Full-color micro led micro display array driving method and device

Abstract

The present application relates to the technical field of display driving, and particularly relates to a full-color Micro LED micro display array driving method and device. The method comprises the following steps: performing multi-frequency full-pixel excitation testing and pixel point excitation threshold sensitivity analysis on the micro display array, so as to obtain the excitation threshold sensitivity of each pixel point; performing pixel point linkage change identification on the micro display array, and performing pixel array topology driving adjustment according to the excitation threshold sensitivity, to construct pixel array topology driving parameters; performing real-time driving control according to the pixel array topology driving parameters, and performing spectral feedback regulation, to obtain a spectral regulation image output stream; performing interframe pixel movement tracking and sub-pixel level light emission track rewriting on the spectral regulation image output stream, to obtain a visual consistency image output stream. The present application improves the picture quality stability and clear visual effect of the micro display array.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of display driving, and in particular to a full-color Micro LED micro display array driving method and device. BACKGROUND

[0002] With the continuous evolution of new display technologies, Micro LED (Micro Light Emitting Diode) has become an important development direction of new generation display devices due to its high brightness, high contrast, ultra-low power consumption, and ultra-long service life, and many other advantages. Especially in the high-precision and high-integration display application scenarios such as augmented reality (AR), virtual reality (VR), wearable devices, and micro projection, Micro LED display technology shows broad application prospects. Among them, full-color Micro LED micro display technology, as one of the core technologies to achieve high-resolution and high-quality display effects, has attracted widespread attention and in-depth research in the industry.

[0003] Compared with traditional LCD and OLED display technologies, Micro LED micro display does not require a backlight source, and can achieve higher brightness output, faster response speed, and lower energy consumption, and performs particularly well in micro display scenarios. However, realizing the full-color display capability of Micro LED display faces a series of technical problems. Among them, the high-precision driving problem of the pixel array is one of the key bottlenecks restricting its industrialization and large-scale application. Since full-color Micro LED usually needs to be controlled by red, green, and blue sub-pixels to achieve color mixing and display, how to uniformly, efficiently, and accurately drive micro LED devices with different wavelengths and different driving characteristics has become an important technical challenge.

[0004] In the prior art, traditional driving methods mostly use passive matrix or simple active matrix control, and cooperate with external control chips to perform timing driving and brightness control on each sub-pixel. Although this method performs well in some low-resolution or monochrome display scenarios, it faces many problems such as large signal interference, poor driving precision, and difficult power consumption control in high-integration and high-resolution full-color micro display applications. At the same time, due to the significant differences in material properties, electrical characteristics, and response behavior of red, green, and blue Micro LED, the traditional driving architecture has limited performance in coordinating the response speed and brightness consistency between different sub-pixels, often leading to display quality problems such as color distortion and brightness deviation. SUMMARY

[0005] To solve the above technical problems, the present application provides a full-color Micro LED micro display array driving method and device to solve at least one of the above technical problems.

[0006] To achieve the above object, the application provides a full-color Micro LED micro display array driving method, comprising the following steps:

[0007] Step S1: performing multi-frequency full-pixel excitation test and pixel point excitation threshold sensitivity analysis on the micro display array, so as to obtain the excitation threshold sensitivity of each pixel point;

[0008] Step S2: performing pixel point linkage change identification on the micro display array, and performing pixel array topology driving adjustment according to the excitation threshold sensitivity, to construct pixel array topology driving parameters;

[0009] Step S3: performing real-time driving control according to the pixel array topology driving parameters, and performing spectral feedback regulation, to obtain a spectral regulation image output stream;

[0010] Step S4: performing inter-frame pixel movement tracking and sub-pixel level light-emitting track rewriting on the spectral regulation image output stream, to obtain a visual consistency image output stream;

[0011] Step S5: performing multi-index change deep fitting on the visual consistency image output stream, and performing adaptive correction, to construct an adaptive driving state adjustment strategy;

[0012] Step S6: performing partition intelligent driving control on the visual consistency image output stream, and performing intelligent collaborative driving control based on the adaptive driving state adjustment strategy, to execute micro display array driving operation.

[0013] The present application can accurately identify the electrical-optical response threshold of each pixel point under different waveform conditions by applying multi-frequency and multi-amplitude current pulse to each pixel point in the full-color Micro LED micro display array for systematic excitation test; by recording the response curve under different frequencies, the excitation threshold sensitivity parameters of the pixel points are extracted, which can accurately describe the starting ability and light output efficiency of each pixel under current excitation; the pixel cascade recognition mechanism is introduced, which can capture the optical behavior coupling between pixels caused by content change or color dynamics in the full array range; combined with the obtained excitation threshold sensitivity, a differentiated regulation strategy is adopted to establish a topological mapping relationship between high-sensitivity pixels and low-sensitivity pixels; the constructed pixel array topological driving parameter model can dynamically describe the response behavior mode between pixel groups, providing regional collaborative path for driving control; through real-time detection and feedback of the spectral characteristics of the output image (such as color temperature, saturation, color gamut boundary offset), the PWM dimming, pulse width and current amplitude of the sub-pixel are dynamically adjusted; the generated spectral regulation image output stream can better adapt to the color restoration degree required by the real picture; the inter-frame pixel tracking technology is introduced, and the pixel position change and motion trajectory between frames are identified through methods such as optical flow estimation; on this basis, the sub-pixel level light emitting trajectory is written back, and the light emitting behavior of high-speed motion or edge detail area is reconstructed; it can effectively eliminate the problems such as ghosting, residual image and mosaic in the motion image due to the delayed response of the pixels; from the perspective of display output, multiple optical indicators including brightness balance, color saturation offset and local contrast change are extracted; through deep neural network or dynamic surface fitting algorithm, these indicators are analyzed by nonlinear regression fitting, and the actual influence of driving parameters on visual results is identified; an iteratively updated driving state adjustment strategy model is constructed to realize the dynamic optimization of the pixel array; the closed-loop control capability of the display system is strengthened, which has the potential of self-learning and self-repairing, effectively improving the picture quality stability and adaptive ability under long-term operation. The regional driving strategy is adopted, the display array is divided into several logical partitions, and local driving tasks are performed according to the visual consistency results and driving state strategy respectively; the dynamic load, current distribution and response speed of each region are customized for driving optimization; through intelligent collaborative mechanism (such as task scheduling, power allocation and brightness compensation), efficient collaborative driving of the whole display panel is realized.

[0014] In the present specification, a full-color Micro LED micro display array driving device is provided for performing the method as described above, comprising:

[0015] A multi-frequency excitation test module is used for multi-frequency full-pixel excitation test and pixel point excitation threshold sensitivity analysis of the micro display array, so as to obtain the excitation threshold sensitivity of each pixel point;

[0016] a topological driving module for identifying pixel linkage changes of the micro display array and adjusting topological driving between pixel arrays according to the excitation threshold sensitivity, and constructing topological driving parameters of the pixel array;

[0017] a spectral feedback regulation module for real-time driving control according to the topological driving parameters of the pixel array, and spectral feedback regulation to obtain a spectral regulation image output stream;

[0018] a trajectory backwriting module for tracking inter-frame pixel movement and sub-pixel level light-emitting trajectory backwriting of the spectral regulation image output stream to obtain a visual consistency image output stream;

[0019] a driving state adjustment module for multi-index change deep fitting of the visual consistency image output stream and adaptive correction to construct an adaptive driving state adjustment strategy;

[0020] a partitioned intelligent driving module for partitioned intelligent driving control of the visual consistency image output stream and intelligent collaborative driving control based on the adaptive driving state adjustment strategy to perform micro display array driving operations.

[0021] The application lays the foundation for subsequent accurate driving by accurately modeling the micro electrical characteristics of each pixel point and obtaining the excitation threshold of each pixel point at different frequencies. The individual differences of pixels caused by inconsistent manufacturing processes are solved, and fine calibration is realized at the pixel level. This test method is more representative than single-frequency static testing, can simulate the response behavior under actual working conditions, and improves the matching degree between testing and application. By constructing a dynamic response relationship topology network between pixels, the actual imaging linkage characteristics are better reflected. The driving control optimization from point to surface and from individual to whole is realized, and the overall consistency and synergy of the display array are improved. It can dynamically adapt to non-ideal behavior between pixels, effectively suppress local non-uniform problems such as hot spots and dark spots. Combined with the spectral feedback mechanism, accurate color restoration and dynamic white balance adjustment are realized, and the overall visual quality is improved. It can adaptively adjust the color shift caused by environmental light changes and material aging, prolong the service life of the device and improve color stability. It supports HDR display requirements and realizes dynamic range expansion in brightness and color saturation. It accurately tracks high-speed moving image content, solves the problems of smearing and distortion in moving images, and significantly improves dynamic clarity. The sub-pixel level write-back capability improves the spatial resolution performance of the image, especially in edge details and fast switching scenarios. Combined with the information of the previous modules, the whole machine has the driving ability of "understanding image motion", which improves the subjective visual consistency of the human eye. It can dynamically adapt to different image content (static / dynamic, bright / dark field, etc.), providing optimized driving for multiple scene adaptation. Through machine learning or deep fitting, the display behavior is predicted and the driving state is adjusted in advance to reduce flicker, delay and other problems. The robustness of the system is enhanced, and it has self-repairing and self-adjusting capabilities, reducing the need for manual maintenance and complex calibration. Avoid energy waste and refresh resource overload caused by full-screen synchronous driving; high-frequency change areas respond first, and low dynamic areas maintain energy-saving mode, greatly improving the energy efficiency ratio of the display system; support for cross-area optical consistency control to avoid phenomena such as "image tearing" and "local lag"; the construction of a hybrid control system of "partition driving + strategy unification" is a key technology for future large-scale micro display array control. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 A step flowchart of a full-color Micro LED micro display array driving method of the application is shown in the figure.

[0023] Figure 2 A detailed implementation step flowchart of step S1 is shown in the figure.

[0024] Figure 3 A detailed implementation step flowchart of step S2 is shown in the figure.

[0025] Figure 4 A detailed implementation step flowchart of step S3 is shown in the figure. DETAILED DESCRIPTION

[0026] It is to be understood that the specific examples described herein are merely exemplary and are not intended to limit the application.

[0027] The application example provides a full-color Micro LED micro display array driving method and device. The execution subject of the full-color Micro LED micro display array driving method and device includes but is not limited to mechanical equipment, a data processing platform, a cloud server node, a network upload device and the like which can be regarded as a general computing node of the application.

[0028] Please refer to Figures 1 to 4 The application provides a full-color Micro LED micro display array driving method, which includes the following steps:

[0029] Step S1: performing multi-frequency full-pixel excitation test and pixel point excitation threshold sensitivity analysis on the micro display array, so as to obtain the excitation threshold sensitivity of each pixel point;

[0030] Step S2: performing pixel point linkage change identification on the micro display array, and performing pixel array topology driving adjustment according to the excitation threshold sensitivity, to construct a pixel array topology driving parameter;

[0031] Step S3: performing real-time driving control according to the pixel array topology driving parameter, and performing spectral feedback regulation, to obtain a spectral regulation image output stream;

[0032] Step S4: performing interframe pixel movement tracking and sub-pixel level light-emitting track rewriting on the spectral regulation image output stream, to obtain a visual consistency image output stream;

[0033] Step S5: performing multi-index change deep fitting on the visual consistency image output stream, and performing adaptive correction, to construct an adaptive driving state adjustment strategy;

[0034] Step S6: performing partition intelligent driving control on the visual consistency image output stream, and performing intelligent collaborative driving control based on the adaptive driving state adjustment strategy, to execute a micro display array driving operation.

[0035] The present application can accurately identify the electrical-optical response threshold of each pixel point under different waveform conditions by applying multi-frequency and multi-amplitude current pulse to each pixel point in the full-color Micro LED micro display array for systematic excitation test; the excitation threshold sensitivity parameters of the pixel points are extracted by recording the response curves under different frequencies, so as to accurately describe the starting ability and light output efficiency of each pixel under current excitation; the pixel cascade linkage recognition mechanism is introduced, which can capture the optical behavior coupling between pixels caused by content change or color dynamics in the full array range; combined with the obtained excitation threshold sensitivity, a differentiated regulation strategy is adopted to establish a topological mapping relationship between high-sensitivity pixels and low-sensitivity pixels; the constructed pixel array topological driving parameter model can dynamically describe the response behavior mode between pixel groups, providing a regional collaborative path for driving control; the spectral characteristics of the output image are detected and fed back in real time (such as color temperature, saturation, color gamut boundary offset), and the PWM dimming, pulse width and current amplitude of the sub-pixel are dynamically adjusted; the generated spectral regulation image output stream can better adapt to the color restoration degree required by the real picture; the inter-frame pixel tracking technology is introduced, and the pixel position change and motion trajectory between frames are identified by methods such as optical flow estimation; on this basis, the sub-pixel level light emitting trajectory is written back, and the light emitting behavior of high-speed motion or edge detail area is reconstructed; it can effectively eliminate the problems such as trailing, ghosting, mosaic and the like caused by pixel delay response in the motion image; from the perspective of display output, multiple optical indicators including brightness balance, color saturation offset, local contrast change and the like are extracted; the indicators are analyzed by nonlinear regression fitting algorithm through deep neural network or dynamic surface fitting algorithm, and the actual influence of driving parameters on visual results is identified; an iteratively updated driving state adjustment strategy model is constructed to realize the dynamic optimization of the pixel array; the closed-loop regulation ability of the display system is strengthened, which has the potential of self-learning and self-repairing, and effectively improves the picture quality stability and self-adaptive ability under long-term operation. The regional driving strategy is adopted, the display array is divided into several logical partitions, and local driving tasks are performed according to the visual consistency results and driving state strategy respectively; the dynamic load, current distribution and response speed of each region are customized for driving optimization; through the intelligent collaborative mechanism (such as task scheduling, power allocation, brightness compensation), efficient collaborative driving of the whole display panel is realized.

[0036] In the embodiment of the present application, referring to Figure 1 The steps of the full-color Micro LED micro display array driving method are shown in the flowchart, and in this example, the steps of the full-color Micro LED micro display array driving method include:

[0037] Step S1: multi-frequency full-pixel excitation test and pixel point excitation threshold sensitivity analysis are performed on the micro display array, so as to obtain the excitation threshold sensitivity of each pixel point;

[0038] In this embodiment, a micro display array is prepared to ensure it is in normal working condition. High-precision signal generators and measuring instruments such as oscilloscopes and photodetectors are used for excitation testing. These devices should be able to output multi-frequency signals and accurately measure the response of each pixel. Set the experimental parameters, the frequency range of the excitation signal is 100 Hz to 10 kHz, the pulse width is 100 µs, to ensure that a sufficient frequency range is covered, and the effect of different frequencies on pixel excitation is tested. The test is carried out in a stable experimental environment, and the temperature and humidity are controlled to reduce the influence of the external environment on the test results. Ensure that the lighting conditions are consistent to facilitate subsequent brightness and response measurements. Use a darkroom or light shielding device to ensure that the micro display array is not disturbed by external light sources during testing, thereby improving the accuracy of the measurements. Use a signal generator to generate a multi-frequency full-pixel excitation signal to ensure that the signal can act on all pixels of the micro display array simultaneously. Set up pulse signals of different frequencies to test the excitation response of the pixels. Set the signal generator output frequency to 1 kHz pulse signal, and adjust the frequency step by step (such as 2 kHz, 5 kHz, 10 kHz), record the excitation signal characteristics at each frequency. Apply the generated excitation signal to all pixel points of the micro display array. Ensure that each pixel is tested under the same excitation conditions to obtain consistent experimental data. While applying the signal, use an oscilloscope to monitor the quality and stability of the signal to ensure that the signal is not distorted, thereby ensuring the accuracy of the test results. While applying the excitation signal, use a photodetector or camera to monitor the light output response of each pixel in real time. Record the brightness change data of each pixel at different frequencies. For each frequency of the excitation signal, record the response intensity of each pixel and form a data set for subsequent analysis. Ensure that at least 100 frames of response data are recorded at each frequency to increase the reliability of the data. Define the excitation threshold as the minimum current or voltage value required for the pixel to start emitting light. By analyzing the response data at different frequencies, determine the excitation threshold of each pixel. Use statistical methods to analyze the relationship between the brightness response of each pixel and the applied current or voltage, plot the corresponding curve, and identify the threshold point. Calculate the excitation threshold sensitivity of each pixel point, defined as the degree of response of the pixel brightness to the change in excitation current or voltage. The following formula can be used: Sensitivity = (ΔBrightness / ΔCurrent). By performing linear regression analysis on the response data of each pixel under different excitation conditions, the sensitivity value of each pixel is calculated to ensure that the model has good fitting degree.

[0039] Step S2: identifying the pixel point linkage change of the micro display array, and adjusting the topology driving of the pixel array according to the excitation threshold sensitivity to construct the topology driving parameter of the pixel array;

[0040] In this embodiment, we are preparing to identify the linkage changes in the micro-display array. The entire process needs to be carried out in a controlled environment to minimize the impact of external factors such as light and temperature on pixel performance. Initialize the data acquisition device and set a high sampling rate (e.g. 1000 Hz) to ensure that the rapidly changing pixel response can be captured. Ensure that the system can process the brightness changes of each pixel in real time. Define the linkage change of the pixel as the brightness change of multiple adjacent pixels under the same excitation condition. The identification of linkage characteristics can be based on the light intensity difference between adjacent pixels, using a threshold method to determine whether there is a significant linkage change. Set a linkage change threshold (e.g. 10%), if the brightness change of a certain pixel exceeds the threshold and there is a related change with its adjacent pixels, it is considered that there is a linkage relationship between these pixels. When applying the excitation signal, use a photodetector to collect the real-time brightness change of each pixel and record the data of these changes. By comparing the brightness changes of adjacent pixels, identify the linkage changes. If the brightness of pixel A changes from 50 to 60 (increases by 20%), and the brightness of pixel B changes from 55 to 65 (increases by 18%), it can be judged that there is a linkage change between the two pixels. Use statistical analysis methods (such as correlation coefficient) to quantify the linkage strength. Record all identified linkage changes in the database to form a data table containing "pixel ID", "linkage change strength", "adjacent pixel ID". By analyzing these data, we can understand which areas have strong pixel linkage, thereby providing a basis for subsequent topological adjustment. Generate a linkage feature report to record all the pixels with linkage changes and their linkage strengths, helping to identify high-linkage areas. According to the identified linkage change characteristics, define the topological driving parameters. These parameters include the driving current, pulse width, and refresh rate of each pixel, ensuring that the driving of adjacent pixels is coordinated and consistent, avoiding display unevenness caused by improper driving. In high-linkage areas, the driving current of adjacent pixels can be set to the same value to ensure that they can emit light synchronously when excited. Combine the excitation threshold sensitivity with the linkage change characteristics to analyze the performance of each pixel when driven. If a pixel has high sensitivity and strong linkage, it needs to be paid special attention when driving, and its driving current or pulse width may need to be adjusted. Set a strategy: if the sensitivity of a pixel is greater than a certain threshold (e.g. 20%) and the linkage strength is high, the driving current of the pixel should be increased, such as from 1.5 mA to 2.0 mA. According to the above analysis, calculate the topological driving parameters of each pixel and optimize them. Ensure that the image can be displayed uniformly and stably during the excitation process, avoiding brightness unevenness or ghosting phenomenon. For a group of pixels with strong linkage, set their common driving current to 2.5 mA and adjust their refresh rate to 60 Hz uniformly to improve the display effect. After implementing the adjustment of the topological driving parameters, monitor the performance of each pixel in real time. Use a photodetector to obtain real-time brightness data to ensure that the adjusted driving parameters can effectively improve the display quality.The adjusted pixel brightness is continuously monitored to ensure that it remains consistent in dynamic scenes and any abnormalities are recorded for subsequent analysis.

[0041] Step S3: Real-time driving control is performed according to the pixel array topology driving parameters, and spectral feedback regulation is performed to obtain a spectral regulation image output stream.

[0042] In this embodiment, all hardware components of the microdisplay array, such as the driving circuit, controller, and photodetector, are ensured to work properly. According to the topological driving parameters obtained in step S2, the driving control system is initialized, and the driving current, pulse width, and refresh rate of each pixel are set. The driving current of the high-linkage area is set to 2.5 mA, and the refresh rate is 60 Hz, while the driving current of the low-linkage area is set to 1.5 mA, and the refresh rate is 30 Hz. These parameters need to be accurately configured in the driving control software to ensure that the system can respond quickly. A signal generator is used to generate a driving signal suitable for the microdisplay array, ensuring that the frequency and amplitude of the signal can cover the excitation requirements of all pixels. The frequency range of the signal is set to 1 kHz to 10 kHz, and the pulse width is set to 100 µs, ensuring that different characteristics of the pixels can be effectively excited. The stability and quality of the signal are monitored to ensure that the output signal has no distortion and avoids negatively affecting the display effect. The driving test is carried out in a controlled environment, ensuring that the temperature and humidity remain within a stable range. A temperature control device is used to monitor the environmental conditions to prevent interference with the spectral response and driving effect due to temperature changes. The laboratory temperature is set to 22°C, and the relative humidity is maintained at around 45% to ensure that the equipment operates in the best state. The driving control system is started, and the preset driving signal is applied to the microdisplay array. Ensure that the signal can be transmitted synchronously to the entire array to achieve full-pixel excitation. Monitor the response of each pixel to ensure that its brightness change during the driving process meets the expectations. Record the light output of all pixels at a certain time to ensure that all pixels can quickly reach a stable brightness within the specified time after the signal is applied. Use a photodetector or high-speed camera to monitor the spectral output of each pixel in real time. The monitoring frequency is set to 1000 Hz to quickly capture the response of each pixel and record the spectral data. During the monitoring process, record the spectral intensity of each pixel at different time points and ensure that the data can be fed back to the control system in real time for real-time adjustment. Analyze the real-time collected spectral data to extract the spectral features (such as wavelength, intensity, etc.) of each pixel. Use specialized spectral analysis software for data processing to ensure that the spectral features of each pixel can be accurately identified. Analyze the spectral data to identify the main wavelength and peak intensity to help judge whether the current driving effect meets the expected standard. According to the spectral analysis results, design a spectral feedback regulation strategy. When the spectral output of some pixels deviates from the preset standard, the system needs to automatically adjust its driving parameters (such as current or pulse width). Set a feedback mechanism that automatically increases the driving current of a pixel if its spectral intensity is detected to be lower than the set threshold (such as 80%). Implement the spectral feedback regulation strategy in the real-time driving control system to ensure that the system can quickly respond and adjust the driving parameters according to the real-time spectral data. Monitor the adjusted spectral output to ensure that it meets the expected standard. If the spectral intensity of a pixel is again below the threshold after adjustment, the system should automatically perform continuous adjustment to ensure the stability of the spectral output.According to the real-time driving result after feedback regulation, the spectral regulation image output stream is generated. The color and brightness of the output image can accurately reflect the result after spectral regulation. The characteristic data of the generated spectral regulation image is recorded and compared with the previous image to verify the effectiveness of the regulation effect.

[0043] Step S4: Inter-frame pixel movement tracking and sub-pixel level light emission trajectory writing are performed on the spectral regulation image output stream to obtain a visual consistency image output stream.

[0044] In this embodiment, ensure that the high-resolution camera or high-speed video camera can capture the spectral control image output stream at a rate of at least 120 frames per second. Set the image resolution to 1920x1080 to capture the details of each pixel. Before the experiment, calibrate the camera to ensure that the imaging quality meets the expectations. Check the color reproduction ability using standard color cards and ensure that the lens is clean to reduce image distortion. Choose a suitable inter-frame pixel movement tracking algorithm, such as the optical flow method or feature point matching algorithm (e.g., Lucas-Kanade optical flow method). These algorithms can effectively identify the pixel displacement in consecutive frames. Before implementation, set the algorithm parameters, such as window size and pyramid layer number, to ensure that the algorithm can work effectively at different motion speeds. A typical setting is a 5x5 window and 3 layers of pyramid. When collecting images, perform preliminary processing, including noise removal and contrast enhancement. Use a Gaussian filter to remove low-frequency noise to ensure the accuracy of subsequent analysis. Apply a 3x3 Gaussian filter to each frame of image to smooth the image and reduce noise interference, thereby improving the effect of the movement tracking algorithm. Start the image acquisition system to obtain the continuous frames of the spectral control image output stream in real time. Ensure that the image acquisition system can run stably throughout the process to avoid frame loss caused by external factors. Record the timestamp of each frame to accurately correspond to the displacement of each pixel during subsequent analysis. Use the selected optical flow algorithm to process consecutive frames to calculate the displacement of each pixel between adjacent frames. The algorithm will analyze the brightness change of each pixel in the image to estimate its motion trajectory. If a pixel is located at (x1, y1) in frame 1 and (x2, y2) in frame 2, the displacement can be calculated as Δx = x2 - x1 and Δy = y2 - y1, forming the motion vector of each pixel. According to the inter-frame pixel movement tracking results, design a sub-pixel level light emission trajectory rewriting strategy. The goal is to reduce the smearing and blurring effects caused by motion by fine-tuning the light emission position of the pixels. Set a rewriting precision (e.g., 0.1 pixels) to ensure that fine adjustments can be made when rewriting the light emission trajectory. Such precision can effectively improve visual consistency. Adjust the light emission position of each pixel based on the previously recorded motion vector of each pixel. By fine-tuning the light intensity of each pixel, ensure that the visual effect of the image remains consistent during motion. If a pixel has a displacement of Δx = 0.2 and Δy = 0.1, adjust the light emission position of the pixel by 0.2 units in the corresponding direction and adjust the brightness to compensate for the light intensity change caused by motion. Generate a new visual consistency image output stream and monitor the adjusted effect in real time. Use a photodetector to monitor the brightness of each pixel to ensure that it can maintain high consistency in dynamic scenes. Compare the adjusted image with the original image to ensure that the visual effect is improved in dynamic scenes, especially in areas with faster motion.

[0045] Step S5: Multi-index change deep fitting is performed on the visual consistency image output stream, and adaptive correction is performed to construct an adaptive driving state adjustment strategy;

[0046] In this embodiment, the relevant data of the visually consistent image output stream obtained in the previous step is ensured, including the brightness, color information and displacement data of each pixel. These data are the basis for multi-index deep fitting. Organize the data set and divide the data into training set and test set. The training set contains a large number of samples (such as 5000 frames of images), and the test set is used for subsequent verification of the fitting effect. Ensure that the data covers different scenes and dynamic changes to improve the generalization ability of the model. Determine the multiple indexes that need to be fitted, including brightness uniformity, color saturation, dynamic response time and energy consumption, etc. The selection of each index should be based on its degree of influence on the display effect. Brightness uniformity can be measured using standard deviation, color saturation can be calculated by the ratio of the average value to the maximum value of the RGB channel, and dynamic response time can be obtained by recording the delay time of image update. Select an appropriate deep learning model for multi-index fitting. Consider using a multi-layer perceptron (MLP) or a convolutional neural network (CNN), and select the appropriate network structure according to the data characteristics. Set the model parameters, such as learning rate (initially set to 0.001), batch size (such as 32), and training period (such as 100 cycles), to ensure that the model can converge within a reasonable time. Use the training set data to train the selected deep learning model. The input data includes the brightness, color and displacement information of each pixel, and the output is the corresponding multi-index change value. During the training process, use cross-entropy loss function or mean square error loss function to evaluate the performance of the model, and regularly monitor the training loss and validation loss to ensure that the model gradually converges. Use the pre-prepared test set to verify the trained model and evaluate the model's performance on new data. Focus on the model's prediction ability for different indexes. Calculate the mean square error (MSE) of the model on the test set, and if the MSE value is lower than the set threshold (such as 0.01), the model is considered to have good fitting ability. Record the fitting results of the model, including the difference between the predicted value and the actual value of each index, and analyze the performance of the model. Identify the shortcomings of the model in some indexes, and provide the basis for subsequent adjustment. If the difference between the predicted value and the actual value of the brightness uniformity is large, further analysis is needed to consider whether there is a problem of data imbalance or improper feature selection. According to the fitting results, design an adaptive correction strategy. This strategy should be able to dynamically adjust the driving parameters (such as current, pulse width, refresh rate) according to the model's predicted index changes. When the model predicts that the brightness uniformity of a certain area is lower than the set threshold (such as 0.8), automatically increase the driving current of that area to improve the brightness. Design a feedback control path to enable the system to monitor the image output in real time and compare it with the model's prediction to make automatic adjustments. Set a feedback cycle to check the difference between the current brightness output and the model's predicted value every 100 milliseconds, and if the difference exceeds the set threshold, trigger the adaptive adjustment. Implement the adaptive correction strategy into the driving control system to ensure that the system can respond and adjust the driving parameters in real time.The effect after adjustment is monitored to ensure that the display quality is improved. The luminance and color after adjustment are monitored in real time using a photodetector to ensure that they perform well in dynamic scenes with good consistency and stability.

[0047] Step S6: Perform partition intelligent driving control on the visual consistency image output stream, and perform intelligent collaborative driving control based on an adaptive driving state adjustment strategy to execute micro display array driving operations.

[0048] In this embodiment, before implementing the partitioned intelligent driving control, the hardware system of the micro display array is first configured to ensure that all driving modules and controllers are working normally. According to the previous analysis, the micro display array is divided into multiple independent regions (such as high, medium, and low regions), and each region is classified according to its content characteristics and dynamic characteristics. The driving parameters of each region are set, including current, pulse width, refresh rate, etc. The driving current of the high dynamic region is set to 2.5 mA, and the refresh rate is 60 Hz; the medium dynamic region is set to 2.0 mA, and the refresh rate is 45 Hz; the low dynamic region is set to 1.5 mA, and the refresh rate is 30 Hz. The partitioned driving control algorithm is designed to ensure that each region can be independently driven while maintaining consistency as a whole. The algorithm needs to respond to content changes in each region in real time and dynamically adjust the driving parameters. A feedback mechanism is set up to check the brightness and color output of each region regularly (such as every 100 ms) and adjust the driving current according to the preset standards. Ensure that the partitioned driving control is carried out in a stable experimental environment, control the temperature and humidity to reduce the influence of the external environment on the test results. Set the laboratory temperature to 22°C and the relative humidity to about 45% to ensure that the equipment operates in the best state. Start the partitioned driving control system and apply the preset driving parameters to each partitioned region. Different driving signals are transmitted to each region through the controller to ensure that each region can operate independently according to the set parameters. First, apply 2.5 mA current and 60 Hz refresh rate to the high dynamic region to ensure that the region can quickly respond to dynamic content. While the partitioned driving control system is running, monitor the brightness and color output of each region in real time and use a photodetector to collect data. Set the sampling frequency to 1000 Hz to ensure that changes in each region can be captured quickly. If the brightness of the high dynamic region is found to be lower than the set value, the system will automatically increase the current of that region to ensure that the brightness reaches the standard. Record the real-time monitoring data and spectral data of each region in the database for subsequent analysis. Ensure that the data is complete and has timeliness to support intelligent collaborative driving control. Record the current, brightness, color output, and other information of each region to form a detailed data log for subsequent optimization and analysis. On the basis of partitioned intelligent driving control, apply adaptive driving state adjustment strategies. According to the real-time monitoring data and model prediction results, dynamically adjust the driving parameters of each region to ensure the consistency of the overall display effect. If the brightness uniformity of a region is less than 0.8, the system will automatically adjust the driving current of that region to 2.0 mA to improve display quality. Design an intelligent collaborative control mechanism to ensure that the driving of each region is coordinated and consistent. When the brightness of a region increases, the brightness of adjacent regions should also be adjusted accordingly to avoid visual discomfort caused by brightness differences.In implementation, a neighboring area feedback mechanism is set, if the brightness of a certain area changes more than a set threshold (such as 10%), the driving parameters of the neighboring areas should also be adjusted accordingly. By continuously monitoring the output of each area, the system can real-time feedback and adjust the driving state of each area. The feedback period is set to 100 ms, ensuring that the system can quickly respond to dynamic changes. If the brightness of a high dynamic area increases sharply at a certain moment, the system should be able to adjust the brightness of the medium and low dynamic areas in real time to maintain the visual consistency of the overall picture.

[0049] In this embodiment, refer to Figure 2 For the detailed implementation procedure of step S1, in this embodiment, the detailed implementation steps of step S1 include:

[0050] Define a multi-cycle pulse micro-excitation current value; based on the multi-cycle pulse micro-excitation current value, perform multi-frequency full-pixel excitation test on the micro display array, and extract the excitation response parameters of each pixel point of the micro display array;

[0051] Calculate the charge response characteristics of each pixel point excitation response parameter;

[0052] According to the charge response characteristics, calculate the brightness intensity of each pixel point to generate pixel point brightness intensity of different cycle pulses;

[0053] Perform multi-cycle brightness nonlinear relationship change analysis on the pixel point brightness intensity to obtain the pixel point change rule of different cycles;

[0054] According to the pixel point change rule of different cycles, analyze the excitation threshold sensitivity of each pixel point, so as to obtain the excitation threshold sensitivity of each pixel point.

[0055] In this embodiment, the specific parameters of the multi-cycle pulse micro-excitation current value are determined, including pulse amplitude, frequency, and duration. Suitable current values are selected to ensure that the micro-display array is not damaged while effectively exciting each pixel. The pulse amplitude is set to 1 mA, the frequency is 1 kHz, and the duration is 100 µs. Such parameter design aims to ensure effective excitation of the micro-display array while avoiding thermal damage caused by excessive current. The required multi-cycle pulse signal is generated using a signal generator, and the repetition period and duty cycle of the pulse are set. The stability and accuracy of the signal are ensured to accurately reflect the response of each pixel in subsequent tests. The repetition period is selected as 10 ms, and the duty cycle is 10% to ensure sufficient spacing between each pulse to facilitate cooling and stable response of the system. Before conducting the experiment, the test equipment is calibrated to ensure synchronization and accuracy between the signal generator, oscilloscope, and micro-display array. Multiple tests are conducted to confirm the reliability of the equipment and the accuracy of the measurement results. The oscilloscope is used to monitor the pulse current signal to ensure that the output current matches the set value and to record any deviations. The micro-display array is connected to the excitation signal source to ensure that each pixel can be independently excited. The test program is set to excite each pixel point one by one and record its response data. Each pixel in the array is sequentially excited, and the appearance time and intensity response of each pixel point are recorded to ensure that the excitation signal is completely transmitted to each pixel. High-precision measuring instruments such as photodetectors or cameras are used to collect the excitation response parameters of each pixel, and the light output intensity of each pixel versus time is recorded. The accuracy and completeness of the collected data are ensured. The response intensity of each pixel under different excitation currents is recorded to form a data set for subsequent analysis. The collected excitation response data is sorted and preprocessed to remove noise and outliers, ensuring the quality of the data for subsequent analysis. The data is standardized for comparison. The response data is smoothed using a mean filter to ensure that the response data of each pixel is representative in statistics. According to the excitation response data, the charge response characteristics of each pixel point are calculated to quantify the degree of response of the pixel to the excitation current. This process usually involves integration or peak detection methods. The current value of the response peak of each pixel is calculated, and its charge characteristics are calculated to form a charge response characteristic matrix. The extracted charge response characteristics are standardized for subsequent analysis. The response characteristics of different pixel points can be effectively compared. The charge response characteristics of each pixel are normalized to the range of 0 to 1 to facilitate comparison of the sensitivity of different pixels. Based on the charge response characteristics, appropriate algorithms are selected to calculate the brightness intensity of each pixel point. Common methods include linear mapping and nonlinear mapping. A linear relationship between brightness intensity and charge response characteristics is set, and a formula is used for calculation. For each pixel point, the corresponding brightness intensity is calculated based on its charge response characteristics. Ensure that the characteristics and excitation conditions of each pixel point are considered in the calculation process.If the charge response of a pixel is 0.5, its luminance intensity might be calculated as 50% of the display luminance. A suitable statistical method (such as polynomial fitting or curve fitting) is chosen to analyze the nonlinear relationship between luminance intensity changes and current excitation over different periods. A quadratic polynomial is used for fitting to capture the nonlinear relationship between luminance intensity and excitation current. Luminance intensity data from different periods are fitted, and their variation patterns are analyzed to identify nonlinear characteristics and trends. The luminance intensity changes under different periods are analyzed to determine if there is a saturation effect or nonlinear response characteristics, and the fitting results and correlation coefficients are recorded. An excitation threshold is defined for each pixel, typically the minimum current value required for the pixel to begin emitting light. The excitation threshold is determined based on the relationship between luminance intensity and excitation current. By analyzing the relationship between luminance intensity changes and current, a specific current value is identified as the excitation threshold. The sensitivity is calculated based on the excitation threshold of each pixel. Sensitivity is typically defined as the rate of change in luminance caused by a change in excitation current. If the excitation current of a pixel increases from 1 mA to 1.5 mA, and the luminance intensity increases from 10% to 30%, then its sensitivity is 40% / 0.5 mA. The excitation threshold sensitivity of each pixel is recorded in a database, generating a corresponding analysis report to help further optimize the driving strategy of the microdisplay array. A data table containing "pixel ID," "excitation threshold," and "sensitivity" is generated to facilitate subsequent performance optimization and design improvements.

[0056] In this embodiment, see Figure 3 The diagram below illustrates the detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include:

[0057] The brightness intensity of pixels with different period pulses is used to identify the brightness difference between neighboring pixels, and the brightness difference feature of neighboring pixels is obtained.

[0058] Pixel linkage change recognition is performed based on the brightness difference characteristics of neighboring pixels, and pixels with linkage brightness changes are extracted;

[0059] The pixels with brightness-linked changes are dynamically divided into pixel regions to obtain multiple pixel-related subnets.

[0060] Based on the excitation threshold sensitivity, the topology driving adjustment between pixel arrays is performed on the associated subnets of multiple pixels to construct pixel array topology driving parameters.

[0061] In this embodiment, the brightness intensity data of each pixel under different periodic pulse excitation is collected and organized. The integrity and accuracy of the data are ensured, and it is organized into a two-dimensional matrix for subsequent analysis. An 8x8 pixel array is set up to record the brightness intensity of each pixel under a specific pulse period, forming a data matrix containing 64 elements. The brightness difference is calculated for each pair of adjacent pixels (e.g., vertically or horizontally adjacent pixels). The absolute difference is used to quantify the brightness difference between adjacent pixels. If the brightness of pixel A is 80 and the brightness of pixel B is 70, then the brightness difference between them is |80-70|=10. The brightness differences of all adjacent pixels are recorded in a data table. Based on the calculated brightness difference values, the brightness difference features of adjacent pixels are extracted. A threshold can be set to identify pixel pairs with significant brightness differences. A brightness difference threshold of 5 is set; if the brightness difference of a pair of pixels exceeds this value, it is marked as "significant difference," and their position information is recorded. The characteristics of pixel linkage changes are defined, which typically refer to the ability of brightness changes in some pixels to cause corresponding changes in the brightness of other adjacent pixels under specific excitation conditions. If an increase in the brightness of a pixel causes a corresponding increase in the brightness of its surrounding pixels, this can be considered a linked change. Using the previously extracted brightness difference features, analyze which pixels exhibit linked brightness changes. Pixels with strong linkages can be identified by calculating correlation coefficients or using clustering analysis. Use the Pearson correlation coefficient to calculate the correlation of brightness changes between adjacent pixels; if the correlation coefficient is higher than 0.7, the pixels are considered to have significant linked changes. Record all identified pixels with linked brightness changes to form a list of linked change pixels. These pixels will be used for subsequent dynamic region partitioning. Record the pixel IDs, brightness change values, and neighboring pixel information of linked change pixels to form a dataset containing all linked change features. Based on the pixel information of linked brightness changes, select an appropriate method for dynamic pixel region partitioning. Common methods include cluster-based region partitioning (such as K-means clustering) or graph theory methods. Use K-means clustering to divide the linked change pixels into multiple regions to facilitate the analysis of their common features and interrelationships. Perform dynamic region partitioning on the pixels with linked brightness changes to form multiple pixel association subnets. Each subnet contains pixels with significant interconnected changes. If pixels in a certain region exhibit similar brightness change characteristics under excitation, these pixels are grouped into the same subnet, forming a pixel region. A topology-driven adjustment strategy is designed based on excitation threshold sensitivity. This strategy aims to optimize the driving parameters of the pixel array to improve overall display performance and response speed. If some pixels have low excitation thresholds, they can be connected to surrounding pixels with larger brightness changes to enhance their interconnectedness. Based on the previously collected excitation threshold sensitivity data, the driving parameters for each pixel under the new topology are calculated.This can be achieved by dynamically adjusting the voltage or current. For each pixel in the subnet, the required current value is calculated to ensure that the expected brightness is achieved under excitation without exceeding the excitation threshold. After implementing the topology-driven adjustment, the response of the pixel array is monitored to evaluate the effect of the adjustment. It is ensured that the new driving parameters improve overall performance and reduce linkage inconsistencies. The uniformity of pixel brightness and response speed after adjustment are monitored, and data on the improvement are recorded to ensure the effectiveness of the optimization.

[0062] In this embodiment, the specific steps for adjusting the pixel array topology driving parameters by performing inter-pixel array topology driving adjustment on multiple pixel point associated subnets based on the excitation threshold sensitivity are as follows:

[0063] Define a preset standard brightness value;

[0064] Based on the excitation threshold sensitivity and the preset standard brightness value, an adaptive driving current amplitude is designed to obtain the adaptive driving current amplitude for each pixel.

[0065] Calculate the pixel linkage response period of the subnet region for multiple pixel points associated with subnets, and generate the linkage response period of different pixel subnets;

[0066] The current drive cycle is adjusted according to the linkage response cycle to generate a current drive cycle;

[0067] The topology driving parameters of the pixel array are constructed by adjusting the adaptive driving current amplitude and current driving period between the pixel arrays.

[0068] In this embodiment, a preset standard brightness value is determined as the target brightness of the microdisplay array. This value should be set according to application requirements and user expectations, and is usually expressed in a standard unit (such as nits). The standard brightness value is set to 300 nits, which will serve as the basis for subsequent drive design, ensuring that the display effect meets expectations. The brightness response characteristics of all pixels are compared with the preset standard brightness value to ensure that the brightness output of each pixel reaches or approaches this standard. This process helps to evaluate the excitation requirements of each pixel. If the current brightness of a pixel is 250 nits, its drive current amplitude needs to be determined to achieve the target brightness of 300 nits. Based on the excitation threshold sensitivity of each pixel, its responsiveness to changes in drive current is analyzed. This sensitivity will affect the design of the adaptive current amplitude. If the excitation threshold sensitivity of a pixel is 20% / mA, it indicates that its brightness is very sensitive to changes in current. Based on the preset standard brightness value and excitation threshold sensitivity, the adaptive drive current amplitude of each pixel is calculated. This calculation can be achieved by setting a linear or non-linear relationship. Assuming a linear relationship between standard brightness and current, if the current driving current is 1 mA, the required current amplitude can be calculated using the formula: Required current = Current current × (Standard brightness / Current brightness). The pixel linkage response period is defined as the time period required for a pixel group to reach a specific brightness change under specific driving conditions. This period can be used to optimize the driving strategy. The response period is set as the time required from the excitation of a pulse signal to the stable output of all associated linked pixels reaching the target brightness. Period calculations are performed for each pixel subnet, its response time is recorded, and the linkage response characteristics of different subnets are analyzed. The brightness change of each pixel can be recorded using timestamps. If the time required for multiple pixels in a subnet to reach stable brightness under excitation is 50 ms, this time is taken as the linkage response period of that subnet. An adjustment strategy for the current driving period is designed based on the linkage response periods of different subnets. This ensures that the periods of each subnet are coordinated during the driving process to improve the display effect. The driving period is set to 1.5 times the linkage response period to ensure that there are no brightness fluctuations during the response process. Based on the previously calculated linkage response period, a new current driving period is calculated for each subnet. Ensure the calculated cycle balances display performance and power consumption. If the linkage response cycle of a subnet is 50 ms, the current drive cycle can be set to 75 ms. Design the topology drive parameters of the pixel array based on the adaptive drive current amplitude and current drive cycle. These parameters should optimize the drive efficiency of the entire array. Set the proportional relationship between the drive current of each pixel and the drive current of its neighboring pixels to ensure uniform brightness output during excitation. Perform detailed calculations on the topology drive parameters of each pixel and subnet to ensure effective application in actual driving. Consider the mutual influence of current between pixels to ensure the calculated parameters are reasonable.If the adaptive current amplitude of a pixel is 2 mA, then the current of its adjacent pixels should be adjusted according to the set ratio.

[0069] In this embodiment, see Figure 4 The diagram below illustrates the detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include:

[0070] The micro-display array is driven and controlled in real time based on the pixel array topology driving parameters, and the image output stream of the display array is acquired.

[0071] Ultrashort time window interferometric sampling is performed on the output stream of the display array image to extract ultrashort time window spectral sampling information;

[0072] Identify the underlying image spectral distribution of the ultrashort time window spectral sampling information;

[0073] The spectral fluctuation frequency of the basic image spectral distribution is calculated, and frequency modeling is performed to generate the image spectral fluctuation curve;

[0074] The dynamic spectral distribution shift of the image spectral fluctuation curve is calculated at high speed to generate dynamic spectral distribution shift data.

[0075] The display array image output stream is spectrally controlled by dynamic spectral distribution offset data to obtain a spectrally controlled image output stream.

[0076] In this embodiment, a real-time drive control system based on pixel array topology drive parameters is deployed to ensure that each pixel can be precisely controlled according to the set drive parameters. This system should be able to respond quickly to user input and changes in the external environment. The drive current for each pixel is set to 2 mA to ensure that the pixels can exhibit the expected brightness and color under different display conditions. A high-precision image acquisition device (such as a high-speed camera or optical sensor) is used to capture the image output stream of the microdisplay array in real time. The acquisition frequency is ensured to be high enough to capture rapidly changing image information. The acquisition frequency is set to 1000 frames / second to obtain a smooth image output stream and record the color and brightness information of each frame. An ultra-short time-window interferometric sampling scheme is designed to capture spectral information in a very short time. An appropriate sampling window is selected to ensure that effective spectral data can be extracted. The sampling window is set to 1 ms to capture the instantaneously changing spectral features in the image output stream. An interferometer or spectrometer is used to perform spectral analysis on the acquired images to extract spectral information within the ultra-short time window. The extracted spectral data is ensured to have high resolution and accuracy. The sampled data is processed using Fourier transform techniques to obtain the spectral information of each pixel in the frequency domain, which is then recorded in a database. The extracted spectral data is analyzed to identify the spectral distribution characteristics of the base image. Statistical methods (such as mean and variance) can be used to quantify the spectral features. The average light intensity at each wavelength is calculated, and the main peak positions of the spectral distribution are identified. The identified spectral distribution characteristics of the base image are recorded in the database, forming a dataset of spectral features for subsequent frequency calculations and modeling. The light intensity, peak position, and width at each wavelength are recorded to form a comprehensive spectral description. A suitable frequency analysis method (such as Fast Fourier Transform, FFT) is selected to calculate the frequency of the base image's spectral distribution. It is ensured that the main frequency components of the spectral fluctuations can be identified. The extracted spectral data is processed using FFT to obtain its frequency domain representation, and the frequency components are identified. Based on the calculated frequency information, an image spectral fluctuation curve is constructed. This curve should reflect the influence of different frequency components on the image output. A fluctuation curve is generated to show the changes in light intensity at different frequencies, helping to understand the dynamic characteristics of the image. A suitable algorithm is selected to dynamically shift the spectral fluctuation curve to identify changes in spectral distribution under high-speed output conditions. The Dynamic Time Warping (DTW) algorithm is employed to compare spectral distributions across different time periods and identify their shifts. Based on the spectral fluctuation curve, the spectral distribution shift for each time period is calculated, generating dynamic spectral distribution shift data. The shift amount and its impact at each time point are recorded. The displacement of the main spectral peaks at specific time points is recorded and quantified for subsequent analysis. The dynamic spectral distribution shift data is recorded in a database, and a detailed analysis report is generated for subsequent spectral feedback and control.A data table containing "time points" and "spectral offsets" is created to facilitate subsequent optimization and adjustments. Based on the dynamic spectral distribution offset data, a spectral feedback control strategy is designed to automatically adjust the image output stream, ensuring that the output spectral distribution meets the expected standards. A feedback loop is set up to automatically adjust the pixel drive current when an offset exceeds a certain threshold. The spectral feedback control strategy is implemented in the real-time drive control system, and its impact on the image output stream is monitored. It is ensured that adjustments are reflected in the output image in a timely manner. The spectral distribution of the output image is monitored in real time to ensure that it meets standard brightness and color requirements.

[0077] In this embodiment, step S4 includes the following steps:

[0078] The spectral-controlled image output stream is decomposed into continuous image frames to extract time-series image frames.

[0079] Perform inter-frame pixel motion tracking on time-series image frames to extract the pixel motion trajectory of consecutive frames;

[0080] Motion vector estimation is performed on the pixel movement trajectory of consecutive frames, and the motion vector values ​​of multiple pixels are extracted.

[0081] Image blur detection by pixel movement is performed on the image output stream of the display array to generate image dispersion and trailing information;

[0082] Based on the image dispersion and trailing information, subpixel-level luminous trajectory rewriting is performed on the motion vector values ​​of multiple pixels to obtain a visually consistent image output stream.

[0083] In this embodiment, the spectrally modulated image output stream is collected to ensure high quality and high frame rate. A high-speed camera or high-performance image acquisition device is used, and the acquisition frequency is set to ensure the integrity of continuous image frames. The acquisition frequency is set to 120 frames / second to ensure that rapidly changing image output is captured for subsequent analysis. The continuous image output stream is decomposed into frames to extract time-series image frames. This process can be implemented using software tools to ensure that each frame can be saved independently for easy subsequent processing. Image processing software is used to convert the continuous image stream into single-frame image files, forming a set of time-series image data for subsequent analysis. The extracted time-series image frames are organized in chronological order and stored in a database, ensuring that the timestamp of each frame is clearly recorded for subsequent motion analysis. A suitable motion tracking algorithm, such as optical flow or feature point matching algorithm, is selected to ensure effective identification of pixel movement between frames. These algorithms can quantify the displacement of pixels between different frames. The Lucas-Kanade optical flow method is used, which can effectively handle pixel movement within a small range and is suitable for the analysis of high frame rate images. The selected motion tracking algorithm is applied to each pair of adjacent time-series image frames to extract the movement trajectory between pixels. Record the displacement of each pixel between frames. Calculate the displacement of each pixel between adjacent frames and record it as a dataset of "pixel position," "frame number," and "displacement amount." Visualize the extracted pixel movement trajectory to help understand the dynamic changes of pixels in the time series. This can be achieved by drawing vector graphics or trajectory diagrams. Based on the inter-frame pixel movement data, use vector estimation methods to calculate the motion vector of each pixel. The motion vector represents the displacement and direction of the pixel between image frames. Calculate the magnitude and direction of the motion vector using the pixel displacement data and record it as a dataset of "vector magnitude" and "vector direction." Record the calculated motion vector values ​​in a database for subsequent image blur detection and visual consistency adjustment. Generate a data table containing "pixel ID," "motion vector X," and "motion vector Y" for subsequent analysis and optimization. Choosing a suitable blur detection algorithm can effectively identify blurred areas and chromatic aberration in the image. Commonly used methods include gradient-based detection algorithms. Use the Laplacian operator or Sobel operator to process the image and calculate the gradient value of each pixel to identify blurred areas. Extract chromatic aberration information from the image using the blur detection algorithm. This process requires analyzing each pixel of the image to identify blurred areas affecting visual quality. The blur level and motion blur information of each pixel are recorded, forming a dataset containing "pixel ID" and "blur level". Based on the extracted motion vector values ​​and dispersive motion blur information, a sub-pixel-level emission trajectory write-back strategy is designed to ensure visual consistency. The emission position of each pixel is fine-tuned according to the motion vector to achieve a smoother visual output effect.Subpixel-level brightness adjustment is performed on each pixel to ensure that the emission position can be dynamically adjusted during motion to eliminate motion blur. Fine-tuning of the brightness of each pixel is made based on the motion vector, for example, increasing the brightness of a pixel from 100 to 105 to compensate for motion-induced blurring. A final visually consistent image output stream is generated based on the adjusted emission trajectory, and its performance in real-world applications is monitored. The adjustments are ensured to effectively improve image quality. The final image output stream is generated and its quality is evaluated to check if the expected visual effect has been achieved.

[0084] In this embodiment, step S5 includes the following steps:

[0085] Perform multi-frame image color recognition on the visually consistent image output stream and extract color features from the multi-frame images;

[0086] Calculate the real-time brightness values ​​of multiple pixels in the visually consistent image output stream;

[0087] The brightness decay rate of the pixels is calculated by performing a brightness decay rate calculation on the real-time brightness values ​​of the multiple pixels.

[0088] Identify the pixel heating status of the microdisplay array; perform deep fitting of multiple index changes based on the color features of the multi-frame images, pixel brightness decay rate, and pixel heating status to obtain a driving state deviation map;

[0089] Based on the driving state deviation map, the pixel array topology driving parameters are dynamically corrected by time window and current driving pulse width, and an adaptive driving state adjustment strategy is constructed.

[0090] In this embodiment, a high-speed camera is used to continuously sample the visual consistency image output stream at a frame rate of 120 frames per second to capture rapidly changing color information. To ensure image quality, appropriate exposure time and gain settings are selected to reduce noise interference. The acquired images are preprocessed, including denoising, contrast enhancement, and equalization. Denoising can be achieved using methods such as median filtering, while contrast enhancement can be achieved through histogram equalization to highlight color features. Each frame is converted to the HSV color space, which is more suitable for color analysis. The main color features are extracted by calculating the average hue, saturation, and brightness of each frame. For each frame, the number of pixels in different color regions is counted to identify the distribution of the main colors, ensuring that the extracted data reflects the color features of the image. When identifying color features, relative color changes are considered rather than absolute values; color drift or inconsistencies are detected by comparing color differences between adjacent frames. Euclidean distance is used to calculate the differences in color features between adjacent frames to ensure that minute changes are captured. Real-time brightness measurements are performed on multiple pixels in the visual consistency image output stream. Using a high-precision photometer or image analysis software, the luminance value is obtained by calculating the RGB value of each pixel. The luminance value can be calculated using the formula Y=0.299R+0.587G+0.114B. The measurement cycle is set to once per frame of image acquisition to ensure the real-time nature and accuracy of the luminance value. The luminance value of each pixel needs to be dynamically updated during image processing. For each pixel, its luminance at a specific time point is recorded and compared across frames to analyze the trend of luminance change. By calculating the real-time luminance value, the magnitude of luminance value change is identified, with particular attention paid to pixels with significant luminance decay. This data will provide the basis for subsequent luminance decay rate calculation. The luminance decay rate is defined as the magnitude of pixel luminance change over a certain period of time. The calculation formula is: Decay Rate = (Initial Luminance - Current Luminance) / Initial Luminance × 100%. An appropriate time window (1 second) is selected to monitor luminance changes. By monitoring the initial and current luminance of each pixel in real time, the decay rate of each pixel is calculated. Pixels with high decay rates are given special attention to analyze possible causes, such as overdrive or heat accumulation. Set a threshold (e.g., 10%) to highlight problematic pixels to facilitate subsequent optimization strategies. Monitor the brightness decay rate of each pixel over time, analyzing for specific patterns, such as a significant increase in decay rate during certain time periods, indicating potential thermal issues or driver instability. Use an infrared thermal imager to monitor pixel temperature, ensuring accurate capture of the heat generation of each pixel. Infrared thermal imagers provide high-resolution temperature distribution maps to help identify heat-generating areas. Set a monitoring temperature threshold, such as 70°C; any pixel exceeding this temperature requires special attention. Record pixel temperature changes in each image frame, especially the number and location of high-heat-generating pixels.By analyzing heat distribution, we can identify areas where performance degradation or failure may occur due to prolonged high temperatures. Cross-analysis of heat data and brightness decay rate is performed to identify whether high-heat areas correspond to pixels with high brightness decay rates, thus identifying potential causal relationships and helping to optimize the driving strategy. A multivariate regression model is designed, with color features, brightness decay rate, and heat status as independent variables, and driving state deviation as the dependent variable. The model is designed to capture the combined impact of multiple factors on the driving state. An appropriate fitting method, such as least squares, is selected to ensure good model fit and reliable predictions. Historical data is used to train the model, ensuring it effectively reflects driving characteristics under different states. Cross-validation is used to evaluate the model's generalization ability, ensuring accuracy on new datasets. Using the trained model, the driving deviation for the current state is calculated, generating a driving state deviation map. This map visually displays the deviation of the driving state of different pixels from the ideal state, helping to quickly identify problem areas. Based on the driving state deviation map, a dynamic time window strategy is designed to flexibly adjust driving parameters under different states. By analyzing the deviation map, we can identify time periods that require special attention or adjustment. If the deviation value remains consistently high over a certain period, the time window can be shortened to allow for more frequent adjustments to the drive parameters. The current drive pulse width is dynamically adjusted based on real-time monitoring of heat generation and brightness decay rate. If the number of high-heat pixels increases, the pulse width is shortened promptly to reduce power consumption and heat generation. Specific rules are established, such as shortening the pulse width from 200 µs to 150 µs when the temperature exceeds 70°C, thereby reducing heat output. After implementing the adaptive drive state adjustment strategy, the adjustment effect is monitored in real time to ensure that pixel brightness and temperature remain within acceptable ranges. The effectiveness of the strategy is verified by observing changes in brightness and heat after adjustment. If pixels still exhibit excessive heat generation or unsatisfactory brightness after adjustment, further optimization of the adjustment strategy or consideration of hardware improvements is necessary.

[0091] In this embodiment, step S6 includes the following steps:

[0092] Perform deep semantic analysis on the visually consistent image output stream to identify the semantic features of different regions.

[0093] Image content is evaluated based on the image semantic features to generate multi-region content attention levels;

[0094] The image refresh rate is differentiated based on the content attention of different regions to generate different image refresh rates for different regions;

[0095] Based on the image refresh rate of different regions, partitioned intelligent drive control is performed to generate energy-efficient partitioned drive strategies.

[0096] Intelligent collaborative drive control is implemented based on energy consumption optimization partitioning drive strategy and adaptive drive state adjustment strategy to execute micro-display array drive operations.

[0097] In this embodiment, a high-resolution camera is used to acquire the visually consistent image output stream, ensuring image clarity and detail richness. The acquisition resolution is set to 1920x1080 to capture sufficient image information. The acquired images are preprocessed, including noise reduction, contrast enhancement, and color correction. Gaussian filtering is used to remove noise, and histogram equalization is applied to improve image contrast, laying the foundation for subsequent semantic analysis. A deep learning model (such as a convolutional neural network CNN) is used to perform semantic segmentation on the preprocessed images to identify the semantic features of different regions. The model is trained using labeled datasets (such as Pascal VOC or COCO) to ensure good recognition capabilities. The model infers from the images, generating category labels for each pixel, identifying the main objects in the image (such as people, backgrounds, objects, etc.), and recording the semantic features of each region. Based on the semantic segmentation results, the image is divided into regions, and the features of each region (such as area, shape, color, etc.) are extracted. The features of each region are recorded in a data table, providing basic data for subsequent evaluation. Based on the identified regional characteristics, content evaluation metrics are set, including visual importance, dynamism, and user attention. These metrics will be used to generate the content attention score for each region. The dynamism metric is defined as the number of moving objects within the region, while visual importance can be measured by the region's size and color saturation. Based on the set evaluation metrics, the content of each region is comprehensively evaluated to generate multi-region content attention scores. A weighted algorithm is used to assign weights according to the importance of different metrics. Dynamic regions (such as areas with human activity) are given higher weights, while static regions (such as the background) are given lower weights. A final attention score is generated, ranging from 0 to 1. Based on the multi-region content attention scores, the refresh requirements of different regions are analyzed. Dynamic content regions require higher refresh rates, while static content regions can have lower refresh rates to optimize energy consumption. The refresh rate for dynamic content regions is set to 60 Hz, while the refresh rate for static content regions can be set to 30 Hz. Content attention is correlated with refresh rate to generate differentiated refresh rate design schemes. Based on the attention score, regions are divided into high, medium, and low categories, and different refresh rates are assigned to each category. For areas with a sensitivity value higher than 0.7, the refresh rate is set to 60Hz; for areas with a sensitivity value between 0.4 and 0.7, it is set to 45Hz; and for areas with a sensitivity value lower than 0.4, it is set to 30Hz. A zoned intelligent drive control system is designed to dynamically adjust the refresh rate according to the different zones. The system should support independent control of the drive parameters for each zone to achieve differentiated driving. The system should be able to automatically identify zones and dynamically adjust the drive current and pulse width according to the preset refresh rate. Corresponding drive parameters, including current amplitude, pulse width, and refresh rate, are configured for each zone. It is ensured that the drive parameters for high refresh rate zones can support higher energy output.To optimize energy consumption and display quality, a current amplitude of 2.5mA is set for the 60 Hz area and 1.5 mA for the 30 Hz area. A zoned intelligent drive control system is activated to monitor each area in real time, ensuring that drive parameters adapt to changes in image content. Dynamic adjustments are made based on monitoring data to maintain image quality. The performance of high-dynamic areas is observed to ensure reduced energy consumption when static content appears, with timely adjustments to current and refresh rate to cope with changes. An intelligent collaborative drive control scheme is designed, combining energy-optimized zoned drive strategies with adaptive drive state adjustment strategies. This scheme should be able to automatically adjust the drive state of each area based on real-time image content changes. When high-dynamic content appears, the refresh rate and current output of that area are automatically increased to ensure the display effect is not affected. A dynamic adjustment mechanism is implemented to monitor image content changes in real time and automatically switch drive parameters and refresh rates for each area. This ensures the system can quickly respond to the user's visual needs. If the content in a certain area suddenly becomes dynamic, the system will immediately increase the refresh rate of that area to 60 Hz and adjust the current to ensure brightness consistency. The implementation effect of intelligent collaborative drive control is monitored in real time to evaluate the stability and consistency of image output. User feedback is collected to further optimize the drive strategy. System operation data is analyzed regularly to evaluate the balance between energy consumption and image quality in different areas, so as to adjust the strategy and improve overall performance.

[0098] In this embodiment, a full-color Micro LED microdisplay array driving device is provided for performing the method described above, including:

[0099] The multi-frequency excitation test module is used to perform multi-frequency full-pixel excitation tests and pixel excitation threshold sensitivity analysis on the micro-display array, thereby obtaining the excitation threshold sensitivity of each pixel.

[0100] The topology driving module is used to identify the linked changes of pixels in the micro-display array, and to adjust the topology driving between pixel arrays according to the excitation threshold sensitivity, thereby constructing pixel array topology driving parameters.

[0101] The spectral feedback control module is used to perform real-time drive control based on the pixel array topology driving parameters and to perform spectral feedback control to obtain a spectrally controlled image output stream.

[0102] The trajectory write-back module is used to perform inter-frame pixel movement tracking and sub-pixel level emission trajectory write-back on the spectral modulated image output stream to obtain a visually consistent image output stream.

[0103] The driving state adjustment module is used to perform deep fitting of multiple index changes on the visual consistency image output stream, and perform adaptive correction to build an adaptive driving state adjustment strategy.

[0104] The partitioned intelligent drive module is used to perform partitioned intelligent drive control on the visually consistent image output stream, and to perform intelligent collaborative drive control based on an adaptive drive state adjustment strategy to execute micro-display array drive operations.

[0105] This invention precisely models the microscopic electrical characteristics of each pixel to obtain its excitation threshold at different frequencies, laying the foundation for subsequent precise driving. It addresses individual pixel differences caused by inconsistent manufacturing processes, achieving pixel-level fine-grained calibration. This testing method is more representative than single-frequency static testing, simulating response behavior under actual working conditions and improving the matching degree between testing and application. By constructing a dynamic response relationship topology network between pixels, it better reflects the actual imaging linkage characteristics. It achieves driving control optimization from point to surface, from individual to overall, improving the overall consistency and synergy of the display array. It can dynamically adapt to non-ideal behaviors between pixels, effectively suppressing local non-uniformity issues such as hot spots and dark spots. Combined with a spectral feedback mechanism, it achieves accurate color reproduction and dynamic white balance adjustment, improving overall visual quality. It can adaptively adjust to color shifts caused by changes in ambient light and material aging, extending device lifespan and improving color stability. It supports HDR display requirements, achieving dynamic range expansion in brightness and color saturation. It accurately tracks high-speed moving image content, solving the problems of ghosting and distortion in moving images, significantly improving dynamic clarity. Subpixel-level write-back capability improves the spatial resolution of images, especially in edge details and fast-switching scenes. Combined with information from the previous modules, this enables the entire system to "understand image motion," enhancing the consistency of subjective visual perception. It can dynamically adapt to different image content (static / dynamic, bright / dark fields, etc.), providing optimized drivers for multiple scenarios. Through machine learning or deep fitting, it predicts display behavior and adjusts the drive state in advance, reducing flickering and latency issues. This enhances system robustness, providing self-healing and self-adjusting capabilities, reducing the need for manual maintenance and complex calibration. It avoids energy waste and refresh resource overload caused by full-screen synchronous driving; it prioritizes response in high-frequency changing areas and maintains energy-saving modes in low-dynamic areas, greatly improving the energy efficiency ratio of the display system; it supports cross-regional optical consistency control, avoiding phenomena such as "image tearing" and "local stuttering"; and it constructs a hybrid control system of "partitioned driving + unified strategy," a key technology for future large-scale micro-display array control.

[0106] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0107] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A driving method for a full-color Micro LED microdisplay array, characterized in that, Includes the following steps: Step S1: Perform multi-frequency full-pixel excitation test and pixel excitation threshold sensitivity analysis on the micro-display array to obtain the excitation threshold sensitivity of each pixel; Step S2: Identify the pixel linkage changes of the micro-display array, and adjust the topology driving between the pixel arrays according to the excitation threshold sensitivity to construct the pixel array topology driving parameters; Step S3: Perform real-time drive control based on the pixel array topology driving parameters and perform spectral feedback modulation to obtain a spectrally modulated image output stream; Step S4: Perform inter-frame pixel motion tracking and sub-pixel level luminescence trajectory rewriting on the spectral modulated image output stream to obtain a visually consistent image output stream; Step S5: Perform deep fitting of multiple index changes on the visual consistency image output stream, and perform adaptive correction to construct an adaptive state adjustment strategy. Step S6: Perform partitioned intelligent drive control on the visually consistent image output stream, and perform intelligent collaborative drive control based on the adaptive drive state adjustment strategy to execute the micro-display array drive operation.

2. The full-color Micro LED microdisplay array driving method according to claim 1, characterized in that, The specific steps of step S1 are as follows: Define multi-cycle pulse micro-excitation current values; perform multi-frequency full-pixel excitation tests on the micro-display array based on the multi-cycle pulse micro-excitation current values, and extract the excitation response parameters of each pixel of the micro-display array; Calculate the charge response characteristics of the excitation response parameters for each pixel; The brightness intensity of each pixel is calculated based on the charge response characteristics to generate the brightness intensity of pixels with different periodic pulses; A multi-cycle nonlinear relationship analysis of the brightness intensity of the pixels is performed to obtain the pixel variation law in different cycles; Based on the pixel change patterns in different periods, the excitation threshold sensitivity of each pixel is analyzed individually to obtain the excitation threshold sensitivity of each pixel.

3. The full-color Micro LED microdisplay array driving method according to claim 1, characterized in that, The specific steps of step S2 are as follows: The brightness intensity of pixels with different period pulses is used to identify the brightness difference between neighboring pixels, and the brightness difference feature of neighboring pixels is obtained. Pixel linkage change recognition is performed based on the brightness difference characteristics of neighboring pixels, and pixels with linkage brightness changes are extracted; The pixels with brightness-linked changes are dynamically divided into pixel regions to obtain multiple pixel-related subnets. Based on the excitation threshold sensitivity, the topology driving adjustment between pixel arrays is performed on the associated subnets of multiple pixels to construct pixel array topology driving parameters.

4. The full-color Micro LED microdisplay array driving method according to claim 3, characterized in that, The specific steps for adjusting the pixel array topology driving parameters by performing inter-pixel array topology driving adjustment on multiple pixel-related subnets based on the excitation threshold sensitivity are as follows: Define a preset standard brightness value; Based on the excitation threshold sensitivity and the preset standard brightness value, an adaptive driving current amplitude is designed to obtain the adaptive driving current amplitude for each pixel. Calculate the pixel linkage response period of the subnet region for multiple pixel points associated with subnets, and generate the linkage response period of different pixel subnets; The current drive cycle is adjusted according to the linkage response cycle to generate a current drive cycle; The topology driving parameters of the pixel array are constructed by adjusting the adaptive driving current amplitude and current driving period between the pixel arrays.

5. The full-color Micro LED microdisplay array driving method according to claim 1, characterized in that, Step S3 is as follows: The micro-display array is driven and controlled in real time based on the pixel array topology driving parameters, and the image output stream of the display array is acquired. Ultrashort time window interferometric sampling is performed on the output stream of the display array image to extract ultrashort time window spectral sampling information; Identify the underlying image spectral distribution of the ultrashort time window spectral sampling information; The spectral fluctuation frequency of the basic image spectral distribution is calculated, and frequency modeling is performed to generate the image spectral fluctuation curve; The dynamic spectral distribution shift of the image spectral fluctuation curve is calculated at high speed to generate dynamic spectral distribution shift data. The display array image output stream is spectrally controlled by dynamic spectral distribution offset data to obtain a spectrally controlled image output stream.

6. The full-color Micro LED microdisplay array driving method according to claim 1, characterized in that, The specific steps of step S4 are as follows: The spectral-controlled image output stream is decomposed into continuous image frames to extract time-series image frames. Perform inter-frame pixel motion tracking on time-series image frames to extract the pixel motion trajectory of consecutive frames; Motion vector estimation is performed on the pixel movement trajectory of consecutive frames, and the motion vector values ​​of multiple pixels are extracted. Image blur detection by pixel movement is performed on the image output stream of the display array to generate image dispersion and trailing information; Based on the image dispersion and trailing information, subpixel-level luminous trajectory rewriting is performed on the motion vector values ​​of multiple pixels to obtain a visually consistent image output stream.

7. The full-color Micro LED microdisplay array driving method according to claim 1, characterized in that, The specific steps of step S5 are as follows: Perform multi-frame image color recognition on the visually consistent image output stream and extract color features from the multi-frame images; Calculate the real-time brightness values ​​of multiple pixels in the visually consistent image output stream; The brightness decay rate of the pixels is calculated by performing a brightness decay rate calculation on the real-time brightness values ​​of the multiple pixels. Identify the pixel heating status of the microdisplay array; perform deep fitting of multiple index changes based on the color features of the multi-frame images, pixel brightness decay rate, and pixel heating status to obtain a driving state deviation map; Based on the driving state deviation map, the pixel array topology driving parameters are dynamically corrected by time window and current driving pulse width, and an adaptive driving state adjustment strategy is constructed.

8. The full-color Micro LED microdisplay array driving method according to claim 1, characterized in that, The specific steps of step S6 are as follows: Perform deep semantic analysis on the visually consistent image output stream to identify the semantic features of different regions. Image content is evaluated based on the image semantic features to generate multi-region content attention levels; The image refresh rate is differentiated based on the content attention of different regions to generate different image refresh rates for different regions; Based on the image refresh rate of different regions, partitioned intelligent drive control is performed to generate energy-efficient partitioned drive strategies. Intelligent collaborative drive control is implemented based on energy consumption optimization partitioning drive strategy and adaptive drive state adjustment strategy to execute micro-display array drive operations.

9. A full-color Micro LED micro-display array driving device, characterized in that, For performing the full-color Micro LED microdisplay array driving method as described in claim 1, comprising: The multi-frequency excitation test module is used to perform multi-frequency full-pixel excitation tests and pixel excitation threshold sensitivity analysis on the micro-display array, thereby obtaining the excitation threshold sensitivity of each pixel. The topology driving module is used to identify the linked changes of pixels in the micro-display array, and to adjust the topology driving between pixel arrays according to the excitation threshold sensitivity, thereby constructing pixel array topology driving parameters. The spectral feedback control module is used to perform real-time drive control based on the pixel array topology driving parameters and to perform spectral feedback control to obtain a spectrally controlled image output stream. The trajectory write-back module is used to perform inter-frame pixel movement tracking and sub-pixel level emission trajectory write-back on the spectral modulated image output stream to obtain a visually consistent image output stream. The driving state adjustment module is used to perform deep fitting of multiple index changes on the visual consistency image output stream, and perform adaptive correction to build an adaptive driving state adjustment strategy. The partitioned intelligent drive module is used to perform partitioned intelligent drive control on the visually consistent image output stream, and to perform intelligent collaborative drive control based on an adaptive drive state adjustment strategy to execute micro-display array drive operations.

Images