A method for performance optimization of a privacy film manufacturing process

Abstract

The present application relates to the technical field of intelligent manufacturing, and is a performance optimization method for a privacy protection film manufacturing process, comprising: collecting real-time optical parameters of a micro-prism array on the surface of the privacy protection film and obtaining a current alignment phase deviation value through convolutional neural network processing; using the optimized alignment phase to simulate and calculate the light transmission path to obtain controllable privacy shielding strength; using the privacy shielding strength in combination with front surface brightness data to perform iterative optimization using a genetic algorithm to obtain high privacy shielding configuration with low brightness loss; using the array layout for suppressing optical interference to perform dynamic verification on the privacy angle to obtain a final set of controllable adjustment parameters; and using the final set of controllable adjustment parameters to integrate the light transmission model and the privacy shielding strength to determine the overall performance indicators of the privacy protection film and output an adjustment instruction when the preset business target is met.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing technology, and in particular to a performance optimization method for the manufacturing process of privacy films. Background Technology

[0002] Privacy screen protectors, as an important optical product for protecting screen privacy, are increasingly widely used in electronic devices such as mobile phones and computers. Their core function is to limit the side viewing angle by controlling the direction of light propagation, thereby preventing information leakage. The development of this field is directly related to the improvement of user privacy and display experience, and has key market value and practical significance.

[0003] Existing privacy screen protectors mainly rely on periodic microstructures to achieve directional light blocking. However, in pursuit of a narrower privacy angle, these structures often result in a significant decrease in the transmittance of light from the front, and the brightness is noticeably reduced when the user looks directly at the screen. At the same time, the periodic arrangement can easily cause optical interference with the screen pixel grid, forming colored stripes that affect visual comfort. In addition, the production of high-precision molds is complex and it is difficult to achieve wide-width continuous manufacturing, resulting in high overall costs.

[0004] In optical design, the introduction of composite microprism arrays is intended to improve performance. However, the relative alignment phase between the upper and lower layers is difficult to control precisely, which means that the effective privacy angle cannot be flexibly changed after production and is fixed once formed. This sensitivity of the alignment phase further exacerbates the occurrence of moiré interference, because tiny misalignments will amplify the interference effect between the structural period and the screen pixels. For example, in the actual screen protector application process, if the alignment phase deviation is only a few micrometers, the side privacy protection strength will be unexpectedly weakened, and obvious rainbow stripes will appear on the front, which seriously affects the adaptability of the product in different usage scenarios.

[0005] Therefore, how to achieve controllable adjustment of the privacy angle and effectively suppress optical interference while maintaining low brightness loss and high privacy shielding has become a key issue in improving the practicality and market competitiveness of privacy films. Summary of the Invention

[0006] This invention provides a performance optimization method for the manufacturing process of privacy films, mainly including:

[0007] Real-time optical parameters of the microprism array on the surface of the privacy film are collected and processed by a convolutional neural network to obtain the current alignment phase deviation value. Based on the current alignment phase deviation value, the matching relationship between the microstructure period and the screen pixel grid is obtained. When the deviation exceeds a preset threshold, the relative position of the microprism array is adjusted to obtain an optimized alignment phase. The optimized alignment phase is used to simulate and calculate the light transmission path to obtain a controllable privacy shielding intensity. The privacy shielding intensity is combined with front brightness data and iteratively optimized using a genetic algorithm to obtain a high privacy shielding configuration with low brightness loss. Based on the high privacy shielding configuration, potential points of moiré interference are obtained. When the number of potential points exceeds a preset threshold, the arrangement of the microstructure period is modified to obtain an array layout that suppresses optical interference. The array layout that suppresses optical interference is used to dynamically verify the privacy angle to obtain the final controllable adjustment parameter set. The final controllable adjustment parameter set is used to integrate the light transmission model and the privacy shielding intensity to determine the overall performance index of the privacy film, and an adjustment command is output when the preset business objectives are met. Furthermore, the step of acquiring real-time optical parameters of the microprism array on the privacy film surface and processing them through a convolutional neural network to obtain the current alignment phase deviation value includes: acquiring real-time optical parameters of the microprism array on the privacy film surface through an optical sensor to obtain a parameter data set; inputting the parameter data set into a convolutional neural network for feature extraction processing to obtain the current alignment phase deviation value; if the current alignment phase deviation value exceeds a preset threshold, obtaining surface structure analysis results through surface scanning to determine the source of the deviation; extracting a calibration mechanism from the neural network training data to obtain adjustment parameters based on the source of the deviation; updating the microprism array structure through the adjustment parameters and then inputting the parameter data set back into the convolutional neural network for iterative processing to obtain the updated current alignment phase deviation value. Furthermore, the step of obtaining the matching relationship between the microstructure period and the screen pixel grid based on the current alignment phase deviation value and adjusting the relative position of the microprism array to obtain the optimized alignment phase when the matching relationship deviation exceeds a preset threshold includes: determining the matching relationship between the microstructure period and the screen pixel grid based on the current alignment phase deviation value to obtain the matching relationship deviation; determining whether the matching relationship deviation exceeds a preset threshold to obtain the deviation exceeding condition; if the deviation exceeding condition is established, adjusting the relative position of the microprism array to obtain the position adjustment result; obtaining the light field distribution characteristics based on the position adjustment result to determine the degree of light field distribution optimization; and determining the re-matching relationship between the array and the grid based on the degree of light field distribution optimization to obtain the optimized alignment phase.Furthermore, the step of using the optimized alignment phase to simulate and calculate the light transmission path to obtain the controllable privacy shielding intensity includes: simulating and calculating the light transmission path based on the optimized alignment phase to obtain distribution data of refraction angle and propagation direction; determining the distribution characteristics of the light transmission path based on the distribution data; obtaining the limitation range of the side viewing angle based on the distribution characteristics; determining the boundary value of the limitation range by comparing the angular deviation within the limitation range; adjusting the dynamic parameters of privacy shielding based on the boundary value and fusing a brightness threshold and a contrast threshold to obtain the adjustment basis for the shielding intensity; and fusing the matching relationship of the light field distribution based on the adjustment basis to obtain the controllable privacy shielding intensity. Furthermore, the step of obtaining a high privacy masking configuration with low brightness loss by iteratively optimizing the privacy masking strength and front brightness data using a genetic algorithm includes: determining initial balance parameters based on the privacy masking strength and front brightness data; using a genetic algorithm to iteratively process the balance relationship between masking strength and brightness loss with low brightness loss as a constraint to obtain an iterative optimization sequence; evaluating the brightness loss based on the iterative optimization sequence and fusing the evaluation results to obtain a configuration parameter adjustment scheme; and outputting a high privacy masking configuration if the configuration parameter adjustment scheme meets preset conditions. Furthermore, the step of obtaining potential moiré interference points based on the high privacy masking configuration and modifying the microstructure periodic arrangement to obtain an array layout that suppresses optical interference when the number of potential points exceeds a preset threshold includes: obtaining the number of potential moiré interference points by scanning an optical pattern based on the high privacy masking configuration; determining whether the number of potential points exceeds a preset threshold; if it does, determining a microstructure periodic modification scheme based on the optical occurrence threshold; periodically adjusting the interference potential points according to the microstructure periodic modification scheme to obtain suppression optical arrangement parameters; and performing hierarchical integration adjustment based on the suppression optical arrangement parameters to obtain an array layout that suppresses optical interference. Furthermore, the method of dynamically verifying the privacy angle using the array layout for suppressing optical interference to obtain the final controllable set of adjustable parameters includes: scanning the display layer to obtain the distribution of moiré pattern occurrence points based on the array layout for suppressing optical interference; determining a microstructure periodic adjustment scheme by comparing parameters based on the distribution of moiré pattern occurrence points through the matching relationship between interference intensity and period; performing angle simulation on the privacy angle using a dynamic verification process based on the microstructure periodic adjustment scheme to obtain angle change tracking results; determining whether the adjustment range covers the specified scenario requirements based on the angle change tracking results; and if it covers the requirements, fusing the suppression optical parameters and superimposing the parameters to obtain the final controllable set of adjustable parameters.Furthermore, the step of integrating the light transmission model and privacy shielding intensity with the final controllable adjustable parameter set to determine the overall performance index of the privacy film and outputting an adjustment command when the preset business objectives are met includes: integrating the final controllable adjustable parameter set with the light transmission model and privacy shielding intensity to obtain the overall performance index; determining the angle-dependent performance evaluation value based on the overall performance index; judging whether the index meets the business objectives by fusing the shielding dynamic response based on the angle-dependent performance evaluation value; and outputting an adjustment command based on the judgment result if the business objectives are met. Furthermore, the step of collecting real-time optical parameters of the microprism array on the privacy film surface and processing them through a convolutional neural network to obtain the current alignment phase deviation value includes: acquiring a parameter data set containing the refractive characteristics and light intensity distribution of the microprisms in real time through an optical sensor; inputting the parameter data set into a pre-trained convolutional neural network model for multi-layer feature extraction and mapping; outputting the current alignment phase deviation value characterizing the degree of phase misalignment between the microprism array and the pixel grid by the convolutional neural network model; and triggering subsequent surface structure deviation source analysis and structure calibration adjustment processes when the current alignment phase deviation value exceeds a preset range. Furthermore, the step of using a genetic algorithm to iteratively optimize the privacy occlusion configuration with low brightness loss by combining the privacy occlusion strength with the frontal brightness data includes: using the privacy occlusion strength as the main component of the fitness function of the genetic algorithm; converting the frontal brightness data into a brightness loss penalty term and constructing a multi-objective optimization function together with the privacy occlusion strength; performing selection, crossover, and mutation processing on the structural control parameters of the microprism array through multi-generation genetic iterative operations; and outputting the corresponding high privacy occlusion configuration with low brightness loss when the genetic algorithm converges to a solution that satisfies the brightness loss constraint and maximizes the privacy occlusion strength.

[0008] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0009] This invention discloses a performance optimization method for the manufacturing process of privacy films, addressing the unique business scenario problem of balancing privacy protection and visual effect. Specifically, it addresses how to achieve high privacy shielding while ensuring low loss of frontal brightness, effectively suppressing optical interference such as moiré patterns, and meeting the dynamic adjustment requirements of different scenarios. This invention acquires optical parameters of a microprism array in real time, uses a convolutional neural network to calculate the alignment phase deviation, adjusts the microstructure period and pixel grid matching relationship, optimizes the light transmission path to control the side viewing angle, and combines a genetic algorithm to balance brightness loss and shielding intensity. Furthermore, it suppresses interference by modifying the array layout, and finally integrates the parameter set to determine the overall performance indicators and output adjustment commands. This invention achieves controllable adjustment and efficient privacy protection of privacy films in multiple scenarios, significantly improving optical performance and user experience. Attached Figure Description

[0010] Figure 1 This is a flowchart of a performance optimization method for the manufacturing process of a privacy film according to the present invention. Detailed Implementation

[0011] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0012] like Figure 1 The performance optimization method for the privacy film manufacturing process in this embodiment may specifically include:

[0013] Step S101: By collecting the real-time optical parameters of the microprism array on the surface of the privacy film, the collected parameter data is processed by a convolutional neural network to obtain the current alignment phase deviation value.

[0014] Real-time optical parameters of the microprism array on the surface of the privacy film are acquired using an optical sensor to obtain a parameter data set. A convolutional neural network is used to process this parameter data set to determine the alignment phase deviation value. If the alignment phase deviation value exceeds a preset threshold, surface structure analysis results are obtained through surface scanning to determine the source of the deviation. For the source of the deviation, a calibration mechanism is extracted from the neural network training data to obtain adjustment parameters. The microprism array structure is updated using these adjustment parameters to obtain the current alignment phase deviation value. The parameter data set acquired by the optical sensor is input into the convolutional neural network for processing. The network extracts features from the real-time data processing to determine the alignment phase deviation value. If the alignment phase deviation value exceeds the threshold, surface structure analysis is combined to obtain deviation source data. This source data is used for neural network training and adjustment, extracting the phase calibration mechanism to generate adjustment parameters. After obtaining the adjustment parameters, the microprism array structure in the privacy film is updated, and the parameter data input is iteratively processed to obtain the deviation value output. To ensure the real-time nature of the optical parameter acquisition, the output is fed back to the sensor to obtain the current alignment phase deviation value. These are fused to obtain the current alignment phase deviation value.

[0015] In one embodiment, real-time optical parameter acquisition of the microprism array on the privacy film surface is achieved through a dedicated optical sensor. The microprism array refers to the tiny prism structures arranged on the privacy film, used to control the direction of light propagation to limit the side-viewing angle and ensure privacy protection.

[0016] Specifically, the array consists of multiple micrometer-scale prism units, each with a specific tilt angle and spacing. On the production line, sensors such as CCD cameras or laser scanners capture optical signals from the array surface in real time, including light reflection intensity, refraction angle, and phase difference. These parameters reflect the alignment of the array; for example, reflection intensity indicates the flatness of the prism surface. Continuous acquisition forms time-series data for subsequent processing. This approach is suitable for online inspection scenarios of display privacy films, ensuring precision control of the manufacturing process. Based on the above acquisition steps, a convolutional neural network is further used to process the parameter data. A convolutional neural network is a deep learning model that processes image or sequence data through convolutional layers for feature extraction, pooling layers for dimensionality reduction, and fully connected layers for classification. In this embodiment, the network input is two-dimensional array data of optical parameters, such as mapping reflection intensity and refraction angle to an image matrix. The network first scans the data through multiple convolutional kernels to extract local features such as edges and textures, which correspond to the geometric deviations of the microprism array. Then, pooling operations reduce computation while retaining key information. Finally, the fully connected layer outputs an aligned phase offset value, which represents the phase offset between array cells, typically in radians. This process enables an automated mapping from raw parameters to the offset value, improving detection efficiency.

[0017] For example, in the field of privacy screen protector quality control, consider one possible implementation where optical parameters include light intensity distribution and polarization state. The acquisition process uses a polarized light source to illuminate the array surface, and a sensor records the polarization changes of the reflected light to quantify phase deviation. Specifically, the light source emits linearly polarized light, which is scattered after hitting a microprism. The sensor separates the different polarization components using a filter and calculates the phase difference. Such details ensure parameter accuracy, providing reliable input to the network. In practical applications, this can be used in the production of privacy screen protectors for laptops to ensure that array alignment deviations are less than a preset threshold.

[0018] It's important to note that the training process for a convolutional neural network involves dataset preparation and parameter optimization. The dataset consists of privacy screen samples with known bias values, such as those labeled with phase biases through manual measurement. During training, the network minimizes the mean squared error between the predicted and actual biases by adjusting the weights through backpropagation. This training method enhances the network's ability to generalize to real-time data.

[0019] In one embodiment, the network architecture includes three convolutional layers and two pooling layers, with a 3x3 kernel size and a ReLU activation function. Through iterative training, the network learns the nonlinear relationship between optical parameters and bias, achieving accurate estimation. In another embodiment, the acquisition parameters are adjusted to suit different privacy screen protector thicknesses and business scenarios.

[0020] For example, for thinner materials, the optical sensor increases the sampling frequency to 100 times per second to capture dynamic changes. During processing, the network input is expanded to three channels, including light intensity, angle, and thickness-related parameters. After network processing, the output deviation value is used to provide feedback control for production equipment, such as adjusting laminating machine parameters to correct array alignment. This flexibility demonstrates the versatility of the technology within the same field, such as its application on mobile phone screen privacy film assembly lines.

[0021] Preferably, the deviation value calculation is incorporated into the post-processing step.

[0022] Specifically, after the network outputs the original deviation estimate, noise is removed using a smoothing filter such as a Gaussian filter to ensure value stability. Then, the deviation value is compared to a threshold; if it exceeds the threshold, an alarm is triggered. In business operations, this facilitates real-time monitoring, for example, in the roll-to-roll production of privacy film, where deviations are continuously calculated to avoid batch defects. Such a mechanism provides a complete chain from data acquisition to decision-making.

[0023] Understandably, the core of this technical solution lies in the feature extraction process of optical parameters using a convolutional neural network. This process can be explained in detail as follows: the network's convolutional layers apply filters to capture local patterns in the microprism array, such as identifying angular differences between adjacent prisms. Through multiple convolutions, high-level features, such as the phase distribution map of the overall array, are extracted. These features are then flattened and input into fully connected layers, where weights are used to calculate deviation values. In the privacy film inspection business, this process can effectively identify minute deviations, such as a 0.1 radian offset, thereby improving the accuracy of product quality control. This detailed implementation supports the broad protection of the claims. Furthermore...

[0024] In one embodiment, the integrated system includes a data acquisition module and a processing module. The data acquisition module is deployed on the production line and uses infrared sensors to acquire parameters, avoiding visible light interference. The processing module runs a convolutional neural network to calculate deviation values ​​in real time at edge devices.

[0025] For example, during the coating process of privacy films for flat panel displays, the system processes 100 samples per minute and outputs a deviation statistics report. This report includes the average deviation, standard deviation, and proportion of outliers, generated in real-time by calculating the difference between the coating thickness of each sample and the standard value, and is formatted as an Excel spreadsheet. This modular design facilitates expansion to similar optical film material applications.

[0026] For example, in privacy film aging tests, optical parameters are collected after long-term exposure, and the network processes the data to assess changes in phase deviation. The process involves simulating accelerated aging in a simulated environment, periodically collecting data from sensors, and comparing the initial and current deviations through the network to quantify the degree of degradation. This provides data support for maintenance operations, ensuring the durability of the film material.

[0027] In one embodiment, multi-sensor fusion is used to improve accuracy.

[0028] Specifically, parameters are simultaneously collected using visible light and ultraviolet sensors to form a composite dataset, which is then input into the network. The network integrates the data through a fusion layer to calculate a comprehensive deviation value. In terms of operational effectiveness, this enables more accurate alignment monitoring, for example, in the production of high-end display devices, reducing the need for human intervention.

[0029] Step S102: Based on the current alignment phase deviation value, obtain the matching relationship between the microstructure period and the screen pixel grid. If the matching relationship deviation exceeds a preset threshold, adjust the relative position of the microprism array and determine the optimized alignment phase.

[0030] By using the current alignment phase deviation value, the matching relationship between the microstructure period and the screen pixel grid is obtained, and the matching relationship deviation is determined. Based on the matching relationship deviation, it is determined whether it exceeds a preset threshold, and the deviation exceeds the threshold. If the deviation exceeds the threshold, the relative position of the microprism array is adjusted, and the position adjustment result is determined. Based on the position adjustment result, the light field distribution characteristics are obtained, and the degree of light field distribution optimization is determined. From the degree of light field distribution optimization, the re-matching relationship between the array and the grid is obtained, and the optimized alignment phase is determined.

[0031] In one implementation, the current alignment phase deviation value is first obtained, which represents the phase difference between the microstructure period and the screen pixel grid.

[0032] Specifically, the alignment phase deviation value can be calculated by comparing the repeating units of the microstructure period with the arrangement spacing of the pixel grid.

[0033] For example, the deviation value is obtained by analyzing image data acquired through optical sensors, reflecting the degree of synchronization between the two. This deviation value helps identify potential moiré patterns or image distortion problems, playing a role in the assembly process of display devices such as LCD or OLED screens. Furthermore, based on the acquired alignment phase deviation value, the matching relationship between the microstructure period and the screen pixel grid is determined. The microstructure period refers to the repeating spacing of prism units in a microprism array, while the screen pixel grid refers to the regular arrangement of pixels on the display screen. The matching relationship can be evaluated by calculating the period ratio or phase offset.

[0034] For example, if the period ratio is close to an integer, the match is good; otherwise, it may cause optical interference.

[0035] It should be noted that obtaining this matching relationship involves data processing steps, such as Fourier transform analysis of the period frequency to quantify the degree of deviation, thus providing a basis for subsequent adjustments. In display applications, this step ensures the sharpness of high-resolution images.

[0036] For example, if the matching deviation exceeds a preset threshold, such as a threshold set to 0.05 period units, an adjustment mechanism is activated. The preset threshold is determined based on the accuracy requirements of the display device to avoid image quality degradation. Adjusting the relative position of the microprism array can be achieved using precision mechanical devices, such as using a stepper motor to drive the array to move slightly within the xy-plane, gradually changing its position relative to the screen pixel grid. During this process, the deviation value is monitored in real time until the matching relationship falls within the threshold.

[0037] In one possible implementation, determining the optimized alignment phase involves iterative optimization steps.

[0038] Specifically, after the initial adjustment, the deviation value is recalculated, and if it still exceeds the threshold, the position is fine-tuned.

[0039] For example, the movement is 0.01 micrometers at a time, and optimization is achieved through a feedback loop. This iterative approach, when applied to flat panel display production lines, can effectively reduce optical defects.

[0040] Preferably, software algorithms can be used to assist in the adjustment, such as using gradient descent to simulate position changes, to accelerate the convergence process. Furthermore, this technical solution can be expanded for application in display device maintenance scenarios.

[0041] For example, when screen alignment misalignment occurs, the deviation value is obtained using a portable optical inspection instrument, and the position of the microprism array is adjusted manually or automatically to achieve phase optimization. This method improves the durability of the device and image stability.

[0042] Understandably, through the steps described above, this technical solution demonstrates versatility in the display field. For example, it allows for parameter adjustments on screens of different resolutions to accommodate various microstructure designs, thereby ensuring consistent matching results. In implementation, this process enables precise alignment control, reducing image distortion.

[0043] Step S103: Using the optimized aligned phase, simulate and calculate the light transmission path to determine the limitation range of the side viewing angle and obtain the controllable and adjustable privacy shielding intensity.

[0044] Using an optimized aligned phase, simulation calculations are performed on the light transmission path. Distribution data is obtained from the refraction angle and propagation direction of the light transmission path to determine its distribution characteristics, including the mean and variance of the angular distribution. From these distribution characteristics, a preliminary limit range for the side viewing angle is obtained. The boundary values ​​of the limit range are determined and refined by comparing the angular deviation within the preliminary range with a preset threshold of 5 degrees. For these boundary values, the dynamic parameters of privacy occlusion are adjusted. These parameters are derived from the simulation data, and deviation data is fused from the brightness and contrast thresholds of the dynamic parameters. Specifically, the adjustment basis for the occlusion intensity is calculated using the formula I = (L - 50) * 0.6 + (C - 0.5) * 0.4, where I is the adjustment basis, L is the current brightness, and C is the current contrast. Based on this adjustment basis, the matching relationship of the light field distribution obtained from the light path simulation is fused, and the adjustment basis is integrated from the phase deviation of the matching relationship to obtain a controllable and adjustable privacy occlusion intensity.

[0045] In one implementation, an optimized alignment phase is used as a basis for simulating the light transmission path in the display device. This alignment phase refers to the adjusted phase matching state between the microprism array and the screen pixel grid, which can reduce optical interference. The simulation process first involves establishing a light model.

[0046] For example, by tracing the path of incident light rays originating from the screen's self-emissive pixels, passing through the pixel layer and the microprism structure.

[0047] Specifically, the simulation of light transmission paths can be based on the principles of geometric optics, considering the refraction and reflection behavior of light. In the assembly of display devices such as OLED screens, this step is achieved through software tools. After inputting optimized alignment phase values ​​(i.e., optimized phase difference values ​​of the microprism array, obtained through preliminary optical modeling calculations), the propagation trajectory of light at different angles is calculated, thus providing a data basis for subsequent judgments. This method is applied in the production of high-resolution displays to ensure accurate simulation of light paths. Furthermore, the specific process of simulation calculation includes dividing the light path into multiple segments.

[0048] For example, the linear propagation from the light source to the pixel grid, and then to the refraction of the microprism.

[0049] It's important to note that simulating the light transmission path requires considering the effect of the periodic structure of the microprisms on light deflection. For example, when light enters at a normal incident angle, the path is relatively straight; however, at an oblique incident angle, scattering or converging may occur. By iteratively calculating the path changes under different incident angles, a light distribution map is obtained. This simulation plays a role in the privacy mode design of OLED screens, helping to identify potential leaky light paths.

[0050] In one possible implementation, a ray tracing algorithm is used to simulate the path, calculating the interaction points between the light rays and the structure point by point until the light rays exit the display surface. This process emphasizes the optimization of phase alignment, enabling the simulation results to more closely resemble actual optical behavior and demonstrating versatility in the quality control stage of display devices. Based on the simulation results, the limitation range of the side viewing angle is determined.

[0051] Specifically, the viewing angle refers to the angle threshold at which image information is still discernible when viewed from the side of the screen. The determination process involves analyzing the intensity distribution of simulated light paths.

[0052] For example, it calculates whether light can reach the observer's position at different side angles. If the light intensity exceeds a preset threshold at a certain angle, that angle is considered within the visible range. Furthermore, this judgment can be combined with vector analysis to assess the degree of deflection of the light path, thereby determining the limiting range, such as a gradual change from 0 degrees for frontal viewing to 45 degrees for side viewing. In flat panel display applications, this step ensures the effectiveness of privacy features and prevents side peeping.

[0053] Preferably, after determining the side viewing angle, a controllable and adjustable privacy shielding intensity is obtained. This intensity represents the degree of suppression of side light leakage and is adjusted by adjusting the alignment phase or microprism parameters. For example...

[0054] In one embodiment, the tilt angle of the microprism array is dynamically changed according to the determined limitation range to control the diffusion of the light path, thereby adjusting the shielding intensity from low to high.

[0055] It should be noted that the controllable adjustment process involves a feedback mechanism, such as real-time monitoring of angle changes and corresponding adjustments, to ensure that privacy levels are maintained under different usage environments. This technical solution demonstrates flexibility in the privacy protection of smartphone screens.

[0056] Understandably, this method can be extended to various display device scenarios.

[0057] For example, similar simulation and judgment steps can be used in the assembly process of laptop screens to achieve personalized adjustments to privacy shading.

[0058] Specifically, the process begins by inputting the optimized alignment phase, then simulating the light path to determine the angle range, and finally adjusting the intensity parameters. This versatility is reflected in its adaptability to screens with different resolutions, ensuring consistent light control. In another implementation, for curved displays, the simulation calculations need to additionally consider the impact of curvature on the light path.

[0059] For example, by incorporating surface geometry factors into path simulation and evaluating the additional scattering from curved surfaces when determining side angles, more precise adjustment of shading intensity can be achieved. This extension enhances the applicability of the technical solution in emerging display technologies.

[0060] Step S104: The privacy occlusion strength is obtained and iteratively optimized by combining the front brightness data. A genetic algorithm is used to process the balance between brightness loss and occlusion strength to obtain a high privacy occlusion configuration with low brightness loss.

[0061] By obtaining the privacy occlusion strength, initial balance parameters are acquired from the frontal brightness data to obtain a low brightness loss threshold. Using this low brightness loss threshold, a genetic algorithm is used to process the occlusion strength relationship to determine an iterative optimization sequence. Based on the iterative optimization sequence, data fusion is performed on the brightness loss assessment to obtain a configuration parameter adjustment scheme. If the configuration parameter adjustment scheme exceeds a preset threshold, a high privacy target verification is performed to determine the occlusion configuration generation output. The occlusion configuration generation output is obtained, and the balance parameter adjustment is integrated into the device integration application to obtain a high privacy occlusion configuration.

[0062] In one implementation, the privacy masking strength is first obtained by quantifying the degree of image blurring or contrast reduction at a side-view angle of the display device.

[0063] Specifically, privacy masking strength can be calculated based on optical simulation models, such as utilizing the pixel arrangement and light refraction principles of the display screen to assess the risk level of information leakage when viewed from the side. This strength value is usually expressed as a percentage; a higher value indicates that the image is more difficult to discern from the side, thus providing higher privacy protection. In this way, masking parameters are initially set while ensuring that frontal viewing is unaffected. Further, iterative optimization is performed using frontal brightness data. Frontal brightness data comes from the display device's brightness sensor or a preset brightness distribution map; this data reflects the light intensity value from the user's frontal viewpoint. During the iteration process, privacy masking strength and brightness data are used as input parameters to repeatedly adjust the masking configuration.

[0064] For example, gradient descent can be used as an aid to gradually reduce brightness loss, but the core relies on a subsequent genetic algorithm to globally search for the optimal solution. This optimization aims to balance two metrics, avoiding excessive attenuation of frontal brightness caused by simply increasing occlusion intensity.

[0065] Preferably, a genetic algorithm is used to handle the balance between brightness loss and occlusion intensity. A genetic algorithm is an optimization technique that simulates natural evolution, solving multi-objective problems through population initialization, selection, crossover, and mutation operations. In this embodiment, a population is first initialized, with each individual representing an occlusion configuration, including parameters such as occlusion layer thickness and material refractive index. Then, a fitness function is defined, which comprehensively evaluates brightness loss (calculated as a percentage decrease in frontal brightness) and occlusion intensity (calculated as the reciprocal of side-view visibility). For example, fitness value = w1 * (1 - brightness loss rate) + w2 * occlusion intensity, where w1 and w2 are weighting coefficients that can be adjusted according to the application scenario. Through multiple generations of iteration, individuals with high fitness are selected for crossover, for example, by single-point crossover to exchange configuration parameters. Mutation is then introduced to increase diversity, such as randomly changing the value of a parameter. This process continues until convergence, yielding the configuration scheme on the Pareto optimal front. The Pareto optimal front refers to the set of optimal compromise solutions found when solving a problem with multiple conflicting objectives (e.g., simultaneously pursuing the lowest cost, highest performance, and lowest energy consumption). The advantage of this algorithm lies in its ability to handle nonlinear equilibrium relationships and avoid local optima traps, making it particularly suitable for complex optical environments in the field of display privacy protection.

[0066] In one possible implementation, for privacy masking applications on laptop displays, the initial population size is set to 50 individuals, with each generation iterating 100 times. Brightness loss is quantified by measuring changes in frontal luminous flux; for example, with an initial brightness of 400 nits, the optimized loss is controlled to within 5%. Simultaneously, the masking strength requires image blurring to exceed 80% at a side-viewing angle greater than 30 degrees. Through a genetic algorithm, the system automatically generates various configurations, such as adjusting the polarizing film angle or adding microstructure layers, to achieve low loss under high masking conditions.

[0067] It should be noted that the balance processing of genetic algorithms also includes multi-objective optimization strategies.

[0068] Specifically, an NSGA-II variant can be introduced, which maintains population diversity through non-dominated ranking and crowding comparison. During ranking, individuals are divided into different frontier layers, with the first layer representing Pareto-optimal individuals. Then, the crowding distance for each individual is calculated to ensure uniform distribution. This method can effectively explore the brightness-occlusion trade-off space in business applications. For example, in mobile phone screen privacy modes, it allows users to customize weight preferences, such as increasing the w1 value to prioritize brightness, thereby generating configurations adapted to different lighting conditions. In another embodiment, for the privacy function of tablet computers, iterative optimization is combined with real-time brightness data feedback. After the sensor collects the frontal brightness, it is dynamically adjusted in the fitness function of the genetic algorithm.

[0069] For example, if the ambient light is dim, the algorithm tends to reduce the occlusion intensity to maintain brightness. Through multiple simulations, this configuration has been shown to minimize brightness loss while maintaining privacy, making it suitable for indoor office scenarios.

[0070] Understandably, once a high privacy masking configuration with low brightness loss is obtained, this configuration can be applied to display control systems.

[0071] Specifically, the final output includes a set of parameters, such as masking layer density and brightness compensation value, which are implemented through hardware drivers.

[0072] For example, in LCD screens, backlight partitioning is adjusted to compensate for losses, thereby ensuring a better user experience. This technique provides an efficient privacy protection mechanism without significantly impacting display quality. Furthermore, in one implementation, scenarios with different display resolutions are considered, such as the difference between a 4000-resolution screen and a full HD screen. During the initialization of the genetic algorithm population, an initial population is generated from a random partition configuration, incorporating a resolution factor as a variable. The fitness function evaluates the balance between privacy protection strength and display quality loss, the optimization objective is to minimize the overall loss, and the output is optimized partitioning parameters suitable for various devices. This extension improves the compatibility of the solution with devices of different resolutions.

[0073] For example, in e-reader applications, the privacy shielding strength is calculated by combining the reflectivity of the e-ink screen with front brightness data for iteration. When processing with a genetic algorithm, the mutation rate is set to 0.01 and the crossover probability to 0.8. After optimization, a configuration is obtained that achieves high privacy with low power consumption.

[0074] Preferably, the entire process can be integrated into a software framework to ensure seamless execution. Through the above steps, this method achieves a balanced optimization in the field of display privacy, providing reliable technical support.

[0075] Step S105: Based on the high privacy masking configuration, obtain the potential occurrence points of moiré interference. If the number of potential occurrence points exceeds a preset threshold, modify the periodic arrangement of the microstructure to determine the array layout for suppressing optical interference.

[0076] By employing a high-privacy masking configuration, potential sources of moiré interference are identified through optical pattern scanning, and the number of these potential sources is determined. If the number of potential sources exceeds a preset threshold, an optical occurrence threshold is used to compare parameters and determine a microstructure periodicity modification scheme. Based on this microstructure periodicity modification scheme, the potential interference sources are periodically arranged and adjusted to obtain suppression optical arrangement parameters. These suppression optical arrangement parameters are then fused with the display layer and adjusted through hierarchical integration to determine the final configuration for array layout suppression.

[0077] In one implementation, based on a high privacy masking configuration, potential locations of moiré interference are first identified. Moiré interference refers to striped visual artifacts that occur when a microstructure layer is superimposed on a periodic pattern of a pixel array in a display device; this interference can reduce the effectiveness of privacy masking.

[0078] Specifically, the high-privacy masking configuration includes previously optimized masking layer parameters, such as microstructure thickness or refractive index, which are used to simulate optical paths. Using an optical simulation model, the system analyzes the propagation of light at side-view angles and identifies areas that may cause interference.

[0079] For example, in laptop displays, ray tracing is used to track light paths and calculate frequency matching points between microstructure periods and pixel spacing. These matching points are potential interference points, and their number is counted using a counting algorithm to quantify the level of interference risk. Ray tracing is a core concept in computer graphics, optical design, and display technology. This acquisition process ensures the integrity of privacy protection configurations and provides foundational data for subsequent steps. Furthermore, if the number of potential interference points exceeds a preset threshold, the microstructure period arrangement needs to be modified. The preset threshold can be set according to the type of display device; for example, in a mobile phone screen privacy mode, the threshold is set to 5 points per square centimeter to avoid visible interference. The modification process involves adjusting the period parameters of the microstructures, such as changing the spacing or angle of the arrangement, to disrupt the frequency matching. Specifically...

[0080] In one possible implementation, an iterative adjustment strategy is employed, starting with an initial period value and gradually increasing or decreasing the spacing value, for example, adjusting the period from 10 micrometers to 12 micrometers, to resimulate the distribution of interference points until the number drops below a threshold. This modification helps maintain high privacy masking strength while suppressing optical interference.

[0081] Preferably, the array layout for suppressing optical interference is determined through a comprehensive evaluation of the modified configuration. Array layout refers to the two-dimensional arrangement of microstructures on the display screen, such as randomization or non-uniform distribution, to further reduce moiré patterns.

[0082] It should be noted that in the privacy protection function of the tablet computer, the system generates multiple layout schemes, such as a quasi-periodic arrangement where the spacing of the microstructural units is gradually distributed. The suppression effect is verified through simulation. This layout determination process includes comparing the number of interference points of different schemes to ensure that the selected layout achieves low interference output while maintaining privacy.

[0083] For example, in privacy protection scenarios for e-readers, when identifying potential sources of interference, the reflective characteristics of the e-ink screen are considered, and the path of side-view light reflection is simulated to identify high-risk areas. If the number of sources exceeds a threshold, such as 10 points per screen, the microstructure periodicity is modified, for example, from a uniform arrangement to an interlaced arrangement, thereby determining an array layout that suppresses interference, such as a honeycomb pattern. This method, when applied in low-light indoor environments, can effectively balance privacy and display clarity.

[0084] Understandably, the entire process can be integrated into the display control system.

[0085] In one embodiment, for a 4K resolution display, the potential occurrence points are acquired by combining the resolution factor with a higher threshold to accommodate finer pixels. The periodic arrangement is then modified to a non-linear mode, ultimately determining the array layout as a multi-layered stack. This integrated approach demonstrates the versatility of the technology across different display devices. Furthermore, in another embodiment, for privacy masking on curved displays, acquiring potential moiré pattern occurrence points requires considering the curvature effect. Interference points in curved regions are calculated using a curved optical model. If the threshold is exceeded, the microstructure periodicity is modified to an adaptive arrangement, such as gradually varying the spacing along the curvature direction, determining the array layout as dynamically adjustable. This process enables stable privacy protection for curved devices without introducing additional optical artifacts.

[0086] For example, in the application of office monitors, the system first simulates interference points based on a high privacy configuration. If the number exceeds a threshold after counting, it makes periodic modifications, such as introducing random offsets. The final layout is determined to be a hybrid array, ensuring high privacy when viewed from the side and no interference visible from the front.

[0087] It should be noted that the logical sequence of these steps, from acquisition to layout determination, forms a continuous optimization chain, providing a reliable interference suppression mechanism in the field of display privacy.

[0088] Step S106: Using an array layout that suppresses optical interference, dynamically verify the privacy angle, determine whether the adjustment range covers the requirements of the specified scenario, and obtain the final set of controllable adjustment parameters.

[0089] By configuring privacy masking, moiré pattern occurrence points are obtained through scanning the display layer, resulting in a moiré pattern occurrence point distribution. Optical interference is assessed based on this distribution, and parameters are compared by matching interference intensity with period to determine a microstructure period adjustment scheme. According to this microstructure period adjustment scheme, a dynamic verification process is used to simulate the privacy angle, and the adjustment range verification results are obtained by tracking angle changes. Based on the adjustment range verification results and scenario requirements, if the adjustment range covers the specified scenario requirements, the optical parameters for suppression are fused and adjusted hierarchically. The array layout suppression configuration is determined by parameter superposition. Based on the array layout suppression configuration, a controllable parameter set is obtained by fusing the display layer to obtain an array layout that suppresses optical interference.

[0090] In one implementation, an array layout that suppresses optical interference is employed, first dynamically verified for privacy angles. The array layout that suppresses optical interference refers to a two-dimensional pattern formed by a previously modified periodic arrangement of microstructures, such as a quasi-periodic or honeycomb distribution, used to reduce moiré patterns in the display device.

[0091] Specifically, dynamic verification involves simulating light paths at different viewing angles to assess the effectiveness of privacy shielding. In the privacy mode application of laptops, the system uses optical simulation software to track light propagation at side-view angles and calculate the degree of leakage of visible information, such as the privacy protection range from 30 degrees to 60 degrees. By comparing the simulation results with standard thresholds, it verifies whether the array layout maintains high shielding strength within the specified angle. This verification process ensures the reliability of the layout in actual use and provides a data basis for subsequent judgments. Furthermore, determining whether the adjustment range covers the requirements of the specified scenario is achieved by analyzing the verification results. The adjustment range refers to the range of adjustable parameters in the array layout, such as the variation in microstructure spacing or angle, which affects the coverage of the privacy protection effect.

[0092] It should be noted that the specified scenario requirements include privacy needs in environments such as indoor offices or outdoor mobile settings. For example, in mobile phone screen privacy protection, the requirement might be to cover a viewing angle from the front to a 45-degree side angle. The judgment process involves checking whether the boundary values ​​of the adjustment range meet these requirements. For example, if the range allows the spacing to be adjusted from 8 micrometers to 15 micrometers, its occlusion performance in low-light scenes is evaluated. Through a logical comparison algorithm, the system determines whether the range is comprehensive. If the coverage is insufficient, the parameter range needs to be expanded. This judgment helps optimize the adaptability of the layout.

[0093] Preferably, the final set of controllable adjustment parameters is obtained by integrating verification and judgment results. The set of controllable adjustment parameters refers to a set of optimized parameter combinations, including spacing values, arrangement angles, and thicknesses, used to adjust the array layout in real time.

[0094] In one possible implementation, for the privacy protection function of a tablet computer, the system extracts effective parameters from dynamic verification data, such as setting the spacing to a gradual mode to cover multiple scenario requirements. Obtaining the parameter set involves an iterative filtering process: first, potential parameters are listed, then items that are not covered are eliminated based on the judgment results, ultimately forming a streamlined set.

[0095] For example, in e-reader applications, the parameter set might include three spacing options to accommodate the characteristics of a reflective display. This method of acquisition allows for precise control of the parameters.

[0096] Understandably, the entire process can be integrated into the display control system.

[0097] Specifically, in privacy shielding for curved displays, an array layout is used to dynamically verify the privacy angle, considering the impact of curvature on light, and determining whether the adjustment range covers the required curved area. If satisfied, a set of parameters, such as adaptive pitch values, is obtained. This integration demonstrates the versatility of the technology in display devices.

[0098] For example, in the implementation of office monitors, the system dynamically verifies the shading effect of the side-view angle based on the array layout that suppresses interference, determines the adjustment range to cover the indoor scene requirements, and the final parameter set includes multi-layer arrangement options to ensure the stability of privacy protection.

[0099] Step S107: By integrating the light transmission model and privacy shielding intensity through the final controllable adjustable parameter set, the overall performance index of the privacy film is determined, and an adjustment command is output after determining that the index meets the business objectives.

[0100] An overall performance index is obtained by integrating a light transmission model and privacy occlusion intensity through a controllable adjustable parameter set. Based on this overall performance index, a material transmittance angle evaluation value is obtained to determine angle-dependent performance. The angle-dependent performance is then fused with the occlusion dynamic response to determine if the index meets business objectives. If the index meets business objectives, an adjustment command is output. Based on the adjustment command, a parameter optimization set is obtained to determine the final adjustment sequence.

[0101] In one implementation, the performance optimization process of the privacy film first involves the construction of a final set of controllable adjustable parameters.

[0102] Specifically, the set of controllable parameters is formed by collecting relevant variables of the privacy screen protector material, such as film thickness, refractive index, and surface microstructure parameters. These parameters are derived from experimental data or simulation calculations and can be dynamically adjusted by the user or system to affect the overall performance of the film.

[0103] It should be noted that the integration of controllable parameter sets aims to provide flexibility. For example, in display applications, the parameter set can include adjustable values ​​for the light incident angle to ensure consistent performance of the film under different lighting conditions. Through this parameter set, the system can achieve precise control over the optical properties of the privacy film, thus laying the foundation for subsequent model integration. Furthermore, the establishment of a light transmission model is a crucial step in the integration process. The light transmission model, based on optical principles, simulates the propagation path of light within the privacy film.

[0104] Specifically, the model takes into account the refraction, reflection, and scattering of light, for example, by using Snell's law to describe the interfacial behavior of light entering the film from the air, without involving complex numerical calculations.

[0105] In one possible implementation, the light transmission model can represent the multi-layered material of the privacy screen using a layered structure, with the optical constants of each layer, such as transmittance, parameterized. By integrating these elements, the model can predict light transmission efficiency under specific parameters; for example, in smartphone screen privacy applications, the model evaluates high transmittance at a frontal viewing angle and low transmittance at a side viewing angle. The principle behind this model is to balance the transmission and blocking of the visible spectrum, ensuring user privacy without sacrificing display clarity.

[0106] For example, if the parameter set adjusts the film thickness, the model updates the simulation of the transmission path accordingly, demonstrating how parameter changes affect the overall optical effect. A detailed description of this process helps in understanding how the model, in conjunction with the privacy shielding strength, forms the core function of the privacy screen protector.

[0107] Preferably, the quantification of privacy shielding strength and its integration with the light transmission model constitute the innovative aspect of performance optimization. Privacy shielding strength refers to the membrane's ability to block lateral peeping, and is typically expressed by measuring the light attenuation rate at the lateral viewing angle.

[0108] Specifically, the integration process involves superimposing the output data of light passing through the model with the occlusion intensity index, for example, by calculating a synthesis function in which the light path data is provided by the model and the occlusion intensity is based on the angle-dependent absorption coefficient.

[0109] In one embodiment, for a privacy screen protector for a laptop screen, the system first extracts relevant variables from a parameter set, such as microstructure density. Then, it inputs a light transmission model to simulate the difference in luminous flux between frontal and side views, and subsequently calculates the shading intensity value, such as a threshold of side-view transmittance below 10%. The principle of this integration is to use the parameter set as a bridge to achieve dynamic linkage between models, ensuring that the shading effect is enhanced without reducing frontal transmittance. In this way, the privacy screen protector can effectively prevent others from peeping at the screen content when used in public places such as coffee shops, while maintaining the clarity of the user's viewing angle. The technical effect of this integration step is to improve the adaptability of the film, allowing it to be adjusted according to different business objectives, such as high privacy requirements or high brightness requirements. Based on the above integration, determining the overall performance indicators of the privacy screen protector becomes the next step. Overall performance indicators typically include composite values ​​such as transmittance efficiency, shading effectiveness, and durability.

[0110] Specifically, these metrics are derived from the model output using a weighted average method; for example, penetration efficiency accounts for 40% of the metric, and occlusion effectiveness accounts for 50%. In one implementation, the system aggregates the model results after parameter set adjustments to generate a numerical metric, such as a performance score ranging from 0 to 100. Further, determining whether the metric meets business objectives involves threshold comparisons.

[0111] For example, if the business objective requires a performance score exceeding 80, the system checks whether the calculated metric meets the target. This judgment process is simple and automated, ensuring a rapid response. The final step is to output adjustment instructions after the judgment.

[0112] For example, if the indicators are not met, the instructions may include a recommendation to increase the film thickness to guide further optimization.

[0113] Understandably, in another embodiment, the process is applied to a tablet privacy screen protector, with the parameter set focusing on touchscreen compatibility, model integration emphasizing multi-angle simulation, and similarly outputting targeted instructions.

[0114] It should be noted that this method is versatile in the field of display devices. For example, in ATM screen applications, the integrated model can adapt to changes in outdoor lighting.

[0115] In one embodiment, the overall process starts from the parameter set and ends with the instruction output, forming a closed-loop optimization and improving the practical value of the privacy film.

[0116] The above-disclosed embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of the invention. Those skilled in the art will understand that implementing all or part of the above-described embodiments and making equivalent changes in accordance with the claims of the present invention are still within the scope of the invention.

Claims

1. A method for performance optimization in the manufacturing process of privacy films, characterized in that, include: Real-time optical parameters of the microprism array on the surface of the privacy film are collected and processed by a convolutional neural network to obtain the current alignment phase deviation value; The matching relationship between the microstructure period and the screen pixel grid is obtained based on the current alignment phase deviation value. When the matching relationship deviation exceeds a preset threshold, the relative position of the microprism array is adjusted to obtain an optimized alignment phase. The optimized alignment phase is used to simulate and calculate the light transmission path to obtain a controllable privacy masking intensity. The privacy masking intensity is combined with the front brightness data and a genetic algorithm is used for iterative optimization to obtain a high privacy masking configuration with low brightness loss. The potential occurrence points of moiré interference are obtained based on the high privacy masking configuration. When the number of potential occurrence points exceeds a preset threshold, the arrangement of the microstructure period is modified to obtain an array layout that suppresses optical interference. The array layout for suppressing optical interference is used to dynamically verify the privacy angle and obtain the final set of controllable adjustment parameters. The light transmission model and privacy shielding intensity are integrated through the final set of controllable adjustment parameters to determine the overall performance index of the privacy film and output adjustment instructions when the preset business objectives are met.

2. The performance optimization method for the manufacturing process of a privacy film as described in claim 1, characterized in that, The process of acquiring real-time optical parameters of the microprism array on the surface of the privacy film and processing them through a convolutional neural network to obtain the current alignment phase deviation value includes: acquiring real-time optical parameters of the microprism array on the surface of the privacy film through an optical sensor to obtain a parameter data set; inputting the parameter data set into a convolutional neural network for feature extraction processing to obtain the current alignment phase deviation value; if the current alignment phase deviation value exceeds a preset threshold, obtaining surface structure analysis results through surface scanning to determine the source of the deviation; extracting a calibration mechanism from the neural network training data to obtain adjustment parameters based on the source of the deviation; updating the microprism array structure through the adjustment parameters and then inputting the parameter data set back into the convolutional neural network for iterative processing to obtain the updated current alignment phase deviation value.

3. The performance optimization method for the manufacturing process of a privacy film as described in claim 1, characterized in that, The step of obtaining the matching relationship between the microstructure period and the screen pixel grid based on the current alignment phase deviation value, and adjusting the relative position of the microprism array to obtain the optimized alignment phase when the matching relationship deviation exceeds a preset threshold, includes: determining the matching relationship between the microstructure period and the screen pixel grid based on the current alignment phase deviation value to obtain the matching relationship deviation; determining whether the matching relationship deviation exceeds a preset threshold to obtain a deviation exceeding condition; if the deviation exceeding condition is met, adjusting the relative position of the microprism array to obtain a position adjustment result; obtaining light field distribution characteristics based on the position adjustment result to determine the degree of light field distribution optimization; and determining the re-matching relationship between the array and the grid based on the degree of light field distribution optimization to obtain the optimized alignment phase.

4. The performance optimization method for the manufacturing process of a privacy film as described in claim 1, characterized in that, The process of simulating and calculating the light transmission path using the optimized alignment phase to obtain a controllable and adjustable privacy shielding intensity includes: simulating and calculating the light transmission path using the optimized alignment phase to obtain distribution data of refraction angle and propagation direction; determining the distribution characteristics of the light transmission path based on the distribution data; obtaining the limitation range of the side viewing angle based on the distribution characteristics; determining the boundary value of the limitation range by comparing the angular deviation within the limitation range; adjusting the dynamic parameters of the privacy shielding based on the boundary value and fusing a brightness threshold and a contrast threshold to obtain the adjustment basis for the shielding intensity; and fusing the matching relationship of the light field distribution based on the adjustment basis to obtain a controllable and adjustable privacy shielding intensity.

5. The performance optimization method for the manufacturing process of a privacy film as described in claim 1, characterized in that, The step of obtaining a high privacy occlusion configuration with low brightness loss by iteratively optimizing the privacy occlusion strength and frontal brightness data using a genetic algorithm includes: determining initial balance parameters based on the privacy occlusion strength and frontal brightness data; using a genetic algorithm to iteratively process the balance relationship between occlusion strength and brightness loss with low brightness loss as a constraint to obtain an iterative optimization sequence; evaluating the brightness loss based on the iterative optimization sequence and fusing the evaluation results to obtain a configuration parameter adjustment scheme; and outputting a high privacy occlusion configuration if the configuration parameter adjustment scheme meets preset conditions.

6. The performance optimization method for the manufacturing process of a privacy film as described in claim 1, characterized in that, The step of obtaining potential moiré interference points based on the high privacy masking configuration and modifying the microstructure periodic arrangement to obtain an array layout that suppresses optical interference when the number of potential points exceeds a preset threshold includes: obtaining the number of potential moiré interference points by optical pattern scanning based on the high privacy masking configuration; determining whether the number of potential points exceeds a preset threshold; if it exceeds, determining a microstructure periodic modification scheme based on the optical occurrence threshold; adjusting the periodic arrangement of the interference potential points according to the microstructure periodic modification scheme to obtain suppression optical arrangement parameters; and performing hierarchical integration adjustment according to the suppression optical arrangement parameters to obtain an array layout that suppresses optical interference.

7. The performance optimization method for the manufacturing process of a privacy film as described in claim 1, characterized in that, The process of dynamically verifying the privacy angle using the array layout that suppresses optical interference to obtain the final set of controllable adjustable parameters includes: scanning the display layer to obtain the distribution of moiré pattern occurrence points based on the array layout that suppresses optical interference; determining a microstructure periodic adjustment scheme by comparing parameters based on the interference intensity and periodicity matching relationship based on the distribution of moiré pattern occurrence points; performing angle simulation on the privacy angle using a dynamic verification process based on the microstructure periodic adjustment scheme to obtain angle change tracking results; determining whether the adjustment range covers the specified scenario requirements based on the angle change tracking results; if it covers, then fusing the suppression optical parameters and superimposing the parameters to obtain the final set of controllable adjustable parameters.

8. The performance optimization method for the manufacturing process of a privacy film as described in claim 1, characterized in that, The process of integrating the light transmission model and privacy shielding intensity with the final controllable adjustable parameter set to determine the overall performance index of the privacy screen protector and outputting an adjustment command when the preset business objectives are met includes: integrating the final controllable adjustable parameter set with the light transmission model and privacy shielding intensity to obtain the overall performance index; determining the angle-dependent performance evaluation value based on the overall performance index; judging whether the index meets the business objectives by fusing the shielding dynamic response based on the angle-dependent performance evaluation value; and outputting an adjustment command based on the judgment result if the business objectives are met.

9. The performance optimization method for the manufacturing process of a privacy film as described in claim 1, characterized in that, The process of acquiring real-time optical parameters of the microprism array on the privacy film surface and processing them through a convolutional neural network to obtain the current alignment phase deviation value includes: acquiring a parameter data set containing the refractive properties and light intensity distribution of the microprisms in real time through an optical sensor; inputting the parameter data set into a pre-trained convolutional neural network model for multi-layer feature extraction and mapping; outputting the current alignment phase deviation value, which characterizes the degree of phase misalignment between the microprism array and the pixel grid, from the convolutional neural network model; and triggering subsequent surface structure deviation source analysis and structure calibration adjustment procedures when the current alignment phase deviation value exceeds a preset range.

Images