period and calculate the historical noise mean for each specific time period log.

[0083] Step S303: Quantify the deviation of each historical noise mean relative to the current test reference noise mean, plot these deviation values ​​in chronological order as a curve reflecting the trend of noise baseline drift, calculate the average slope of the curve, and use it as the first correction factor.

[0084] In this embodiment of the invention, the preset reference model is used to provide the baseline noise mean of the capacitive sensor measurement data by the noise conditioning device under different touch operating modes. The model is established based on a large amount of experimental data, covering different types of display units and their noise characteristics under standard operating conditions, and is optimized by combining existing noise analysis and filtering techniques. In the prior art, the noise characteristics of touch systems can be fitted through experimental measurement, statistical modeling, or machine learning methods; therefore, this preset reference model is feasible and can be constructed based on existing technologies.

[0085] The process of building this model typically includes the following aspects:

[0086] First, standardized tests are conducted in a controlled environment for different types of display units, recording noise data under different touch operation modes, including background noise, capacitive sensor response data, and the processing results of the noise conditioning device. Second, based on the collected data, mathematical models are established, such as those based on multivariate regression, probability distribution modeling, or neural network fitting, to analyze the variation law of the noise mean under different conditions. Finally, through data verification and model optimization, the preset reference model can accurately reflect the noise level under standard operating conditions, providing an accurate reference for subsequent noise deviation analysis.

[0087] The specific implementation process of step S303 is as follows: First, extract the historical noise mean from the log of a specific time period, and calculate the deviation of each historical noise mean relative to the current test reference noise mean. The deviation is calculated by subtracting the current test reference noise mean from the historical noise mean, then dividing by the current test reference noise mean, and taking the absolute value to measure the degree of deviation of the historical noise mean from the standard value.

[0088] Secondly, all calculated deviation amplitude values ​​are sorted according to the time points corresponding to the logs of specific time periods, and plotted in chronological order as a curve of the noise baseline drift trend. This curve can intuitively reflect the change law of noise deviation over time, so as to evaluate the long-term stability of the noise conditioning device on the measurement data of the capacitive sensor.

[0089] Finally, the curve is analyzed using linear regression to calculate the average slope of the trend change. The average slope reflects the degree of noise baseline drift. This slope is used as the first correction factor to adjust the initial gain amplitude parameter so that the adjusted gain amplitude parameter can compensate for the measurement error caused by noise baseline drift.

[0090] The significance of this method lies in the fact that noise baseline drift is a dynamic process that changes over time. Data from a single measurement may be affected by random factors and cannot accurately describe the overall trend. Therefore, by analyzing historical noise deviation changes over a period of time, the performance fluctuations of the noise conditioning device can be assessed more stably, thereby improving the correction effect on capacitive sensor measurement data. Furthermore, this method can effectively reduce the impact of single-point outliers on the correction factor, making the calculated first correction factor more reliable and providing a more valuable reference for the subsequent optimization of gain amplitude parameters.

[0091] Furthermore, the performance testing method for the AI-based visual interactive touchscreen also includes the following steps:

[0092] Step S400: Analyze the logs for a specific time period, compare the noise mean within a predetermined time period before and after the calibration time to obtain the calibration error value, and generate a second correction factor based on the changing trends of several calibration error values.

[0093] Specifically, Figure 3 A flowchart for determining the second correction factor is shown.

[0094] The process of parsing logs for a specific time period, comparing the average noise values ​​within a predetermined time period before and after the calibration time to obtain calibration error values, and generating a second correction factor based on the changing trends of several calibration error values ​​specifically includes the following steps:

[0095] Step S401: Analyze the logs for a specific time period and calculate the mean noise before calibration and the mean noise after calibration within the predetermined time period.

[0096] Step S402: Calculate the calibration error value based on the mean noise value before calibration and the mean noise value after calibration;

[0097] Step S403: Plot the corresponding calibration error values ​​in the logs of several specific time periods into a curve reflecting the trend of calibration error change in chronological order, calculate the average slope of the curve, and use it as the second correction factor.

[0098] In this embodiment of the invention, the calibration error value is used to quantify the change in noise level before and after calibration. It is calculated by subtracting the mean noise level after calibration from the mean noise level before calibration, then dividing by the mean noise level before calibration, and taking the absolute value to measure the effect of calibration on noise suppression or correction.

[0099] The calibration error values ​​in the logs of multiple specific time periods are sorted according to the corresponding calibration event time points, and plotted into a curve of calibration error change trend in chronological order, so that the curve can intuitively reflect the change of calibration error of the noise conditioning device over time.

[0100] The average slope of the trend curve is calculated using a linear regression method. This slope reflects the trend of calibration error over time. This average slope is used as a second correction factor to adjust the initial gain amplitude parameter in order to compensate for the attenuation of calibration effect due to long-term use of the equipment or changes in the environment.

[0101] The significance of using this method to determine the second correction factor lies in the fact that the changing trend of the calibration error reflects the stability of the noise conditioning device's processing of capacitive sensor measurement data at different time points. A single calibration error value may be affected by external environmental factors, equipment status, and other factors, making it impossible to accurately describe the overall variation pattern of the calibration error. Therefore, by analyzing the changing trend of the calibration error over a period of time, it is possible to more stably assess whether the calibration effect of the equipment gradually degrades or fluctuates during long-term operation, thus providing a more reliable basis for parameter adjustment. Furthermore, this method can reduce the impact of random errors on calibration correction, making the calculated second correction factor more stable and providing a more accurate reference for optimizing the gain amplitude parameter.

[0102] Furthermore, the performance testing method for the AI-based visual interactive touchscreen also includes the following steps:

[0103] Step S500: Adjust the initial gain amplitude parameter by combining the first correction factor and the second correction factor.

[0104] Specifically, Figure 4 A flowchart is shown to optimize the initial gain amplitude parameter.

[0105] The adjustment of the initial gain amplitude parameter by combining the first correction factor and the second correction factor specifically includes the following steps:

[0106] Step S501: Call the preset gain amplitude parameter adjustment formula, and dynamically correct the initial gain amplitude parameter by combining the first correction factor and the second correction factor to obtain the optimized gain amplitude parameter.

[0107] Step S502: Apply the optimized gain amplitude parameter to the sensitivity adjustment unit of the capacitive sensor of the display unit under test.

[0108] The formula for adjusting the gain amplitude parameter is: ,in This refers to the optimized gain amplitude parameter. This refers to the initial gain amplitude parameter. This refers to the first correction factor, which is the average slope of the curve reflecting the trend of noise baseline drift. This refers to the adjustment coefficient corresponding to the first correction factor. This refers to the second correction factor, which is the average slope of the curve reflecting the trend of calibration error changes. This refers to the adjustment coefficient corresponding to the second correction factor.

[0109] In the formula for adjusting the gain amplitude parameter ,in This refers to the total number of logs within a specific time period. It refers to the first The midpoint of a specific time period log. This refers to the average of the median times of all logs for a specific time period. Indicates the first The magnitude of the deviation between the historical noise mean and the current test baseline noise mean in the log for a specific time period. This refers to the average deviation value corresponding to all logs for a specific time period.

[0110] ,in It refers to the first Historical noise mean of logs for a specific time period This refers to the current benchmark noise mean;

[0111] ,in It refers to the first The time point of calibration events in a specific time period log. This refers to the average time point of calibration events across all logs for a specific time period. It refers to the first The calibration error value between the mean noise before calibration and the mean noise after calibration in the log for a specific time period. This refers to the average of the calibration error values ​​corresponding to all logs for a specific time period;

[0112] ,in It refers to the first Mean noise before calibration in logs for a specific time period. It refers to the first The mean noise after calibration in the logs for a specific time period.

[0113] In this embodiment of the invention, the significance of dynamically correcting the initial gain amplitude parameter by combining the first correction factor and the second correction factor lies in the fact that these two correction factors respectively reflect the degree of noise baseline drift and the changing trend of calibration error, thereby enabling a comprehensive evaluation of the processing effect of the noise conditioning device on the capacitive sensor measurement data, as well as the impact of noise on the measurement data. Unlike existing technologies, the core idea of ​​this invention is that when the noise suppression effect of the noise conditioning device weakens—that is, when the first correction factor indicates a significant drift in the noise baseline and the second correction factor shows an increase in calibration error—it can be reasonably inferred that the anomaly in the capacitive sensor measurement data is related to the noise effect. At this time, the adjustment of the gain amplitude parameter should not solely rely on the fixed gain strategy in existing technologies; otherwise, the noise amplification problem may be overlooked, leading to deviations in the measurement data. This invention adaptively adjusts the gain amplitude parameter by combining it with the processing capability of the noise conditioning device, appropriately reducing the gain amplitude parameter when the noise effect is significant, avoiding excessive amplification of noise interference by the system, and improving the stability and accuracy of signal processing.

[0114] The first and second correction factors work together to form a mechanism that dynamically adjusts the gain amplitude parameter based on the processing capability of the noise conditioning device, making gain adjustment more intelligent and adaptive. For example, a smaller first correction factor indicates a more stable noise baseline, while a smaller second correction factor indicates a lower calibration error in the noise conditioning device. This means that the current capacitive sensor measurement data is less affected by noise, and the gain amplitude parameter can be maintained at a higher level to ensure sensitivity and signal response speed. However, if both the first and second correction factors are large, it indicates a decrease in the noise suppression capability of the noise conditioning device, and the current noise impact is more severe. In this case, the system should appropriately reduce the gain amplitude parameter to reduce noise interference in the signal, making the measurement data more stable and avoiding false triggering or drift problems caused by signal amplification.

[0115] Besides the sensitivity adjustment unit of the capacitive sensor used in the display unit under test, the dynamic adjustment of this gain amplitude parameter can also be applied to other systems involving noise conditioning devices. For example, in the bioelectrical signal detection of high-precision medical equipment, it can be used to optimize the gain adjustment of electrocardiogram or electroencephalogram signals to adapt to the physiological characteristics of different patients and environmental noise levels; in industrial automation systems, it can be used for sensor signal processing optimization to ensure that the equipment can still accurately detect changes in object position or pressure in high-noise environments; in avionics systems, it can be used for data optimization of flight control sensors to maintain accurate measurement capabilities under different flight states and interference environments, while the role of the noise conditioning device is to reduce environmental noise interference in different application scenarios, so that the gain adjustment can play an effective role.

[0116] For example, in practical applications, suppose a smart touchscreen device operates in a factory environment. Due to long-term exposure to high electromagnetic interference, users find that touch operation becomes unstable, occasionally resulting in accidental touches or slow response. System detection reveals that the first correction factor indicates a significant noise baseline drift, suggesting a decrease in the noise conditioning device's suppression capability after prolonged use. The second correction factor indicates a continuously increasing calibration error, meaning that even after multiple calibrations, the noise conditioning device cannot fully compensate for noise interference. Existing technologies might further increase the gain amplitude parameter to enhance sensor sensitivity, but this would amplify noise further, affecting the accuracy of measurement data. The solution of this invention, based on a comprehensive analysis of the two correction factors, identifies the correlation between this anomaly and the decreased processing capability of the noise conditioning device, and appropriately reduces the gain amplitude parameter, making the touch signal more stable and reducing the impact of noise on measurement data. After the optimized gain amplitude parameter is applied to the sensitivity adjustment unit, the touchscreen returns to stability, accidental touches decrease, and device operation becomes more reliable.

[0117] This invention solves the problem of deviation caused by relying solely on gain amplitude compensation in the prior art while ignoring the influence of noise amplification. It makes gain adjustment no longer a fixed strategy, but can dynamically adapt to the processing capability of the noise conditioning device, ensuring the accuracy and reliability of capacitive sensor measurement data, thereby improving the long-term stability and adaptability of the equipment.

[0118] Furthermore, Figure 5 An application architecture diagram of the system provided in an embodiment of the present invention is shown.

[0119] In another preferred embodiment of the present invention, a performance testing system for a visual interactive touchscreen based on AI technology includes:

[0120] The data acquisition module 100 is used to determine the current touch working mode if the acquired capacitive sensor measurement data is abnormal after multi-touch detection of the display unit under test, and to acquire the initial gain amplitude parameter prepared for the capacitive sensor and the historical processing record of the noise conditioning device used to process the capacitive sensor measurement data.

[0121] In this embodiment of the invention, the display unit under test refers to a display device that undergoes multi-touch detection, and this invention is particularly aimed at visual interactive touchscreens. The display unit under test includes a capacitive touch sensor for sensing touch operations and converting them into electrical signals for subsequent processing and analysis. Multi-touch detection refers to the simultaneous detection of multiple touch points on the display unit, and is commonly used in smart devices, industrial control panels, medical touchscreens, and other scenarios to detect the accuracy of touch response, gesture recognition capabilities, and device stability.

[0122] Capacitive sensor measurement data refers to the electrical signal output by the sensor when detecting touch events, including capacitance value, noise level, response time, and other data. When the measured data deviates from the expected range, such as abnormal fluctuations in capacitance value, loss of touch signal, or accidental touch, it indicates that the touch sensor may be malfunctioning. Specific anomaly determination can be based on historical data comparison, threshold detection, or AI anomaly recognition methods.

[0123] The current touch operation mode refers to the working state of the touch sensor under different application scenarios or user operating habits, including touch sensitivity settings, signal sampling mode, and input signal filtering method. Different touch modes may affect the way touch data is collected. Therefore, when detecting anomalies, it is necessary to determine the current touch operation mode so that subsequent corrective measures can be more targeted.

[0124] The initial gain amplitude parameter for this capacitive sensor refers to the signal gain adjustment parameter set for the sensor under default or preset conditions, used to optimize the acquisition accuracy and noise suppression capability of the touch signal. This parameter can be obtained through experimental data, sensor calibration, or automatic system adjustment. In the prior art, there are already methods based on automatic gain control (AGC) to dynamically adjust the sensor gain.

[0125] Noise conditioning devices used for processing capacitive sensor measurement data refer to hardware or software modules that process touch signals to reduce noise interference and improve signal accuracy. Their historical processing records refer to the conditioning processes and results performed on the capacitive sensor measurement data by the device at different times and under different test conditions. These records can be obtained from device logs, noise filtering algorithm logs, or the data storage of the sensor signal processing unit.

[0126] Historical processing records should include at least the following: First, each record should indicate the type of the display unit under test to ensure that the comparison is made with the same or similar types of devices during subsequent analysis; second, it should include the processing results of the noise conditioning device in different touch operation modes to compare the noise characteristics under different modes; in addition, it should record the specific time of the calibration event and store the noise data before and after calibration to analyze the impact of calibration on noise; finally, it should record the specific time of each measurement to facilitate subsequent trend analysis in chronological order.

[0127] Furthermore, the performance testing system for the AI-based visual interactive touchscreen also includes:

[0128] The historical record filtering module 200 is used to parse historical processing records and filter out a predetermined number of logs for specific time periods where the processed display unit is of the same type as the tested display unit, the capacitive sensor measurement data is normal, the time interval is consistent, and there is a calibration event.

[0129] In this embodiment of the invention, selecting logs from specific time periods is significant in ensuring that the historical data used to calculate the correction factor is of high quality and comparable. By filtering logs that meet specific conditions, interference from irrelevant factors can be reduced, the impact of abnormal data can be minimized, and the accuracy of the correction factor calculation can be ensured. This guarantees that when adjusting the gain amplitude parameter, the data used can truly reflect the changing trends of noise baseline drift and calibration error, making the final optimized gain amplitude parameter more accurate and reliable.

[0130] Several factors need to be considered when filtering logs, primarily to ensure data stability and representativeness. First, logs consistent with the type of display unit being tested should be selected to avoid data mismatches caused by differences in device characteristics. Second, capacitive sensor measurements should be anomalies-free to eliminate data deviations caused by equipment malfunctions or environmental interference. Third, consistent time intervals should be ensured to guarantee uniform data distribution and prevent trend analysis errors due to uneven data density. Finally, each log should include one calibration event. This is to calculate calibration error and generate a second correction factor, ensuring that the optimized gain amplitude parameter adapts to performance changes in different calibration states. Combining these filtering criteria ensures high consistency of the data used for analysis, making the calculated correction factor more valuable.

[0131] Furthermore, the performance testing system for the AI-based visual interactive touchscreen also includes:

[0132] The first correction factor determination module 300 is used to determine the current test reference noise mean value of the noise conditioning device processing the capacitive sensor measurement data of the display unit under test in the current touch working mode according to the preset reference model, determine the historical noise mean value of the log for each specific time period, and generate the first correction factor according to the changing trend of the deviation of each historical noise mean value relative to the current test reference noise mean value.

[0133] The preset reference model refers to the average baseline noise value obtained under standard operating conditions when the noise conditioning device processes capacitive sensor data measured by different types of display units in each touch operation mode.

[0134] Specifically, Figure 6 The diagram shows a structural block diagram of the first correction factor determination module 300 in the system provided by an embodiment of the present invention.

[0135] In a preferred embodiment provided by the present invention, the first correction factor determination module 300 specifically includes:

[0136] The reference noise mean determination unit 301 is used to retrieve a preset reference model, determine the specific type of the display unit under test, and use the specific type and the current touch working mode as input to determine the corresponding current test reference noise mean.

[0137] The historical noise mean calculation unit 302 is used to parse the logs of the selected specific time period and calculate the historical noise mean corresponding to each specific time period log.

[0138] The first correction factor calculation unit 303 is used to quantify the deviation magnitude of each historical noise mean relative to the current test reference noise mean, plot these deviation magnitude values ​​in chronological order as a curve reflecting the trend of noise baseline drift, calculate the average slope of the curve, and use it as the first correction factor.

[0139] In this embodiment of the invention, the preset reference model is used to provide the baseline noise mean of the capacitive sensor measurement data by the noise conditioning device under different touch operating modes. The model is established based on a large amount of experimental data, covering different types of display units and their noise characteristics under standard operating conditions, and is optimized by combining existing noise analysis and filtering techniques. In the prior art, the noise characteristics of touch systems can be fitted through experimental measurement, statistical modeling, or machine learning methods; therefore, this preset reference model is feasible and can be constructed based on existing technologies.

[0140] The process of building this model typically includes the following aspects:

[0141] First, standardized tests are conducted in a controlled environment for different types of display units, recording noise data under different touch operation modes, including background noise, capacitive sensor response data, and the processing results of the noise conditioning device. Second, based on the collected data, mathematical models are established, such as those based on multivariate regression, probability distribution modeling, or neural network fitting, to analyze the variation law of the noise mean under different conditions. Finally, through data verification and model optimization, the preset reference model can accurately reflect the noise level under standard operating conditions, providing an accurate reference for subsequent noise deviation analysis.

[0142] The specific working process of the first correction factor calculation unit 303 is as follows: First, extract the historical noise mean from the log of a specific time period, and calculate the deviation of each historical noise mean relative to the current test reference noise mean. The deviation is calculated by subtracting the current test reference noise mean from the historical noise mean, then dividing by the current test reference noise mean, and taking the absolute value to measure the degree of deviation of the historical noise mean from the standard value.

[0143] Secondly, all calculated deviation amplitude values ​​are sorted according to the time points corresponding to the logs of specific time periods, and plotted in chronological order as a curve of the noise baseline drift trend. This curve can intuitively reflect the change law of noise deviation over time, so as to evaluate the long-term stability of the noise conditioning device on the measurement data of the capacitive sensor.

[0144] Finally, the curve is analyzed using linear regression to calculate the average slope of the trend change. The average slope reflects the degree of noise baseline drift. This slope is used as the first correction factor to adjust the initial gain amplitude parameter so that the adjusted gain amplitude parameter can compensate for the measurement error caused by noise baseline drift.

[0145] The significance of this method lies in the fact that noise baseline drift is a dynamic process that changes over time. Data from a single measurement may be affected by random factors and cannot accurately describe the overall trend. Therefore, by analyzing historical noise deviation changes over a period of time, the performance fluctuations of the noise conditioning device can be assessed more stably, thereby improving the correction effect on capacitive sensor measurement data. Furthermore, this method can effectively reduce the impact of single-point outliers on the correction factor, making the calculated first correction factor more reliable and providing a more valuable reference for the subsequent optimization of gain amplitude parameters.

[0146] Furthermore, the performance testing system for the AI-based visual interactive touchscreen also includes:

[0147] The second correction factor determination module 400 is used to parse the logs for a specific time period, compare the noise mean within a predetermined time period before and after the calibration time to obtain the calibration error value, and generate a second correction factor based on the changing trend of several calibration error values.

[0148] Specifically, Figure 7 The diagram shows a structural block diagram of the second correction factor determination module 400 in the system provided by an embodiment of the present invention.

[0149] In a preferred embodiment of the present invention, the second correction factor determination module 400 specifically includes:

[0150] The noise mean calculation unit 401 before and after calibration is used to parse logs for a specific time period and calculate the noise mean before calibration and the noise mean after calibration within a predetermined time period.

[0151] The calibration error value determination unit 402 is used to calculate the calibration error value based on the average noise value before calibration and the average noise value after calibration.

[0152] The second correction factor calculation unit 403 is used to plot the corresponding calibration error values ​​in the logs of several specific time periods into a curve reflecting the trend of calibration error change in chronological order, calculate the average slope of the curve, and use it as the second correction factor.

[0153] In this embodiment of the invention, the calibration error value is used to quantify the change in noise level before and after calibration. It is calculated by subtracting the mean noise level after calibration from the mean noise level before calibration, then dividing by the mean noise level before calibration, and taking the absolute value to measure the effect of calibration on noise suppression or correction.

[0154] The calibration error values ​​in the logs of multiple specific time periods are sorted according to the corresponding calibration event time points, and plotted into a curve of calibration error change trend in chronological order, so that the curve can intuitively reflect the change of calibration error of the noise conditioning device over time.

[0155] The average slope of the trend curve is calculated using a linear regression method. This slope reflects the trend of calibration error over time. This average slope is used as a second correction factor to adjust the initial gain amplitude parameter in order to compensate for the attenuation of calibration effect due to long-term use of the equipment or changes in the environment.

[0156] The significance of using this method to determine the second correction factor lies in the fact that the changing trend of the calibration error reflects the stability of the noise conditioning device's processing of capacitive sensor measurement data at different time points. A single calibration error value may be affected by external environmental factors, equipment status, and other factors, making it impossible to accurately describe the overall variation pattern of the calibration error. Therefore, by analyzing the changing trend of the calibration error over a period of time, it is possible to more stably assess whether the calibration effect of the equipment gradually degrades or fluctuates during long-term operation, thus providing a more reliable basis for parameter adjustment. Furthermore, this method can reduce the impact of random errors on calibration correction, making the calculated second correction factor more stable and providing a more accurate reference for optimizing the gain amplitude parameter.

[0157] Furthermore, the performance testing system for the AI-based visual interactive touchscreen also includes:

[0158] The gain amplitude parameter optimization module 500 is used to adjust the initial gain amplitude parameter by combining the first correction factor and the second correction factor.

[0159] Specifically, Figure 8 The diagram shows a structural block diagram of the gain amplitude parameter optimization module 500 in the system provided in an embodiment of the present invention.

[0160] In a preferred embodiment of the present invention, the gain amplitude parameter optimization module 500 specifically includes:

[0161] The initial gain amplitude parameter optimization unit 501 is used to call the preset gain amplitude parameter adjustment formula, and dynamically correct the initial gain amplitude parameter by combining the first correction factor and the second correction factor to obtain the optimized gain amplitude parameter.

[0162] The optimized gain amplitude parameter application unit 502 is used to apply the optimized gain amplitude parameter to the sensitivity adjustment unit of the capacitive sensor of the display unit under test.

[0163] The formula for adjusting the gain amplitude parameter is: ,in This refers to the optimized gain amplitude parameter. This refers to the initial gain amplitude parameter. This refers to the first correction factor, which is the average slope of the curve reflecting the trend of noise baseline drift. This refers to the adjustment coefficient corresponding to the first correction factor. This refers to the second correction factor, which is the average slope of the curve reflecting the trend of calibration error changes. This refers to the adjustment coefficient corresponding to the second correction factor.

[0164] In the formula for adjusting the gain amplitude parameter ,in This refers to the total number of logs within a specific time period. It refers to the first The midpoint of a specific time period log. This refers to the average of the median times of all logs for a specific time period. Indicates the first The magnitude of the deviation between the historical noise mean and the current test baseline noise mean in the log for a specific time period. This refers to the average deviation magnitude values ​​corresponding to all logs for a specific time period;

[0165] ,in It refers to the first Historical noise mean of logs for a specific time period This refers to the current benchmark noise mean;

[0166] ,in It refers to the first The time point of calibration events in a specific time period log. This refers to the average time point of calibration events across all logs for a specific time period. It refers to the first The calibration error value between the mean noise before calibration and the mean noise after calibration in the log for a specific time period. This refers to the average of the calibration error values ​​corresponding to all logs for a specific time period;

[0167] ,in It refers to the first Mean noise before calibration in logs for a specific time period. It refers to the first The mean noise after calibration in the logs for a specific time period.

[0168] It should be understood that although the steps in the flowcharts of the various embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise expressly stated herein, there is no strict order restriction on the execution of these steps, and these steps may be executed in other orders.

Claims

1.A performance detection method of an AI technology-based visual interactive touch screen, the method comprising: determining a first performance of the visual interactive touch screen; determining a second performance of the visual interactive touch screen; and comparing the first performance with the second performance. The method comprises: After multi-point touch detection is performed on the display unit under test, if the obtained capacitive sensor measurement data is abnormal, the current touch working mode is determined, and an initial gain amplitude parameter prepared for the capacitive sensor and a historical processing record of a noise conditioning device used for processing the capacitive sensor measurement data are obtained; The historical processing record is analyzed, and a predetermined number of specific period logs are screened out, which have the same type of display unit as the display unit under test, no abnormal capacitive sensor measurement data, the same time interval, and a calibration event; A current test reference noise mean value of the noise conditioning device for processing the capacitive sensor measurement data of the display unit under test in the current touch working mode is determined according to a preset reference model, a historical noise mean value of each specific period log is determined, and a first correction factor is generated according to a change trend of a deviation amplitude of each historical noise mean value relative to the current test reference noise mean value; The specific period logs are analyzed, a calibration error value is obtained by comparing noise mean values within a predetermined time period before and after a calibration time, and a second correction factor is generated according to a change trend of a plurality of calibration error values; The initial gain amplitude parameter is adjusted by comprehensively considering the first correction factor and the second correction factor. 2.The AI technology-based performance detection method of a visual interactive touch screen according to claim 1, wherein, The preset reference model refers to a reference noise mean value obtained in each touch working mode when the noise conditioning device processes capacitive sensor data measured on different types of display units under standard working conditions. 3.The AI technology-based performance detection method of a visual interactive touch screen according to claim 2, characterized in that, The step of determining a current test reference noise mean value of the noise conditioning device for processing the capacitive sensor measurement data of the display unit under test in the current touch working mode according to a preset reference model, determining a historical noise mean value of each specific period log, and generating a first correction factor according to a change trend of a deviation amplitude of each historical noise mean value relative to the current test reference noise mean value comprises: The preset reference model is called, the specific type of the display unit under test is determined, and the specific type and the current touch working mode are taken as inputs to determine the corresponding current test reference noise mean value; The selected specific period logs are analyzed, and the historical noise mean value corresponding to each specific period log is calculated; The deviation amplitude of each historical noise mean value relative to the current test reference noise mean value is quantified, and these deviation amplitude values are plotted in time sequence to form a curve reflecting the change trend of the noise baseline drift, the average slope of the curve is calculated, and the average slope is taken as the first correction factor. 4.The AI technology-based performance detection method of a visual interactive touch screen according to claim 3, wherein, The step of analyzing the specific period logs, comparing the noise mean values within a predetermined time period before and after a calibration time to obtain a calibration error value, and generating a second correction factor according to a change trend of a plurality of calibration error values comprises: The specific period logs are analyzed, and the noise mean values before and after calibration within a predetermined time period are calculated respectively; The calibration error value is calculated according to the noise mean values before and after calibration; A plurality of specific period logs corresponding to the calibration error values are plotted in time sequence to form a curve reflecting the change trend of the calibration error, the average slope of the curve is calculated, and the average slope is taken as the second correction factor. 5.The AI technology-based performance detection method of a visual interactive touch screen according to claim 4, characterized in that, The step of adjusting the initial gain amplitude parameter by comprehensively considering the first correction factor and the second correction factor comprises: The preset gain amplitude parameter adjustment formula is called to dynamically correct the initial gain amplitude parameter by combining the first correction factor and the second correction factor, so as to obtain an optimized gain amplitude parameter; The optimized gain amplitude parameter is applied to a sensitivity adjustment unit of the capacitive sensor of the display unit under test. 6.The AI technology-based performance detection method of a visual interaction touch screen according to claim 5, wherein, The gain amplitude parameter adjustment formula is: Wherein Refers to the optimized gain amplitude parameter, Refers to the initial gain amplitude parameter, Refers to the first correction factor, that is, the average slope of the curve reflecting the noise baseline drift change trend, Refers to the adjustment coefficient corresponding to the first correction factor, Refers to the second correction factor, that is, the average slope of the curve reflecting the calibration error change trend, Refers to the adjustment coefficient corresponding to the second correction factor. 7.A performance detection system of an AI technology-based visual interactive touch screen, characterized by, The system comprises a data acquisition module, a historical record screening module, a first correction factor determination module, a second correction factor determination module, and a gain amplitude parameter optimization module, wherein: The data acquisition module is configured to, after multi-point touch detection of the display unit under test, determine a current touch working mode if the acquired capacitive sensor measurement data is abnormal, and acquire an initial gain amplitude parameter prepared for the capacitive sensor and historical processing records of a noise conditioning device used for processing capacitive sensor measurement data; The historical record screening module is configured to analyze the historical processing records, and screen out a predetermined number of specific period logs whose display units processed are consistent with the type of the display unit under test, whose capacitive sensor measurement data are normal, whose time intervals are consistent, and which have a calibration event; The first correction factor determination module is configured to determine a current test reference noise mean value of the noise conditioning device for processing capacitive sensor measurement data of the display unit under test in the current touch working mode according to a preset reference model, determine a historical noise mean value of each specific period log, and generate a first correction factor according to a change trend of a deviation amplitude of each historical noise mean value relative to the current test reference noise mean value; The preset reference model refers to reference noise mean values obtained in each touch working mode when the noise conditioning device processes capacitive sensor data measured by different types of display units under standard working conditions; The second correction factor determination module is configured to analyze the specific period logs, compare noise mean values within a predetermined time period before and after a calibration time to obtain a calibration error value, and generate a second correction factor according to a change trend of a plurality of calibration error values; The gain amplitude parameter optimization module is configured to adjust the initial gain amplitude parameter by comprehensively considering the first correction factor and the second correction factor. 8.The AI technology-based visual interactive touch screen performance detection system of claim 7, wherein, The first correction factor determination module specifically comprises: A reference noise mean value determination unit configured to call the preset reference model, determine a specific type of the display unit under test, and take the specific type and the current touch working mode as inputs to determine a corresponding current test reference noise mean value; A historical noise mean value calculation unit configured to analyze the selected specific period logs and calculate a historical noise mean value corresponding to each specific period log; A first correction factor calculation unit configured to quantify a deviation amplitude of each historical noise mean value relative to the current test reference noise mean value, draw a curve reflecting a change trend of noise baseline drift according to the deviation amplitude values in time sequence, calculate an average slope of the curve, and take the average slope as the first correction factor. 9.The AI technology-based visual interactive touch screen performance detection system of claim 8, wherein, The second correction factor determination module specifically comprises: A calibration before and after noise mean value calculation unit configured to analyze the specific period logs and calculate a calibration before noise mean value and a calibration after noise mean value within a predetermined time period, respectively; A calibration error value calculation unit configured to calculate a calibration error value of each specific period log by comparing the calibration before noise mean value and the calibration after noise mean value of the specific period log, and generate the second correction factor according to a change trend of a plurality of calibration error values. The calibration error value determination unit is configured to calculate a calibration error value according to the noise mean value before calibration and the noise mean value after calibration; The second correction factor calculation unit is configured to draw a curve reflecting the change trend of the calibration error according to the corresponding calibration error values in the log of the specific time period in time sequence, calculate the average slope of the curve, and take the average slope as the second correction factor. 10.The AI technology-based visual interactive touch screen performance detection system of claim 9, wherein, The gain amplitude parameter optimization module specifically comprises: The initial gain amplitude parameter optimization unit is configured to call a preset gain amplitude parameter adjustment formula, dynamically correct the initial gain amplitude parameter by combining the first correction factor and the second correction factor, and obtain an optimized gain amplitude parameter. The optimized gain amplitude parameter application unit is configured to apply the optimized gain amplitude parameter to the sensitivity adjustment unit of the capacitive sensor of the display unit under test. The gain amplitude parameter adjustment formula is: Wherein Refers to the optimized gain amplitude parameter, Refers to the initial gain amplitude parameter, Refers to the first correction factor, that is, the average slope of the curve reflecting the noise baseline drift change trend, Refers to the adjustment coefficient corresponding to the first correction factor, Refers to the second correction factor, that is, the average slope of the curve reflecting the calibration error change trend, Refers to the adjustment coefficient corresponding to the second correction factor.

Images