Flexible screen brightness control method and device based on region segmentation

Abstract

The invention relates to the technical field of flexible screens, and discloses a flexible screen brightness control method and device based on region segmentation, and the method comprises the steps: carrying out the multi-view image collection of a plurality of bending regions of a flexible display screen, obtaining target image data, and calculating the deformation compensation partition data of the flexible display screen; gridding brightness sampling is carried out on the deformation compensation partition data to generate brightness uniformity characteristic data, LED driving parameter calculation is carried out based on the brightness uniformity characteristic data, and a driving compensation matrix is output; inputting the driving compensation matrix into a second-order state observer for real-time current estimation and temperature feedback, and generating a partition driving control signal; and performing multi-region collaborative optimization on the partition driving control signal to realize brightness uniformity control of the flexible display screen in different bending states. According to the method, the influence of temperature drift on the display uniformity is eliminated, the stability of the driving current is improved, and the display deviation caused by temperature change is remarkably reduced.

Description

Technical Field

[0001] The present invention relates to the field of flexible screen technology, and in particular to a flexible screen brightness control method and device based on area segmentation. Background Art

[0002] As an important development direction of the new generation of display technology, flexible display technology has been widely used in wearable devices, foldable mobile phones and other fields due to its advantages of being bendable, thin and light. However, when flexible displays are bent, they are prone to uneven brightness in the display area. This unevenness is mainly due to the combined influence of factors such as uneven distribution of mechanical stress caused by bending deformation, drive current distortion and inconsistent temperature distribution.

[0003] Traditional display brightness control methods are mainly designed for rigid flat-panel displays and cannot effectively cope with the complex optical property changes of flexible displays when they are bent. These methods often use a single brightness compensation strategy, ignoring the non-uniform effect of bending deformation on the display area, resulting in unsatisfactory compensation effects and difficulty in adapting to dynamically changing bending states. Summary of the invention

[0004] The present invention provides a flexible screen brightness control method and device based on area segmentation, which are used to eliminate the influence of temperature drift on display uniformity, improve the stability of driving current, and significantly reduce display deviation caused by temperature changes.

[0005] In a first aspect, the present invention provides a flexible screen brightness control method based on region segmentation, the flexible screen brightness control method based on region segmentation comprising: Performing multi-view image acquisition on multiple curved areas of the flexible display screen to obtain target image data, and calculating deformation compensation partition data of the flexible display screen; Performing gridded brightness sampling and LED drive parameter calculation on the deformation compensation partition data, and outputting a drive compensation matrix; The drive compensation matrix is ​​input into a second-order state observer for real-time current estimation and temperature feedback to generate a partition drive control signal; The partition drive control signal is subjected to multi-region collaborative optimization to achieve brightness uniformity control of the flexible display screen under different bending states.

[0006] In a second aspect, the present invention provides a flexible screen brightness control device based on region segmentation, the flexible screen brightness control device based on region segmentation comprising: An acquisition module, used for performing multi-view image acquisition on multiple curved areas of the flexible display screen, obtaining target image data, and calculating deformation compensation partition data of the flexible display screen; A calculation module, used for performing gridded brightness sampling and LED drive parameter calculation on the deformation compensation partition data, and outputting a drive compensation matrix; A feedback module, used for inputting the drive compensation matrix into a second-order state observer for real-time current estimation and temperature feedback, and generating a partition drive control signal; The optimization module is used to perform multi-region collaborative optimization on the partition drive control signal to achieve brightness uniformity control of the flexible display screen under different bending states.

[0007] In the technical solution provided by the present invention, the brightness distribution characteristics of the flexible display screen under different bending states are accurately identified through multi-view image acquisition and adaptive area segmentation technology, thereby improving the accuracy of the characterization of the brightness distribution. The grid-based brightness sampling and dynamic compensation strategy, combined with the real-time calculation of the bending deformation, improves the accuracy of the horizontal and vertical brightness uniformity evaluation, and effectively solves the problem of uneven brightness caused by bending deformation. The LED drive parameter calculation method based on the temperature coupling compensation algorithm eliminates the influence of temperature drift on display uniformity, improves the stability of the driving current, and significantly reduces the display deviation caused by temperature changes. The second-order state observer is introduced for real-time current estimation, combined with cross-coupling filtering technology, the drive response time is shortened, and the dynamic response performance of the system is greatly improved. Through distributed optimization and multi-region collaborative control strategies, a smooth transition of brightness between adjacent display areas is achieved, ensuring the continuity and consistency of the display effect.

[0008] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0009] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic diagram of an embodiment of a flexible screen brightness control method based on area segmentation in an embodiment of the present invention; Figure 2 This is a schematic diagram of an embodiment of a flexible screen brightness control device based on area segmentation in an embodiment of the present invention. DETAILED DESCRIPTION

[0011] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0012] The terms "including" and "having" and any variations thereof mentioned in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device end including a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or device ends.

[0013] To facilitate understanding of this embodiment, a flexible screen brightness control method based on region segmentation disclosed in an embodiment of the present invention is first described in detail. Figure 1 As shown, the method comprises the following steps: 101. Perform multi-view image acquisition on multiple curved areas of the flexible display screen to obtain target image data, and calculate deformation compensation partition data of the flexible display screen; It is understandable that the execution subject of the present invention may be a flexible screen brightness control device based on area segmentation, or may be a terminal or a server, which is not specifically limited here. The embodiment of the present invention is described by taking a server as the execution subject as an example.

[0014] Specifically, the position of the image acquisition point is determined according to the curvature distribution of the flexible display screen to ensure that the collected data can accurately reflect the curvature state of the screen. The curvature information of the screen is measured using a high-precision sensor or computer vision technology, and multiple suitable acquisition points are selected based on the measurement data. The distribution of these acquisition points covers all curved areas as much as possible, so that the subsequent calculation can fully consider the morphological changes of the entire screen. The positions of multiple acquisition points are synchronously triggered and controlled to ensure that the acquired images are recorded at the same time scale, thereby avoiding data errors caused by time delays. The original acquisition images of multiple acquisition points are obtained through a high frame rate camera or a synchronous control system to reflect the morphology and brightness changes of the flexible screen at different positions. The original acquisition images are calibrated for lens distortion parameters and space coordinate correction to eliminate image distortion caused by lens distortion of the camera and ensure the accuracy of subsequent calculations. A calibration plate or other reference object of known geometric shape is used to establish a distortion model by multiple shooting and calculation of optical parameters, and the acquired image is corrected using the model so that the geometric shape in the image is more consistent with the actual physical structure. At the same time, through coordinate correction, each pixel in the image is mapped to a standardized spatial coordinate system, so that images of different perspectives can be analyzed and calculated in the same coordinate system, eliminating the coordinate offset problem caused by different camera perspectives. Based on the position correction image, the contour feature points of the display area are extracted to establish the spatial mapping relationship between the multi-perspective images. Through feature extraction algorithms, such as SIFT (scale-invariant feature transform) or ORB (oriented fast binary feature), the key feature points of the display area are identified, and the corresponding feature points under different perspectives are matched, so as to establish a spatial mapping relationship and describe the spatial correspondence between images taken by different cameras. Based on the position mapping data, the multi-perspective images are spatially registered and coordinate transformed so that all images are unified in the same coordinate system, thereby eliminating the differences caused by different camera shooting angles and positions. Image registration methods such as projection transformation and bilinear interpolation are used to enable all images to be analyzed and processed in a unified geometric framework to construct a globally consistent image data set. The image data in the unified coordinate system is processed to eliminate environmental lighting interference and suppress noise to improve the image quality and reduce the impact of the external environment on brightness calculation. Since the flexible display screen is in different lighting environments, an adaptive brightness equalization algorithm, such as histogram equalization or Retinex algorithm, is used to eliminate the impact of ambient light on image brightness. Wavelet denoising, median filtering and other methods are used to reduce noise interference in the image to obtain a clearer denoised image. The denoised image is contrast enhanced and the display area is divided into blocks. Contrast enhancement uses methods such as Gamma correction and Laplace enhancement to increase the local brightness difference of the image, making the brightness changes in different areas more obvious.The block processing of the display area divides the image into multiple sub-areas according to the physical structure of the screen, so that subsequent calculations can perform independent brightness compensation for different areas. Through the block method, the brightness change characteristics of the screen under different bending states can be analyzed more accurately, thereby improving the accuracy and effect of compensation. The local brightness difference matrix is ​​calculated based on the target image data to quantify the brightness changes between different areas. The local brightness difference matrix is ​​constructed by calculating the average brightness value of each sub-area and comparing it with the brightness value of the adjacent area. By analyzing the matrix, areas with uneven brightness are identified. Based on the brightness difference matrix, the flexible display screen is divided into regions to obtain deformation compensation partition data.

[0015] A brightness gradient operator is constructed for the target image data to accurately characterize the brightness change characteristics of different areas of the screen. The construction of the brightness gradient operator relies on differential operators, such as Sobel, Prewitt or Laplacian operators, which can effectively calculate the brightness change rate of pixels in the image in different directions to form brightness gradient data. Based on the brightness gradient calculation, the brightness change data is obtained by local area brightness scanning operation, and a sliding window or gridding method is used to extract brightness change features in different areas and record their directionality and amplitude information. The brightness change data is orthogonally decomposed to more accurately characterize the brightness change characteristics of the local area. Orthogonal decomposition is to decompose the brightness change data into two mutually perpendicular components, such as horizontal brightness gradient and vertical brightness gradient, and calculate the local brightness difference matrix in combination with the brightness gradient direction of adjacent areas. The role of this matrix is ​​to characterize the brightness change degree and directional relationship between different areas. When calculating the local brightness difference matrix, directional derivatives or least squares-based fitting methods are used to ensure that the brightness change trend can be accurately quantified and fully consider the continuity in space. Regional segmentation is performed based on the local brightness difference matrix and brightness continuity conditions to form initial partition data. The brightness continuity condition is set to ensure the rationality of the region division, that is, pixels with similar brightness change characteristics should be classified into the same region, and the boundaries of regions with obvious brightness changes should be accurately identified. The region division is performed using methods such as region growing algorithm, K-means clustering, and graph cut to ensure that the initial partition can fully reflect the brightness change law of the screen. The initial partition data is judged for region merging to optimize the boundaries of the partitions to make them more consistent with the actual screen bending state. The judgment of region merging is based on the brightness difference threshold of adjacent regions, that is, when the brightness gradient change of two adjacent regions meets certain continuity constraints, they are merged into the same region. The method based on connectivity analysis is used to ensure that the shape and boundaries of the merged region are more natural, and the optimized boundary data is obtained. The curvature analysis of the boundary optimization data is performed to calculate the compensation coefficient in combination with the deformation parameters of the flexible display. The purpose of curvature analysis is to identify the degree of curvature of the screen in different regions and adjust the compensation scheme of region division based on this information. The second-order differential curvature calculation method or the high-order curvature analysis method based on surface fitting is used to obtain the local curvature of the screen at different positions, and the influence of bending on brightness change is calculated in combination with the material characteristics of the flexible screen. Based on this curvature information, a compensation model is established to calculate the compensation coefficient of each area and use it for brightness adjustment. In the process of calculating the compensation coefficient, the optical properties of the screen material, the change in pixel arrangement caused by bending, and the compensation relationship of the LED drive current are considered to ensure that the final brightness compensation can accurately match the physical deformation of the screen. The compensation parameter data is weighted and fused with the boundary optimization data to form the final deformation compensation partition data.

[0016] 102. Perform grid-based brightness sampling and LED drive parameter calculation on the deformation compensation partition data, and output a drive compensation matrix; Specifically, the deformation compensation partition data is gridded according to the display resolution so as to perform refined brightness control on different areas of the screen. The principle of gridding is based on the pixel arrangement structure of the screen and takes into account the degree of curvature of each area to ensure that each grid can accurately characterize the local brightness change. After the gridding is completed, the sampling data matrix is ​​constructed by scanning the brightness value to obtain the gridded brightness data. The brightness values ​​of each grid area are obtained by point-by-point measurement or average calculation through a high-precision optical sensor or camera array, and a complete brightness data matrix is ​​formed. The gridded brightness data is calculated for probability distribution, and the brightness change trend of the screen in different areas is analyzed to obtain the brightness distribution feature vector. The calculation of the brightness distribution feature vector relies on statistical methods, such as histogram distribution analysis, mean and variance calculation, etc., to determine the distribution characteristics of different brightness values ​​on the entire screen. At the same time, in order to more accurately compensate for the brightness non-uniformity caused by deformation, the deformation angle, curvature radius and stress distribution data of the flexible display screen are combined for deformation analysis to quantify the impact of screen bending on brightness. In this process, the deformation data of the screen is measured by structured light scanning, gyroscope or flexible sensor, and the bending deformation data is established by finite element analysis or bending modeling method to describe the deformation characteristics of the screen at different positions. The bending deformation data is subjected to local stress analysis to further understand how the deformation affects the arrangement of pixels and the emission direction of light. The local stress analysis is based on the material properties of the flexible display screen. The stress distribution of each region is calculated by the stress-strain model, and the compensation weight is calculated in combination with the brightness distribution eigenvector to generate the brightness compensation coefficient. The calculation of the compensation weight takes into account multiple factors, such as the degree of screen bending, the arrangement deformation of pixels, and the nonlinear relationship of the LED driving current, to ensure that the compensation scheme can adapt to different deformation conditions and ensure the overall brightness uniformity. The brightness compensation coefficient is subjected to timing analysis to ensure that the compensation scheme can dynamically adapt to the deformation state of the screen at different time points. The timing analysis can identify the time trend of brightness changes, and obtain dynamic compensation parameters through iterative calculation of the compensation amount to improve the real-time performance of the compensation. In this process, an adaptive filtering algorithm or Kalman filter is used to iteratively optimize the compensation amount to eliminate brightness instability caused by environmental changes or screen deformation fluctuations. The dynamic compensation parameters are compensated with the gridded brightness data to generate brightness uniformity feature data, which reflects the brightness distribution of the screen under different bending states and is used to optimize the final brightness control scheme. A regional drive model is established based on the brightness uniformity feature data to calculate the LED drive parameters and finally output the drive compensation matrix. The construction of the regional drive model needs to consider the brightness requirements of different regions and the nonlinear relationship of the LED drive characteristics, so a machine learning model or a physical simulation-based algorithm is used to optimize the drive strategy.At the same time, in order to improve the compensation accuracy, a temperature coupling compensation algorithm is used to correct the brightness deviation of the LED under different temperature conditions. Temperature coupling compensation measures the screen temperature based on a thermistor or infrared sensor, and calculates the temperature compensation parameters of the LED in combination with the temperature-current characteristic curve to eliminate the impact of temperature changes on brightness uniformity. By integrating all compensation parameters into the drive compensation matrix, precise control of the LED drive is achieved, thereby ensuring the brightness uniformity of the flexible display under different bending states.

[0017] The brightness uniformity characteristic data and the volt-ampere characteristic curve of the LED are input into the parameter identification module. Through this module, the regional driving model that meets the actual driving requirements is obtained by using the nonlinear function fitting method. The temperature coefficient compensation matrix is ​​constructed according to the LED junction temperature collection data. The brightness output of the LED is related to its junction temperature. The change of temperature will cause the nonlinear change of the LED brightness, so the influence of temperature on the driving model is considered. The construction of the temperature coefficient compensation matrix collects the LED junction temperature data at different operating temperatures and combines the thermal response characteristics of the LED to correct the temperature compensation. By performing temperature coupling compensation operation on the regional driving model, a temperature correction model is obtained. The model dynamically adjusts the driving parameters according to the change of the LED junction temperature to ensure the brightness stability and uniformity of the LED under different temperature conditions. The temperature correction model is transformed into state space, and a polynomial fitting current-brightness response equation is established to describe the brightness change characteristics of the LED under different current drives. The driving state variable set is generated by fitting the response equation. According to the driving state variable set, the feedback gain coefficient of each region is calculated to adjust the output of the LED driver. The calculation of the feedback gain coefficient depends on the optimization method in control theory. The brightness correction ratio of each region is determined by analyzing the current-brightness response equation. At the same time, combined with the working bandwidth range of the driving circuit, the frequency response characteristics of the system are considered to adjust the compensation control parameters. Inter-regional interference decoupling calculation is performed on the compensation control parameters to eliminate cross-interference between regions and ensure that the brightness adjustment of each region is not affected by other regions. The least squares optimization algorithm is adopted, and the driving signals of each region are gradually adjusted through iterative optimization methods, so that the brightness adjustment of each region is independent of the influence of other regions, and a regional independent driving vector is generated. The regional independent driving vector is structurally reorganized based on the physical connection relationship of the display partition. The physical connection relationship of the display partition is related to the actual physical layout and circuit connection structure of the display screen. The purpose of the reorganization is to ensure the drive coordination of the entire display screen on the basis of independent driving of each region. Through structural reorganization, the brightness compensation of each region can be efficiently transmitted to each LED unit, thereby ensuring that the final output drive compensation matrix can achieve brightness uniformity of the flexible display screen under various bending and deformation states.

[0018] 103. Input the drive compensation matrix into the second-order state observer for real-time current estimation and temperature feedback to generate a partition drive control signal; Specifically, a second-order state observer is constructed based on the dynamic response characteristics of the LED driving current to observe the state variables of the driving compensation matrix and obtain the observer state equation. The dynamic response characteristics of the LED driving current are determined by its current-brightness relationship, temperature drift effect, parasitic inductance, capacitance and other factors. When constructing the observer, these influencing factors are considered to ensure that the observer can accurately track the changes in the LED current. The second-order state observer estimates the current state of the LED in real time, so that the drive system can more accurately adjust the input signal of the LED to improve the brightness uniformity. In the process of state variable observation, Kalman filtering or optimal state estimation algorithm is used to reduce the impact of noise on the accuracy of the observer and ensure that the obtained state equation can effectively describe the actual behavior of the LED driving system. The optimal feedback gain of the observer state equation is calculated according to the Lyapunov stability criterion to generate the observer feedback coefficient. Lyapunov stability analysis is used to determine the stable feedback control strategy of the system to ensure that the drive system always remains stable in dynamic changes. In this process, the Lyapunov function is constructed, and the optimal feedback gain coefficient is calculated according to its decreasing condition to ensure that the observer can effectively track the actual driving state of the LED under the condition of current fluctuation and temperature change. The calculated observer feedback coefficient is used to adjust the LED driving signal to compensate for the brightness deviation caused by external environment changes or screen deformation. The observer feedback coefficient is input into the orthogonal filter network according to the multi-component coupling relationship to extract the fundamental component of the LED from the driving current and generate current characteristic data. The function of the orthogonal filter network is to decompose the different frequency components in the driving signal so as to extract the current fundamental component that best represents the LED brightness output. In the LED driving system, the driving current often contains high-frequency noise components, which will affect the stability of the brightness. The interference signal is eliminated through the filter network, and only the key components that affect the brightness are retained to obtain the current characteristic data. The temperature-current compensation function is constructed based on the current characteristic data and the real-time temperature value of the LED to optimize the adjustment accuracy of the driving signal. Since the brightness output of the LED is greatly affected by temperature, the light efficiency of the LED will decrease as the junction temperature increases. Therefore, a temperature compensation function is established to correct the brightness fluctuation caused by temperature changes. When constructing the temperature-current compensation function, the experimental data fitting method is used to establish a mathematical model based on the LED current-brightness response curve at different temperatures to predict the impact of temperature changes on LED brightness. The cross-coupling filter is used to correct the current response and optimize the drive compensation signal. The cross-coupling filter dynamically adjusts the LED drive current according to the temperature changes in different areas to ensure the uniformity of brightness distribution and reduce the impact of temperature gradient on the display effect. According to the compensation control amount and the effective bandwidth range of the drive circuit, the LED drive parameters are closed-loop adjusted to generate control adjustment data.The core of closed-loop regulation is to ensure that the LED drive signal can adapt to different working conditions in real time and dynamically adjust the output parameters according to feedback information. When implementing closed-loop regulation, PID control, fuzzy control or adaptive control methods are used to optimize the stability of the drive current. Closed-loop regulation also needs to consider the working bandwidth range of the drive circuit to avoid exceeding the dynamic response capability of the circuit, ensure that the adjustment signal can effectively act on the LED drive system, and improve the stability of the entire system. The control adjustment data is reorganized according to the physical distribution of the display area and the drive topology relationship to output the partition drive control signal. Since the flexible display screen is composed of multiple partitions, and the drive signal of each partition is affected by the adjacent area, when organizing the control data, the interconnection relationship of each area is comprehensively considered, and it is ensured that the distribution of the drive signal can achieve uniform control of the overall brightness. In this process, matrix operations or network optimization algorithms are used to construct a reasonable drive signal distribution scheme, so that the final partition drive control signal can adapt to the different deformation states of the flexible screen and ensure the stability and consistency of the display effect.

[0019] 104. Perform multi-region collaborative optimization on the partition drive control signal to achieve brightness uniformity control of the flexible display screen under different bending states.

[0020] Specifically, an inter-partition interference coupling matrix is ​​constructed for the partition drive control signal, which describes the brightness interaction relationship between adjacent regions and reflects how the drive signals of each region affect each other. Since the optical properties and brightness distribution of each region are affected by deformation when the flexible display screen is bent, the physical connection relationship between adjacent display regions is considered to establish a regional state transfer function, so as to accurately characterize the dynamic response characteristics of each partition under different deformation conditions. By calculating the state transfer function, regional association data is obtained. A distributed optimization objective function is established based on the regional association data to ensure the optimization of global brightness uniformity and energy utilization efficiency. When establishing the optimization model, the brightness uniformity constraint and energy loss index are considered at the same time to balance the relationship between display quality and power consumption. The brightness uniformity constraint is used to ensure that the brightness difference of each region does not exceed the set threshold, while the energy loss index is used to optimize the LED drive current so that it can reduce power consumption as much as possible while meeting the brightness requirements. Based on these constraints, a global optimization function is constructed, the goal of which is to minimize the brightness deviation and energy consumption within the range of the entire display screen, so that the brightness changes of all regions are as consistent as possible, and the driving energy efficiency of the LED is optimized. In order to solve the global optimization function, the Lagrange multiplier method is used to ensure that the optimization calculation can meet all constraints. During the calculation process, the optimal control increment is determined according to the gradient direction of the regional control variable to ensure that the optimization adjustment minimizes excessive intervention on the existing driving signal while ensuring brightness uniformity. By calculating the gradient direction, the coordinated control parameters are obtained. These parameters are used to guide the drive adjustment scheme of each region, so that the brightness transition between adjacent regions is smoother and the global brightness uniformity is optimized. Based on the coordinated control parameters, the brightness boundaries of adjacent partitions are dynamically compensated to correct the brightness unevenness caused by screen deformation. An adaptive boundary smoothing algorithm is used to ensure that there is no sudden change in the brightness transition between regions. At the same time, the brightness value of the transition area is dynamically adjusted using an interpolation-based method to ensure the naturalness of the overall display effect. Through this dynamic compensation operation, boundary correction data is obtained, which is used to optimize the brightness connection between regions so that the display can maintain good visual consistency under different bending states. The boundary correction data is input into the adaptive weight allocation network to dynamically adjust the compensation intensity according to the deformation degree of the flexible display. The adaptive weight allocation network assigns different compensation weights to different regions according to the degree of curvature of different regions, so that the compensation strength can be dynamically adjusted as the deformation changes. The network uses neural network or fuzzy control methods to optimize the compensation strength based on historical data and real-time measurements. After calculating the regional compensation vector, the vector is used to reconstruct the data of each partition drive control signal, so that the final drive signal can adapt to the physical deformation of the screen to the greatest extent and ensure the stability and consistency of the display effect.By optimizing and reconstructing all partition drive signals, brightness control data is output, which is directly used for LED drive control to achieve brightness uniformity optimization of the flexible display screen in various bending states.

[0021] In the embodiment of the present invention, the brightness distribution characteristics of the flexible display screen under different bending states are accurately identified through multi-view image acquisition and adaptive area segmentation technology, thereby improving the accuracy of the characterization of the brightness distribution. The grid-based brightness sampling and dynamic compensation strategy, combined with the real-time calculation of the bending deformation, improves the accuracy of the horizontal and vertical brightness uniformity evaluation, and effectively solves the problem of uneven brightness caused by bending deformation. The LED drive parameter calculation method based on the temperature coupling compensation algorithm eliminates the influence of temperature drift on display uniformity, improves the stability of the driving current, and significantly reduces the display deviation caused by temperature changes. The second-order state observer is introduced for real-time current estimation, combined with the cross-coupling filtering technology, the drive response time is shortened, and the dynamic response performance of the system is greatly improved. Through distributed optimization and multi-region collaborative control strategies, a smooth transition of brightness between adjacent display areas is achieved, ensuring the continuity and consistency of the display effect.

[0022] In a specific embodiment, the process of executing step 101 may specifically include the following steps: Determine the position of the image acquisition point according to the curvature distribution of the flexible display screen, obtain a plurality of acquisition points, and perform synchronous trigger control on the plurality of acquisition points to obtain an original acquisition image; Perform lens distortion parameter calibration and space coordinate correction on the original captured image to obtain a position-corrected image; Extracting the feature points of the display area contour based on the position correction image, establishing the spatial mapping relationship between the multi-view images, and obtaining the position mapping data; Based on the position mapping data, the multi-view images are spatially registered and coordinate transformed to obtain image data in a unified coordinate system; Performing ambient light interference elimination and noise suppression processing on the image data in the unified coordinate system to obtain a denoised image, and performing contrast enhancement and display area block processing on the denoised image to obtain target image data; The local brightness difference matrix is ​​calculated according to the target image data, and the flexible display screen is divided into regions to obtain deformation compensation partition data.

[0023] Specifically, the deformation characteristics of the flexible display screen are analyzed. Since the surface curvature of the flexible display screen changes dynamically under different bending states, the curvature distribution calculation method is used to determine the reasonable image acquisition point position. The curvature is calculated by the curvature radius. To define, , and the curvature The spatial coordinate function of the screen surface Calculated by the following formula:

[0024] in, and The screen surface is and The first-order partial derivative in the direction of They are the second-order partial derivatives respectively. According to the curvature distribution, the position where the curvature changes significantly is selected as the image acquisition point to ensure that the acquired information can fully reflect the deformation of the screen. Use a synchronous trigger system to shoot multiple acquisition points simultaneously to ensure that all data is collected at the same time scale to avoid errors caused by time delays. The original acquired image is calibrated for lens distortion parameters and space coordinate correction to eliminate optical distortion and ensure the accuracy of image geometric relationships. Lens distortion is divided into radial distortion and tangential distortion, and its mathematical model is expressed as: ; ; in, is the corrected pixel coordinate, are the original pixel coordinates, is the distance to the optical axis, is the radial distortion coefficient, is the tangential distortion coefficient. The calibration plate and the least squares optimization method are used to calculate these distortion parameters, and the distortion correction is performed to obtain the position-corrected image. The feature points of the display area contour are extracted from the position-corrected image to establish the spatial mapping relationship between the multi-view images. The feature point extraction uses SIFT (Scale Invariant Feature Transform) or ORB (Oriented Fast Binary Feature) algorithms, which can identify stable corners or edge features and match corresponding points under different viewing angles. Assume For the feature points, and their projection relationship at different viewing angles is expressed by the homography matrix To describe: ; in, yes The homography transformation matrix is ​​optimized by the RANSAC (random sampling consensus) algorithm to remove mismatched points, thereby building a stable spatial mapping relationship and obtaining position mapping data. Based on the position mapping data, the multi-view images are spatially registered and coordinate transformed to obtain image data in a unified coordinate system. Spatial registration is achieved by transforming the matrix Align images from different viewpoints so that they share the same geometric reference frame. Registration is done using an affine transformation: ; in, is the translation amount, are the rotation and scaling coefficients. These parameters are optimized by minimizing the mean square error between the two images so that data from all perspectives can be converted into a unified coordinate system. After obtaining the image data in a unified coordinate system, ambient light interference elimination and noise suppression are performed to enhance image quality. Light interference is handled by adaptive histogram equalization, which can enhance local contrast and reduce brightness non-uniformity. Noise suppression uses wavelet denoising or mean filtering to remove high-frequency noise and retain image edge information to obtain a denoised image. The denoised image is contrast enhanced and the display area is divided into blocks to ensure the accuracy of brightness analysis to obtain the target image data. The local brightness difference matrix is ​​calculated based on the target image data to analyze the brightness changes in different areas of the screen. Suppose the brightness value of a certain area is , then the local brightness difference is calculated using the gradient operator:

[0025] Local brightness difference matrix Defined as:

[0026] This matrix is ​​used to characterize the brightness changes in different areas of the screen and serves as the basis for regional division. According to the brightness difference matrix, the flexible display screen is divided into regions to obtain deformation compensation partition data. The regional division adopts the clustering-based K-means algorithm or the graph cut-based segmentation method to ensure that regions with similar brightness characteristics are classified into the same partition.

[0027] In a specific embodiment, the execution step calculates the local brightness difference matrix according to the target image data, and divides the flexible display screen into regions to obtain deformation compensation partition data, which may specifically include the following steps: A brightness gradient operator is constructed for the target image data, and brightness change data is obtained by performing brightness scanning operations in the local area; Orthogonal decomposition is performed on the brightness change data, and the local brightness difference matrix is ​​calculated by combining the brightness gradient directions of adjacent areas; Based on the local brightness difference matrix and the brightness continuity condition, the region is segmented to obtain the initial partition data, and the initial partition data is subjected to the region merging judgment to obtain the boundary optimization data; The curvature analysis is performed on the boundary optimization data, and the compensation coefficient is calculated in combination with the deformation parameters of the flexible display screen to obtain the compensation parameter data. The compensation parameter data is weightedly fused with the boundary optimization data to obtain the deformation compensation partition data.

[0028] Specifically, the target image is spatially differentiated to capture the characteristics of local brightness changes. The brightness gradient operator is a mathematical tool used to calculate the brightness change rate of a pixel in the horizontal and vertical directions. The Sobel operator, Prewitt operator or Laplace operator is used to extract brightness gradient information. The Sobel operator is used to calculate the horizontal and vertical gradients of the image, which are defined as: ; ; in, and Respectively represent images exist Direction and The gradient in the direction, while " indicates the convolution operation. For any pixel point , and its gradient amplitude is calculated by the following formula: ; The gradient direction is calculated by the following formula:

[0029] Through the local area brightness scanning operation, the brightness gradient data of each pixel in the entire image is calculated, and a brightness change matrix is ​​formed, which represents the brightness change amplitude and direction of different regions. The brightness change data is orthogonally decomposed, and the local brightness difference matrix is ​​calculated by combining the brightness gradient direction of adjacent regions. Orthogonal decomposition is to project the brightness gradient data into the main direction and the vertical direction, so as to separate the brightness changes in different directions. Assume that the adjacent regions and The brightness gradient directions between and , then the gradient difference between the two is calculated as: ; in, represents the local brightness difference between two adjacent regions, and Respectively, the regions and The local brightness difference matrix is ​​constructed by calculating the brightness gradient difference between all adjacent regions: ; The region is segmented based on the local brightness difference matrix and the brightness continuity condition to obtain the initial partition data. The brightness continuity condition requires that the brightness change between adjacent regions is within a specific threshold. can be classified into the same area, that is: if ,but and The regions are grouped into the same region. The region segmentation is implemented by K-means clustering or graph cut-based algorithms to ensure the rationality of the partitioning. After obtaining the initial partitioning data, the region merging judgment is performed to optimize the boundary conditions. The merging judgment criteria are based on the brightness similarity and spatial connectivity of the region, that is, if the brightness change gradient of two adjacent regions is less than the set threshold, they are merged into a larger region, thereby reducing unnecessary partitions. The curvature analysis is performed on the boundary optimization data, and the compensation coefficient is calculated in combination with the deformation parameters of the flexible display. The deformation of the screen is calculated using the curvature function It means that the calculation formula is:

[0030] in, are the partial derivatives of the deformed surface, Based on the curvature analysis, the deformation of the screen in different areas is calculated, and the brightness compensation coefficient is adjusted according to the size of the deformation. The deformation variable at , then the compensation coefficient Defined as: ; in, is the deformation sensitivity coefficient, which determines the influence of deformation on brightness compensation. By calculating the compensation coefficient of each area, a compensation parameter matrix is ​​formed: ; The compensation parameter data and the boundary optimization data are weightedly fused to obtain deformation compensation partition data. The weighted fusion method uses interpolation methods, such as bilinear interpolation or B-spline interpolation, to make the compensation parameters smoothly transition between different regions, thereby avoiding the problem of sudden brightness changes. and The brightness compensation weights are and , then the final deformation compensation brightness Calculated as: ; in, and is the original brightness value. Through calculation and optimization, the brightness compensation partition data for different bending states of the flexible display screen is finally obtained.

[0031] In a specific embodiment, the process of executing step 102 may specifically include the following steps: The deformation compensation partition data is gridded according to the display resolution, and a sampling data matrix is ​​constructed by scanning the brightness values ​​to obtain gridded brightness data; The probability distribution of the gridded brightness data is calculated to obtain the brightness distribution feature vector, and the deformation analysis is performed based on the deformation angle, curvature radius and stress distribution of the flexible display screen to obtain the bending deformation data; Perform local stress analysis on the bending deformation data, calculate the compensation weight based on the brightness distribution feature vector, and obtain the brightness compensation coefficient; Perform time series analysis on the brightness compensation coefficient, obtain dynamic compensation parameters through iterative calculation of compensation amount, and perform compensation operation on the dynamic compensation parameters and gridded brightness data to obtain brightness uniformity characteristic data; A regional driving model is established based on the brightness uniformity characteristic data, and the temperature coupling compensation algorithm is used to calculate the LED driving parameters and output the driving compensation matrix.

[0032] Specifically, the deformation compensation partition data is grid-divided according to the display resolution so as to accurately measure the brightness change of the local area. Assuming that the resolution of the flexible display is , it is divided into grid units of uniform size, and the brightness value of each grid is expressed as ,in and Represents the row index and column index respectively. By scanning the brightness value of each grid area , construct the brightness sampling data matrix: ; in, , which means the grid area The brightness value after integration and averaging. Probability distribution calculation is performed on the gridded brightness data to obtain the brightness distribution feature vector. Assume that the brightness value of the pixel on the screen is in the interval The brightness value is divided into intervals, and the probability density of each interval is expressed as:

[0033] in, Indicates that the brightness value falls within the range The probability within is an indicator function, and a value of 1 indicates that the brightness value belongs to this interval, otherwise it is 0. By calculating the probability distribution of all intervals, the brightness distribution feature vector is obtained: ; Based on the brightness distribution feature vector, combined with the deformation angle of the flexible display , radius of curvature and stress distribution Perform deformation analysis to obtain bending deformation data. Deformation angle Reflects the tilt of the local area of ​​the screen, obtained by measuring the change of the normal vector:

[0034] in, is the normal vector of the local surface after bending, is the normal vector in a flat state. The radius of curvature Use the second-order derivative to calculate:

[0035] in, is the local first-order partial derivative of the surface, and is the second-order partial derivative. Stress distribution It reflects the effect of bending on the material and is calculated by the following formula:

[0036] in, is the Young's modulus of the material, is the thickness of the screen, and is the curvature. After obtaining the bending deformation data, the local area stress analysis is performed, and the compensation weight is calculated in combination with the brightness distribution feature vector to obtain the brightness compensation coefficient The compensation coefficient is fitted by the empirical model as follows: ; in, and is the adjustment coefficient, is the maximum brightness value of the screen. This compensation coefficient is used to correct the brightness loss caused by bending, so that the brightness uniformity is improved. The brightness compensation coefficient is analyzed in time series, and the dynamic compensation parameter is obtained by iterative calculation of the compensation amount. Assume that the compensation amount at the current moment is , then at the next time step Its change is described by the first-order difference equation:

[0037] in, is the time step factor, is the rate of change of the compensation amount, which is estimated by historical brightness data. Compensation operation is performed with the gridded brightness data to obtain brightness uniformity characteristic data:

[0038] After obtaining the brightness uniformity characteristic data, a regional driving model is established to calculate the LED driving parameters and output the driving compensation matrix. The regional driving model takes into account the volt-ampere characteristics of the LED and the temperature effect, so a temperature coupling compensation algorithm is used. Assume that the driving current of the LED is , its brightness output It can be expressed as:

[0039] in, is the forward voltage of the LED, It is a temperature-related light efficiency function. Since temperature changes will affect the light output efficiency of LEDs, a temperature compensation function is constructed: ; in, is the reference temperature The light effect below, is the temperature attenuation coefficient. By solving the LED drive current , the drive compensation matrix is ​​calculated:

[0040] The driving compensation matrix is ​​used to adjust the driving current of each LED unit so that it can provide uniform brightness output under different temperatures and deformation states.

[0041] In a specific embodiment, the execution step of establishing a regional driving model based on the brightness uniformity characteristic data, calculating the LED driving parameters using the temperature coupling compensation algorithm, and outputting the driving compensation matrix may specifically include the following steps: The brightness uniformity characteristic data and the LED volt-ampere characteristic curve are input into the parameter identification module, and the regional driving model is obtained by nonlinear function fitting; The temperature coefficient compensation matrix is ​​constructed based on the LED junction temperature collection data, and the temperature coupling compensation operation is performed on the regional driving model to obtain the temperature correction model; Performing state space transformation on the temperature correction model and establishing a polynomial fitting current-brightness response equation to generate a set of driving state variables; The feedback gain coefficient of each region is calculated according to the driving state variable set, and the compensation control parameter is obtained in combination with the working bandwidth range of the driving circuit; Inter-region interference decoupling calculation is performed on the compensation control parameters, and regional independent driving vectors are generated through least squares optimization iteration. The regional independent driving vectors are structurally reorganized based on the physical connection relationship of the display partitions, and the driving compensation matrix is ​​output.

[0042] Specifically, the brightness-current relationship of the LED is modeled. With its driving current The following relationship is satisfied: ; in, It is the temperature-related light efficiency coefficient, which indicates the light conversion efficiency of LED at different temperatures. is the forward voltage of the LED. In the parameter identification module, the experimentally measured data set is used to fit the volt-ampere characteristic curve of the LED using a nonlinear regression method. The volt-ampere characteristic of the LED is expressed as the Shockley diode equation:

[0043] in, is the reverse saturation current of the LED, is the electron charge, is the forward voltage of the LED, is the diode ideality factor, is the Boltzmann constant, is the absolute temperature. In the nonlinear fitting process, the Levenberg-Marquardt optimization algorithm is used to solve the best parameter combination so that the mean square error between the theoretical curve and the experimental data is minimized. The fitted regional drive model is used to calculate the brightness output of the LED under different driving conditions. After obtaining the regional drive model, the temperature coefficient compensation matrix is ​​constructed according to the LED junction temperature collection data, and the temperature coupling compensation operation is performed on the regional drive model to obtain the temperature correction model. Since the brightness of the LED is significantly affected by the junction temperature, the junction temperature Estimate using the following equation: ; in, is the ambient temperature, is the thermal resistance of the LED, is the power loss of the LED, calculated as: ; Based on the junction temperature data collection, the temperature coefficient compensation matrix is ​​constructed , and its calculation formula is: ; in, is the temperature compensation coefficient, is the reference temperature. The temperature correction model is expressed by the following formula: ; The temperature correction model is transformed into state space, and a polynomial fitting current-brightness response equation is established to generate a set of driving state variables. Representing the relationship between LED current, brightness and temperature, the state space equation is written as: ; ; in, is the state matrix of the system, is the input matrix, is the output matrix, is the noise term. In order to obtain an accurate current-brightness response relationship, a polynomial fitting method is used to convert the brightness Expressed as current A high-order polynomial of: ; in, is the fitting coefficient, which is calculated by the least squares method. The set of driving state variables includes , and The mapping relationship between them is used to optimize the LED driving strategy. After generating the driving state variable set, the feedback gain coefficient of each area is calculated according to the variable set, and the compensation control parameter is obtained by combining the working bandwidth range of the driving circuit. Feedback gain coefficient The calculation of is based on the pole configuration method, so that the closed-loop characteristics of the system meet the expected requirements. Define the error signal The target brightness The actual brightness Deviation between: ; Then design the feedback control law: ; in, is the feedback gain matrix, calculated by Lyapunov equation. Considering the working bandwidth of the driving circuit , the compensation control parameters must meet the following constraints: ; in, is the frequency response function of the system. The final calculated compensation control parameters ensure that the dynamic change of the driving current does not exceed the stability range of the system. Inter-region interference decoupling calculation is performed on the compensation control parameters to eliminate the coupling effect between different display areas. The interference between regions is calculated using the cross-influence matrix express:

[0044] in, Representative area The brightness of the area The influence of current. By solving the least squares optimization problem, the optimal solution is found :

[0045] in, is the regularization parameter. After solving the optimization problem, we get the region-independent driving vector Based on the physical connection relationship of the display partitions, the regional independent driving vectors are structurally reorganized to output the driving compensation matrix. Assume that the screen is divided into regions, and the driving current of each region is determined by the vector Given, the final drive compensation matrix Calculated by the following formula: ; The drive compensation matrix is ​​used to control the LED drive current in each area to maintain the brightness uniformity of the screen under different temperature and deformation conditions.

[0046] In a specific embodiment, the process of executing step 103 may specifically include the following steps: Based on the dynamic response characteristics of LED driving current, a second-order state observer is constructed to observe the state variables of the driving compensation matrix and obtain the observer state equation. The optimal feedback gain is calculated for the observer state equation according to the Lyapunov stability criterion to generate the observer feedback coefficient; The observer feedback coefficient is input into the orthogonal filter network according to the multi-component coupling relationship, the LED fundamental component is extracted from the driving current, and the current characteristic data is generated; A temperature-current compensation function is constructed based on current characteristic data and the real-time temperature value of the LED, and the current response is corrected using a cross-coupling filter to obtain the compensation control value; According to the compensation control amount and the effective bandwidth range of the driving circuit, the LED driving parameters are adjusted in a closed loop to obtain control adjustment data; The control adjustment data is reorganized according to the physical distribution of the display area and the driving topology relationship, and the partition driving control signal is output.

[0047] Specifically, a mathematical model of the LED drive system is established. The current-brightness response of the LED has dynamic characteristics, which can be described by a second-order differential equation, and its system state variables are expressed as: ; ; in, Indicates the instantaneous brightness of the LED, represents the rate of change of brightness, is the drive current input, and is the system attenuation coefficient, is the gain parameter. Assume the system output is: ; The system can be expressed in state space equation as: ; ; in, ; In order to estimate the system state in real time, a second-order state observer is constructed, and the state equation of the observer is written as: ; in, is the state variable estimated by the observer, is the observer gain matrix, is the measurement error signal. The optimal feedback gain is calculated according to the Lyapunov stability criterion to ensure the convergence of the observer. Define the Lyapunov function: ; in, is the error state vector, is a positive definite matrix. To ensure the asymptotic stability of the system, The Lyapunov equation must be satisfied: ; in, is a positive definite matrix, selected by the designer. By solving this equation, we get the optimal observer feedback gain matrix , ensuring the rapid convergence and stability of the system. After obtaining the observer feedback coefficient, it is input into the orthogonal filter network according to the multi-component coupling relationship to extract the fundamental component of the LED from the drive current and generate current characteristic data. Usually contains multiple frequency components, the fundamental component It can be extracted by orthogonal filter. The Fourier expansion of is: ; Among them, the fundamental component is obtained by the bandpass filter:

[0048] The current characteristic data extracted by this method is used for subsequent compensation calculations. Based on the current characteristic data and the real-time temperature value of the LED, a temperature-current compensation function is constructed, and the current response is corrected using a cross-coupling filter to obtain the compensation control amount. The light output of the LED is greatly affected by temperature, and its light efficiency is It is expressed as: ; in, is the reference temperature The light effect below, is the temperature attenuation coefficient. Combined with the current characteristic data , calculate the actual brightness of the LED :

[0049] If the actual brightness is different from the expected brightness If there is a deviation, the compensation control amount It can be calculated as:

[0050] In order to eliminate the mutual influence between different regions, a cross-coupling filter is used to correct the current response of different regions. The cross-coupling filter matrix is ​​expressed as:

[0051] in, Indicates area For Region By minimizing the following objective function:

[0052] Optimize the compensation distribution between regions. According to the compensation control amount and the effective bandwidth range of the driving circuit, the LED driving parameters are closed-loop adjusted to obtain the control adjustment data. The closed-loop control strategy adopts PID control, and its control law is expressed as:

[0053] in, is the error signal, They are proportional, integral and differential gains respectively. In order to ensure that the drive current is within the effective bandwidth, it is necessary to meet the following requirements: ; in, is the frequency response function of the system, is the bandwidth upper limit of the circuit. The control adjustment data is reorganized according to the physical distribution of the display area and the driving topology relationship to output the partition driving control signal. Assume that the display screen is divided into The driving current of each partition is determined by the vector Given, the final driving signal matrix It is expressed as:

[0054] The driving control signal ensures that the LEDs in each area can maintain consistent brightness output under different ambient temperatures and deformation conditions.

[0055] Among them, the generation of partition drive control signals also includes a multi-time scale collaborative control step: time scale decomposition of various dynamic response characteristics of the display area, classification of LED drive response, temperature change response and deformation impact response into fast, medium and slow dynamic processes, and obtaining hierarchical control data; a fast response model of the LED driver is established based on the hierarchical control data, and the key driving features are extracted using the state space simplification method to obtain a fast process model; the fast process model is reduced in order, and a medium-speed response subsystem is constructed based on the temperature change characteristics to obtain a temperature process model; a slow response subsystem is established based on the temperature process model and the deformation characteristics of the flexible display screen. Respond to the subsystem, extract the deformation influencing factors, and obtain the deformation process model; hierarchically combine the fast process model, temperature process model and deformation process model to construct a multi-time scale collaborative control framework and obtain a hierarchical control structure; design an adaptive sampling mechanism based on the hierarchical control structure, assign differentiated sampling periods to control loops of different time scales, and obtain sampling timing data; construct a hierarchical progressive solver according to the sampling timing data, and use a semi-analytical solution method to accelerate the calculation of the slow process to obtain the collaborative control parameters; reconstruct the partitioned drive control signal according to the collaborative control parameters according to the time scale, realize the hierarchical collaborative control of fast and slow dynamic processes, and output the optimized control data.

[0056] In a specific embodiment, the process of executing step 104 may specifically include the following steps: Constructing an inter-partition interference coupling matrix for the partition driving control signal, establishing a regional state transfer function according to the physical connection relationship of adjacent display areas, and obtaining regional correlation data; A distributed optimization objective function is established based on the regional correlation data, and the brightness uniformity constraint and energy loss index are added to the optimization model to obtain a global optimization function. Based on the global optimization function, the Lagrange multiplier is solved, and the optimal control increment is calculated according to the gradient direction of the regional control variable to obtain the coordinated control parameters; Based on the collaborative control parameters, dynamic compensation calculation is performed on the brightness boundaries of adjacent partitions to obtain boundary correction data; The boundary correction data is input into the adaptive weight allocation network, and the compensation strength is dynamically adjusted according to the deformation degree of the flexible display screen to obtain the regional compensation vector. The data of each partition driving control signal is reconstructed according to the regional compensation vector to output the brightness control data.

[0057] Specifically, a mathematical model is established to describe the relationship between the brightness of different areas. Assume that the flexible display screen is divided into The brightness of each area is determined by the driving current Control, the interference between regions is expressed as a coupling matrix :

[0058] in, Representative area Brightness For Region Drive current The degree of influence. When the area The driving changes will affect the region The brightness of the flexible display screen is increased, thereby forming a coupling effect between regions. When constructing the regional state transfer function, the influence relationship between adjacent regions is determined according to the physical structure of the flexible display screen. Assume Indicates area At the moment The brightness state of the state transfer equation is expressed as:

[0059] in, Representatives and regions The set of directly adjacent regions, It is a region Affected by the area The weight coefficient of influence, It is a region The driving input is obtained by solving the state transfer equation, and the regional correlation data is used for optimization control calculation. A distributed optimization objective function is established based on the regional correlation data, and the brightness uniformity constraint and energy loss index are incorporated into the optimization model to construct a global optimization function. The brightness uniformity constraint requires that the brightness of all regions is as close to the global average brightness as possible. ,Right now:

[0060] The energy loss index can be defined as:

[0061] In order to optimize brightness uniformity and energy loss simultaneously, the following global optimization function is constructed:

[0062] in, and are the weighting factors for brightness uniformity and energy loss, respectively. and The value of is used to strike a balance between uniformity and energy consumption. In the optimization process, the Lagrange multiplier method is used to solve the global optimization function, and the optimal control increment is calculated according to the gradient direction of the regional control variable to obtain the coordinated control parameters. Define the Lagrange function:

[0063] in, is the Lagrange multiplier, is the LED brightness response function. The optimal control increment is calculated by the gradient descent method:

[0064] in, is the step size factor, and the optimal The value is calculated to ensure the overall brightness uniformity and minimum energy consumption, and the coordinated control parameters are obtained. Based on the coordinated control parameters, the brightness boundaries of adjacent partitions are dynamically compensated to eliminate the brightness discontinuity caused by deformation. The brightness boundary correction data is defined as: ; in, is the average brightness of the adjacent area, is the smooth compensation coefficient. By smoothing the brightness interpolation, the brightness mutation phenomenon is eliminated. The boundary correction data is input into the adaptive weight distribution network, and the compensation strength is dynamically adjusted according to the deformation degree of the flexible display screen to obtain the regional compensation vector. Assume that the deformation degree is determined by the curvature Description, the compensation weight is expressed as:

[0065] in, is the deformation compensation factor, It is a region The local curvature of . The final regional compensation vector is:

[0066] The data of each partition driving control signal is reconstructed based on the regional compensation vector to output the final brightness control data. Assume that the final driving current matrix is ,but: ; The final brightness control data is expressed as: .

[0067] The step of inputting the boundary correction data into the adaptive weight allocation network, dynamically adjusting the compensation strength according to the deformation degree of the flexible display screen, obtaining the regional compensation vector, and reconstructing the data of the driving control signal of each partition according to the regional compensation vector to output the brightness control data includes: performing feature analysis on the boundary correction data, constructing the initial weight distribution matrix based on the spatial connection relationship of the display area, and obtaining the weight distribution data; establishing the deformation variable evaluation function according to the bending angle and curvature radius of the flexible display screen, and quantitatively calculating the stress distribution of the display area to obtain the deformation feature data; converting the deformation feature data into the weight adjustment coefficient, and performing adaptive learning based on the neural network model to obtain the network training parameters; and The training parameters are back-propagated, and the network weights are updated in combination with the brightness uniformity constraint of the display area to obtain the optimized weight data; a compensation intensity function is constructed according to the optimized weight data, and the boundary correction data is adaptively weighted to obtain the compensation adjustment matrix; the compensation coefficient of each display partition is dynamically corrected by using the compensation adjustment matrix, and a compensation weight vector is generated based on the regional topological relationship to obtain the regional compensation vector; the regional compensation vector is mapped and transformed according to the physical position relationship of the display partition, and a spatial distribution model of the compensation data is established to obtain the reconstructed mapping data; the reconstructed mapping data is matched with the partition driving signal, and a control data stream is generated based on the working characteristics of the regional driving circuit to obtain the brightness control data.

[0068] The embodiment of the present invention also includes a flexible screen driving circuit interference suppression step: performing a working spectrum analysis on the LED driving circuits in each area of ​​the flexible display screen, collecting switching frequency, harmonic frequency and amplitude data, and obtaining interference source characteristic data; performing frequency domain decomposition on the interference source characteristic data, identifying interference components in the three working states of the driving circuit being turned on, turned off, and oscillating, and obtaining interference classification data; designing an LC resonant compensation network based on the interference classification data, bypassing and guiding the common-mode interference current in the turn-off interval of the switch tube, and obtaining compensation circuit parameters; associating the compensation circuit parameters with the bending deformation of the flexible screen, constructing a parasitic parameter compensation model under bending stress conditions, and obtaining deformation compensation data; optimizing the driving circuit wiring according to the deformation compensation data, using a double-layer differential routing structure to eliminate electromagnetic induction, and obtaining a wiring optimization scheme; constructing an on-chip electromagnetic shielding structure based on the wiring optimization scheme, and arranging a gradient shielding layer in the bending area of ​​the flexible substrate to obtain shielding layer parameters; performing material stress analysis on the shielding layer parameters, establishing a shielding effectiveness compensation model under bending deformation, and obtaining compensation intensity data; adjusting the LED driving parameters according to the compensation intensity data, optimizing the driving current waveform to a soft switching characteristic, and outputting the final interference suppression control data.

[0069] The above describes the flexible screen brightness control method based on region segmentation in the embodiment of the present invention. The following describes the flexible screen brightness control device based on region segmentation in the embodiment of the present invention. Figure 2 In one embodiment of the present invention, a flexible screen brightness control device based on area segmentation includes: The acquisition module 201 is used to perform multi-view image acquisition on multiple curved areas of the flexible display screen to obtain target image data and calculate deformation compensation partition data of the flexible display screen; The calculation module 202 is used to perform grid brightness sampling and LED drive parameter calculation on the deformation compensation partition data, and output a drive compensation matrix; Feedback module 203, used for inputting the drive compensation matrix into the second-order state observer for real-time current estimation and temperature feedback, and generating a partition drive control signal; The optimization module 204 is used to perform multi-region collaborative optimization on the partition driving control signal to achieve brightness uniformity control of the flexible display screen in different bending states.

[0070] Through the collaboration of the above components, multi-view image acquisition and adaptive area segmentation technology, the brightness distribution characteristics of the flexible display screen under different bending states can be accurately identified, and the accuracy of the brightness distribution characterization can be improved. The grid-based brightness sampling and dynamic compensation strategy, combined with the real-time calculation of the bending deformation, improves the accuracy of the horizontal and vertical brightness uniformity evaluation, and effectively solves the problem of uneven brightness caused by bending deformation. The LED drive parameter calculation method based on the temperature coupling compensation algorithm eliminates the influence of temperature drift on display uniformity, improves the stability of the driving current, and significantly reduces the display deviation caused by temperature changes. The second-order state observer is introduced for real-time current estimation, combined with cross-coupling filtering technology, the drive response time is shortened, and the dynamic response performance of the system is greatly improved. Through distributed optimization and multi-region collaborative control strategies, a smooth brightness transition between adjacent display areas is achieved, ensuring the continuity and consistency of the display effect.

[0071] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0072] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0073] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A flexible screen brightness control method based on area segmentation, characterized in that: The method comprises: Performing multi-view image acquisition on multiple curved areas of the flexible display screen to obtain target image data, and calculating deformation compensation partition data of the flexible display screen; Performing gridded brightness sampling and LED drive parameter calculation on the deformation compensation partition data, and outputting a drive compensation matrix; The drive compensation matrix is ​​input into a second-order state observer for real-time current estimation and temperature feedback to generate a partition drive control signal; The partition drive control signal is subjected to multi-region collaborative optimization to achieve brightness uniformity control of the flexible display screen under different bending states.

2. The flexible screen brightness control method based on area segmentation according to claim 1, characterized in that: The method of performing multi-view image acquisition on a plurality of curved areas of the flexible display screen to obtain target image data and calculating deformation compensation partition data of the flexible display screen includes: Determine the position of the image acquisition point according to the curvature distribution of the flexible display screen to obtain a plurality of acquisition points, and perform synchronous trigger control on the plurality of acquisition points to obtain an original acquired image; Performing lens distortion parameter calibration and spatial coordinate correction on the original captured image to obtain a position-corrected image; Extracting the display area contour feature points based on the position correction image, establishing a spatial mapping relationship between the multi-view images, and obtaining position mapping data; Based on the position mapping data, spatial registration and coordinate transformation are performed on the multi-view images to obtain image data in a unified coordinate system; Performing ambient light interference elimination and noise suppression processing on the image data in the unified coordinate system to obtain a denoised image, and performing contrast enhancement and display area block processing on the denoised image to obtain target image data; A local brightness difference matrix is ​​calculated according to the target image data, and the flexible display screen is divided into regions to obtain deformation compensation partition data.

3. The flexible screen brightness control method based on area segmentation according to claim 2 is characterized in that: The calculating of the local brightness difference matrix according to the target image data and dividing the flexible display screen into regions to obtain deformation compensation partition data includes: Constructing a brightness gradient operator for the target image data, and obtaining brightness change data by performing a local area brightness scanning operation; Performing orthogonal decomposition on the brightness change data, and calculating the local brightness difference matrix by combining the brightness gradient directions of adjacent regions; Performing regional segmentation based on the local brightness difference matrix and the brightness continuity condition to obtain initial partition data, and performing regional merging judgment on the initial partition data to obtain boundary optimization data; The boundary optimization data is subjected to curvature analysis, and a compensation coefficient is calculated in combination with deformation parameters of the flexible display screen to obtain compensation parameter data. The compensation parameter data is weightedly fused with the boundary optimization data to obtain deformation compensation partition data.

4. The flexible screen brightness control method based on area segmentation according to claim 1, characterized in that: The gridding brightness sampling and LED driving parameter calculation of the deformation compensation partition data and outputting the driving compensation matrix include: The deformation compensation partition data is gridded according to the display resolution, and a sampling data matrix is ​​constructed by scanning the brightness values ​​to obtain gridded brightness data; Performing probability distribution calculation on the gridded brightness data to obtain a brightness distribution feature vector, and performing deformation analysis based on the deformation angle, curvature radius and stress distribution of the flexible display screen to obtain bending deformation data; Performing local area stress analysis on the bending deformation data, calculating compensation weights in combination with the brightness distribution feature vector, and obtaining brightness compensation coefficients; Performing a time series analysis on the brightness compensation coefficient, obtaining a dynamic compensation parameter by iterative calculation of the compensation amount, and performing a compensation operation on the dynamic compensation parameter and the gridded brightness data to obtain brightness uniformity characteristic data; A regional driving model is established based on the brightness uniformity characteristic data, a temperature coupling compensation algorithm is used to calculate LED driving parameters, and a driving compensation matrix is ​​output.

5. The flexible screen brightness control method based on area segmentation according to claim 4 is characterized in that: The method of establishing a regional driving model based on the brightness uniformity characteristic data, calculating LED driving parameters using a temperature coupling compensation algorithm, and outputting a driving compensation matrix includes: Input the brightness uniformity characteristic data and the LED volt-ampere characteristic curve into a parameter identification module, and obtain a regional driving model through nonlinear function fitting; A temperature coefficient compensation matrix is ​​constructed based on the LED junction temperature collection data, and a temperature correction model is obtained by performing temperature coupling compensation operation on the regional driving model; Performing state space transformation on the temperature correction model and establishing a polynomial fitting current-brightness response equation to generate a set of driving state variables; Calculate the feedback gain coefficient of each region according to the driving state variable set, and obtain the compensation control parameter in combination with the working bandwidth range of the driving circuit; Inter-region interference decoupling calculation is performed on the compensation control parameters, regional independent driving vectors are generated through least squares optimization iteration, and the regional independent driving vectors are structurally reorganized based on the physical connection relationship of the display partitions to output a driving compensation matrix.

6. The flexible screen brightness control method based on area segmentation according to claim 1, characterized in that: The step of inputting the drive compensation matrix into a second-order state observer for real-time current estimation and temperature feedback to generate a partition drive control signal includes: A second-order state observer is constructed based on the dynamic response characteristics of the LED driving current, and the state variable observation of the driving compensation matrix is ​​performed to obtain the observer state equation; Calculating the optimal feedback gain of the observer state equation according to the Lyapunov stability criterion to generate an observer feedback coefficient; Inputting the observer feedback coefficient into the orthogonal filter network according to the multi-component coupling relationship, extracting the LED fundamental component from the driving current, and generating current characteristic data; A temperature-current compensation function is constructed according to the current characteristic data and the real-time temperature value of the LED, and a cross-coupling filter is used to correct the current response to obtain a compensation control amount; According to the compensation control amount and the effective bandwidth range of the driving circuit, the LED driving parameters are closed-loop adjusted to obtain control adjustment data; The control adjustment data is reorganized according to the physical distribution of the display area and the driving topology relationship, and a partition driving control signal is output.

7. The flexible screen brightness control method based on area segmentation according to claim 1, characterized in that: The multi-region collaborative optimization of the partition drive control signal to achieve brightness uniformity control of the flexible display screen in different bending states includes: Constructing an inter-partition interference coupling matrix for the partition driving control signal, establishing a regional state transfer function according to the physical connection relationship between adjacent display areas, and obtaining regional association data; Establishing a distributed optimization objective function according to the regional association data, adding brightness uniformity constraints and energy loss indicators to the optimization model, and obtaining a global optimization function; Solving the Lagrange multiplier based on the global optimization function, calculating the optimal control increment according to the gradient direction of the regional control variable, and obtaining the coordinated control parameter; Based on the collaborative control parameters, a dynamic compensation operation is performed on the brightness boundaries of adjacent partitions to obtain boundary correction data; The boundary correction data is input into an adaptive weight distribution network, the compensation strength is dynamically adjusted according to the deformation degree of the flexible display screen, a regional compensation vector is obtained, and data of each partition driving control signal is reconstructed according to the regional compensation vector to output brightness control data.

8. A flexible screen brightness control device based on area segmentation, characterized in that: The device is used to execute the flexible screen brightness control method based on area segmentation according to any one of claims 1 to 7, and comprises: An acquisition module, used for performing multi-view image acquisition on multiple curved areas of the flexible display screen, obtaining target image data, and calculating deformation compensation partition data of the flexible display screen; A calculation module, used for performing gridded brightness sampling and LED drive parameter calculation on the deformation compensation partition data, and outputting a drive compensation matrix; A feedback module, used for inputting the drive compensation matrix into a second-order state observer for real-time current estimation and temperature feedback, and generating a partition drive control signal; The optimization module is used to perform multi-region collaborative optimization on the partition drive control signal to achieve brightness uniformity control of the flexible display screen under different bending states.

Images