LED display screen splicing intelligent control method and system

Abstract

The application discloses an LED display screen splicing intelligent control method and system, and relates to the technical field of LED display screen control.The method comprises the following steps: original data is extracted through double-dimension feature capture to generate a feature vector reflecting real-time photoelectric characteristics of lamp beads and a content feature matrix reflecting dynamic characteristics of display content; correction coefficients are generated through a nonlinear mapping engine based on the feature vector and the content feature matrix; and real-time control is performed based on the correction coefficients.The application comprehensively captures real-time photoelectric characteristics of lamp beads and dynamic characteristics of display content through double-dimension feature extraction; ensures display consistency by dynamically performing basic adjustment or global synchronous adjustment through consistency judgment of correction coefficients of adjacent areas; and adaptively updates self-learning coefficients alpha and beta, improves adaptability to complex scenes, reduces deviation of splicing areas, improves display stability and smoothness, has low end-to-end delay, and has high adjustment precision, so the application is suitable for various LED display screen splicing scenes.

Description

Technical Field

[0001] This invention relates to the field of LED display control technology, specifically to an intelligent control method and system for LED display splicing. Background Technology

[0002] In the field of LED display splicing control technology, with the expansion of display area and the increasing complexity of application scenarios, extremely high requirements are placed on the consistency of the image, the uniformity of brightness and the dynamic response speed at the splicing points of the display screens. As a core element to ensure display effect, the performance of dynamic correction technology directly affects the overall display quality.

[0003] In existing technologies, dynamic calibration of LED displays largely relies on preset environmental models and fixed algorithm parameters, adjusting the characteristics of LED chips and the displayed content through pre-defined logic. However, during actual operation, the photoelectric characteristics of LED chips can undergo nonlinear drift due to factors such as temperature changes. Simultaneously, the dynamic characteristics of the displayed content (such as pixel changes in high-speed motion images) have a complex coupling relationship with the characteristics of the LED chips. This fixed-mode calibration method struggles to adapt to these dynamic changes, leading to a significant decrease in calibration accuracy over time and the continuous accumulation of errors. Ultimately, this results in problems such as uneven brightness and image misalignment at splicing seams, severely impacting the stability and consistency of the display effect. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent control method and system for LED display splicing to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an intelligent control method for LED display splicing, comprising the following steps:

[0006] Native feature extraction: Extract raw data through dual-dimensional feature capture to generate feature vectors that reflect the real-time photoelectric characteristics of LED beads and content feature matrices that reflect the dynamic characteristics of the displayed content;

[0007] Dynamic coupling correction coefficient generation: Based on the feature vector and the content feature matrix, correction coefficients are generated through a nonlinear mapping engine. The nonlinear mapping engine includes a base mapping layer, a dynamic correction layer, and an output constraint layer. The nonlinear mapping engine first establishes an initial correction benchmark through the base mapping layer, then calculates the brightness correction amount through the dynamic correction layer and determines whether it belongs to the defined range. Finally, the output constraint layer generates the correction coefficients. The dynamic correction layer processes the data and determines whether its output result belongs to the defined range. If it does, the output constraint layer processes the data; otherwise, it returns to the dynamic correction layer to recalculate.

[0008] Hardware-level real-time execution: Real-time control is performed based on the correction coefficients, and during the execution process, it is determined whether the consistency of the correction coefficients in adjacent regions meets the preset threshold. If so, basic adjustment operations are performed; otherwise, global synchronization adjustment operations are performed.

[0009] Preferably, the native feature extraction process includes:

[0010] Physical dimension feature extraction: Real-time acquisition of the operating current I(t) and junction temperature T(t) of each LED, calculation of the current change rate ΔI / Δt and temperature change rate ΔT / Δt, and generation of the feature vector [I(t), T(t), ΔI / Δt, ΔT / Δt];

[0011] Content dimension feature extraction: Calculate the pixel value change rate R(t) in each display frame to generate the content feature matrix;

[0012] Feature association: The feature vector is associated with and stored in relation to the content feature matrix.

[0013] Preferably, the current change rate ΔI / Δt is the ratio of the current difference between adjacent sampling periods to the sampling period, the temperature change rate ΔT / Δt is the ratio of the temperature difference between adjacent sampling periods to the sampling period, the current sampling accuracy is 1μA, and the temperature sampling accuracy is ±0.5℃.

[0014] Preferably, the pixel value change rate R(t) is calculated by the variance of the pixel difference of three consecutive frames. The content feature matrix includes high-speed motion area identifiers and static area identifiers, wherein R(t) > 0.8 for high-speed motion areas and R(t) < 0.2 for static areas; the pixel activity change amount ΔR is the difference in pixel value change rate between adjacent frames, i.e., ΔR = |R(t) - R(t-1)|.

[0015] Preferably, the dynamic coupling correction coefficient generation process includes:

[0016] Basic mapping layer processing: Based on the photoelectric characteristic curve L of the LED at the time of manufacture. o =f(I o, T o Establish an initial calibration reference, where L o For standard brightness, I o T is the rated operating current of the LED chip. o The rated junction temperature of the LED chip;

[0017] Dynamic correction layer processing: The brightness correction amount ΔL(t) is calculated using a bivariate error compensation function, which is:

[0018] ΔL(t)=α·(I(t)-I o )·(T(t)-T o)+β·R(t)·∫ΔL(t-1)dt where ΔL(t) is the brightness correction amount at time t, α and β are self-learning coefficients, I(t) is the working current of the LED at time t, T(t) is the junction temperature of the LED at time t, R(t) is the pixel value change rate at time t, and ∫ΔL(t-1)dt is the cumulative value of the brightness correction amount before time t-1;

[0019] Range determination and processing: Determine whether the output ΔL(t) of the dynamic correction layer belongs to [0.95L]. o 1.05L o If the value falls within the range [0.95L], then proceed to the output constraint layer for processing; otherwise, return to the dynamic correction layer, attenuate the α and β coefficients by 10%, and recalculate until ΔL(t) belongs to the range [0.95L]. o 1.05L o ] interval;

[0020] Output constraint layer processing: Limits ΔL(t) after range judgment to [0.95L]. o 1.05L o The interval is used to generate the final correction coefficient.

[0021] Preferably, in the hardware-level real-time execution process:

[0022] The basic adjustment operation includes: acquiring I(t), T(t) data and R(t) characteristics every 10ms, adjusting the PWM duty cycle of the drive current based on the correction coefficient, with an adjustment accuracy of 1‰, and maintaining an end-to-end delay of <15ms;

[0023] The global synchronization adjustment operation includes: triggering a global synchronization command, re-executing the dynamic coupling correction coefficient generation step based on a unified time reference to generate synchronization correction coefficients, and then performing the basic adjustment operation based on the synchronization correction coefficients.

[0024] Preferably, the update mechanism for α and β includes:

[0025] When the pixel activity change ΔR > 0.5 or the temperature change rate ΔT / Δt > 2°C / s, update immediately; otherwise, the update cycle is adaptively adjusted according to the cumulative error. The cycle is 2 hours when the error is < 0.05% and 10 minutes when the error is > 0.1%. The initial value is 0.5 for both.

[0026] The present invention also provides an intelligent control system for LED display splicing, comprising:

[0027] The feature extraction module is used to extract raw data through two-dimensional feature capture, and generate feature vectors that reflect the real-time photoelectric characteristics of the LED beads and content feature matrices that reflect the dynamic characteristics of the displayed content.

[0028] The correction coefficient generation module is used to generate correction coefficients based on the feature vector and the content feature matrix through a nonlinear mapping engine. The nonlinear mapping engine includes a basic mapping layer, a dynamic correction layer and an output constraint layer. After processing by the dynamic correction layer, it determines whether the output result belongs to the limited range. If it does, the output constraint layer is executed; otherwise, it returns to the dynamic correction layer for recalculation.

[0029] The execution control module is used to perform real-time control based on the correction coefficient, and during the execution process, it determines whether the consistency of the correction coefficients in adjacent areas meets a preset threshold. If so, it performs basic adjustment operations; otherwise, it performs global synchronization adjustment operations.

[0030] The present invention also provides an electronic device, which is a physical device, comprising:

[0031] The processor and the memory are communicatively connected.

[0032] The memory is used to store at least one executable instruction executed by the processor, which executes the executable instruction to implement the LED display splicing intelligent control method as described above.

[0033] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the LED display splicing intelligent control method as described above.

[0034] Compared with the prior art, the beneficial effects of the present invention are:

[0035] This invention comprehensively captures the real-time photoelectric characteristics of LED beads and the dynamic characteristics of the displayed content through dual-dimensional feature extraction, and generates accurate correction coefficients by combining a nonlinear mapping engine to achieve hardware-level real-time control. By judging the consistency of correction coefficients in adjacent areas, it dynamically executes basic adjustments or global synchronous adjustments to ensure display consistency. The self-learning coefficients α and β can be adaptively updated to improve adaptability to complex scenarios, reduce splicing area deviation, improve display stability and smoothness, and have low end-to-end latency and high adjustment accuracy, making it suitable for various LED display splicing scenarios. Attached Figure Description

[0036] Figure 1 This is a main flowchart of an intelligent control method for LED display splicing provided in an embodiment of the present invention;

[0037] Figure 2 This is a schematic diagram of the structure of an intelligent control system for LED display splicing provided in an embodiment of the present invention;

[0038] Figure 3This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0039] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0040] The method in this embodiment is executed by a terminal, which can be a mobile phone, tablet computer, PDA, laptop or desktop computer, etc. Of course, it can also be other devices with similar functions, and this embodiment does not limit them.

[0041] Please see Figure 1 This invention provides an intelligent control method for LED display splicing, comprising:

[0042] Step 100, Native Feature Extraction: Extract raw data through dual-dimensional feature capture to generate feature vectors reflecting the real-time photoelectric characteristics of the LED beads and content feature matrix reflecting the dynamic characteristics of the displayed content.

[0043] It should be noted that, in this embodiment, dual-dimensional features refer to features extracted from both physical and content dimensions; raw data refers to data obtained directly from the LED display hardware and display content without filtering or processing; real-time photoelectric characteristics of the LED beads refer to the light and electricity-related characteristics exhibited by the LED beads during real-time operation, such as current, temperature, and their changes; feature vector refers to a vector composed of multiple feature values, used to centrally reflect the real-time photoelectric characteristics of the LED beads; dynamic features of the display content refer to the characteristics exhibited by the display content during changes, such as changes in pixel values; and the content feature matrix refers to a matrix composed of the dynamic features of the display content, used to reflect the dynamic changes of the display content.

[0044] The purpose of this step is to acquire raw data through two-dimensional features, and then generate feature vectors that reflect the real-time photoelectric characteristics of the LED beads and a content feature matrix that reflects the dynamic characteristics of the displayed content, providing basic data support for the subsequent generation of correction coefficients. By extracting features from two dimensions, factors affecting the splicing effect of LED displays can be comprehensively captured, thereby improving the accuracy of subsequent corrections and achieving more precise splicing control.

[0045] In one possible implementation, physical dimension feature extraction may include collecting the operating current and junction temperature of the LED beads and calculating their rate of change; content dimension feature extraction may include statistically analyzing the rate of change of pixel values ​​in each display frame and identifying regions with different motion states.

[0046] Specifically, the native feature extraction process in step 100 includes:

[0047] Step 110, Physical dimension feature extraction: Real-time acquisition of the operating current I(t) and junction temperature T(t) of each LED, calculation of the current change rate ΔI / Δt and temperature change rate ΔT / Δt, and generation of the feature vector [I(t), T(t), ΔI / Δt, ΔT / Δt];

[0048] Step 120, Content Dimension Feature Extraction: Calculate the pixel value change rate R(t) in each display frame to generate the content feature matrix;

[0049] Step 130, Feature Association: Associate and store the feature vector with the content feature matrix.

[0050] It should be noted that, in this embodiment, physical dimension feature extraction refers to the process of extracting features from the perspective of the physical working state of the LED; the working current I(t) refers to the real-time working current of the LED at time t; the junction temperature T(t) refers to the PN junction temperature of the LED at time t; the current change rate ΔI / Δt refers to the ratio of the change in current to the sampling period in two adjacent sampling periods; the temperature change rate ΔT / Δt refers to the ratio of the change in temperature to the sampling period in two adjacent sampling periods; the feature vector [I(t), T(t), ΔI / Δt, ΔT / Δt] refers to a vector composed of the above four physical quantities that centrally reflects the real-time photoelectric characteristics of the LED; content dimension feature extraction refers to the process of extracting features from the perspective of the dynamic changes of the displayed content; the pixel value change rate R(t) refers to the proportion of pixel value change in the displayed frame at time t; the content feature matrix refers to a matrix composed of multiple pixel value change rates that reflects the dynamic characteristics of the displayed content; feature association refers to the process of establishing a correspondence between the physical dimension feature vector and the content dimension content feature matrix and storing them. By refining the sub-steps of native feature extraction, the feature extraction process becomes more specific, ensuring comprehensive feature acquisition from both physical and content dimensions. Furthermore, by associative storage, it provides structured input for subsequent correction coefficient generation, thereby improving data utilization efficiency.

[0051] In one possible implementation, the sampling period for physical dimension feature extraction can be adjusted according to the working stability of the LED. For LEDs with high stability, the sampling period can be appropriately extended to reduce the amount of computation. In another possible implementation, timestamp alignment can be used when associating features to ensure that physical features and content features at the same time correspond accurately.

[0052] For example, in one feasible implementation, when the system performs native feature extraction, it first performs physical dimension feature extraction: real-time acquisition of the operating current I(t) = 20mA and junction temperature T(t) = 50℃ for each LED, calculates the current change rate ΔI / Δt = 0.5mA / ms and the temperature change rate ΔI / Δt = 0.2℃ / ms, and generates a feature vector [20mA, 50℃, 0.5mA / ms, 0.2℃ / ms]; then, it performs content dimension feature extraction: statistically analyzes the pixel value change rate R(t) = 0.6 in the display frame, and generates a content feature matrix containing this value; finally, it associates and stores the feature vector with the content feature matrix through timestamps, as input to the dynamic coupling correction coefficient generation step.

[0053] Wherein, the current change rate ΔI / Δt is the ratio of the current difference between adjacent sampling periods to the sampling period, the temperature change rate ΔT / Δt is the ratio of the temperature difference between adjacent sampling periods to the sampling period, the current sampling accuracy is 1μA, and the temperature sampling accuracy is ±0.5℃.

[0054] It should be noted that in this embodiment, the calculation method for the current change rate ΔI / Δt refers to subtracting the current value of the previous sampling period from the current value of the subsequent sampling period, and then dividing by the time interval between the two sampling periods; the calculation method for the temperature change rate ΔI / Δt is similar; the current sampling accuracy refers to the smallest resolvable unit when acquiring current, 1μA means that the error of the acquired current value does not exceed 1 microamp; the temperature sampling accuracy refers to the error range when acquiring temperature, ±0.5℃ means that the deviation between the acquired temperature value and the actual value is within 0.5℃. By clarifying the calculation methods and sampling accuracy of the current change rate and temperature change rate, the accuracy and consistency of physical dimension feature extraction are ensured, providing reliable basic data for the subsequent generation of correction coefficients and avoiding correction deviations caused by feature data errors.

[0055] In one possible implementation, for high-precision display scenarios (such as medical displays), the current sampling accuracy can be improved to 0.5μA and the temperature sampling accuracy to ±0.2℃; in another possible implementation, the sampling period can be set to 1ms to more sensitively capture rapid changes in current and temperature.

[0056] For example, in one feasible implementation, when the system extracts physical dimension features, the sampling period is set to 2ms. The current in the previous cycle is 19.8mA and the current in the next cycle is 20.2mA. The current change rate ΔI / Δt = (20.2-19.8)mA / 2ms = 0.2mA / ms is calculated. The temperature in the previous cycle is 49.7℃ and the temperature in the next cycle is 50.3℃. The temperature change rate ΔI / Δt = (50.3-49.7)℃ / 2ms = 0.3℃ / ms is calculated. The current sampling accuracy is controlled within 1μA and the temperature sampling accuracy is controlled within ±0.5℃ to ensure the accuracy of the feature vector.

[0057] The pixel value change rate R(t) is calculated by the variance of the pixel difference of three consecutive frames. The content feature matrix includes high-speed motion area identifiers and static area identifiers, where R(t) > 0.8 for high-speed motion areas and R(t) < 0.2 for static areas. The pixel activity change ΔR is the difference in pixel value change rate between adjacent frames, i.e., ΔR = |R(t) - R(t-1)|.

[0058] It should be noted that in this embodiment, the variance of the pixel difference over three consecutive frames refers to taking three consecutive display frames, calculating the difference between corresponding pixels in adjacent frames, and then calculating the variance of these differences to quantify the drastic change in pixel values. The pixel value change rate R(t) refers to the ratio of this variance to the maximum pixel value, reflecting the relative magnitude of pixel change. The high-speed motion area marker refers to the area in the content feature matrix where R(t) > 0.8, indicating that the displayed content in that area is moving rapidly. The static area marker refers to the area where R(t) < 0.2, indicating that the displayed content in that area is changing slowly. By clarifying the calculation method of the pixel value change rate and the area marker standard, the content dimension features more accurately reflect the dynamics of the displayed content, facilitating the adoption of differentiated strategies for areas with different motion states during subsequent correction, and improving the consistency of dynamic display.

[0059] In one possible implementation, for rapidly switching video content, the variance of the pixel difference over five consecutive frames can be used to calculate R(t) to capture motion features more stably; in another possible implementation, the threshold for high-speed motion areas can be adjusted to 0.7, and the threshold for static areas can be adjusted to 0.3 to adapt to the display requirements of different content.

[0060] For example, in one feasible implementation, when the system calculates the pixel value change rate R(t), it takes three consecutive frames of displayed content. The variance of the pixel difference between the first and second frames is 20, the variance of the pixel difference between the second and third frames is 25, the average variance is 22.5, and the maximum pixel value is 255. Then, R(t) = 22.5 / 255 ≈ 0.088. In the content feature matrix, the region with R(t) = 0.85 is marked as a high-speed motion region, and the region with R(t) = 0.15 is marked as a static region, providing a dynamic basis for subsequent correction.

[0061] Step 200, Dynamic Coupling Correction Coefficient Generation: Based on the feature vector and the content feature matrix, correction coefficients are generated through a nonlinear mapping engine. The nonlinear mapping engine includes a base mapping layer, a dynamic correction layer, and an output constraint layer. The nonlinear mapping engine first establishes an initial correction benchmark through the base mapping layer, then calculates the brightness correction amount through the dynamic correction layer and determines whether it belongs to the defined range. Finally, the correction coefficients are generated through the output constraint layer. The dynamic correction layer processes the data and determines whether its output result belongs to the defined range. If it does, the output constraint layer is executed; otherwise, the dynamic correction layer is returned to recalculate.

[0062] It should be noted that, in this embodiment, the feature vector refers to the vector generated in the original feature extraction step that reflects the real-time photoelectric characteristics of the LED bead; the content feature matrix refers to the matrix generated in the same original feature extraction step that reflects the dynamic characteristics of the displayed content; the nonlinear mapping engine refers to the tool used to map the feature vector and the content feature matrix into correction coefficients, which includes a basic mapping layer, a dynamic correction layer, and an output constraint layer; the basic mapping layer refers to the layer that establishes an initial correction benchmark based on the photoelectric characteristic curve of the LED bead at the time of manufacture; the dynamic correction layer refers to the layer that corrects the initial correction benchmark through a bivariate error compensation function; the output constraint layer refers to the layer that limits the output result of the dynamic correction layer to a specific range; the correction coefficient refers to the parameter used to adjust the display effect of the LED display screen; and the limited range refers to the range that limits the output result of the dynamic correction layer.

[0063] The purpose of this step is to generate accurate correction coefficients based on the feature vectors and content feature matrices extracted from the native features, using a nonlinear mapping engine to achieve precise correction of the LED display screen. By employing a multi-level nonlinear mapping engine, the correction can be performed by combining the physical characteristics of the LED chips with the dynamic characteristics of the displayed content, improving the adaptability and accuracy of the correction and reducing display deviations caused by environmental changes and dynamic content changes.

[0064] Additionally, it's important to note that determining whether the output result falls within the defined range after processing by the dynamic correction layer ensures that the generated correction coefficients are within a reasonable range, preventing over- or under-correction from affecting the display. If the result is within the range, it proceeds to the output constraint layer for further processing; otherwise, it returns to the dynamic correction layer for recalculation. This process guarantees the reliability of the correction coefficients.

[0065] In one possible implementation, the base mapping layer can establish its own initial calibration reference based on the factory photoelectric characteristic curves of different LEDs; in another possible implementation, the defined range can be adjusted according to different display scenarios and LED characteristics, such as for high-precision display scenarios, the range can be set narrower.

[0066] Specifically, the dynamic coupling correction coefficient generation process in step 200 includes:

[0067] Step 210, Basic Mapping Layer Processing: Based on the photoelectric characteristic curve L of the LED at the time of manufacture. o =f(I o, T o Establish an initial calibration reference, where L o For standard brightness, I o T is the rated operating current of the LED chip. o The rated junction temperature of the LED chip;

[0068] Step 220, Dynamic Correction Layer Processing: Calculate the brightness correction amount using a bivariate error compensation function. ΔL(t) The function is:

[0069] ΔL(t)=α·(I(t)-I o )·(T(t)-T o )+β·R(t)·∫ΔL(t-1)dt

[0070] In the formula, ΔL(t) is the brightness correction amount at time t, α and β are self-learning coefficients, I(t) is the working current of the LED at time t, T(t) is the junction temperature of the LED at time t, R(t) is the pixel value change rate at time t, and ∫ΔL(t-1)dt is the cumulative value of the brightness correction amount before time t-1.

[0071] Step 230, Range Judgment and Processing: Determine whether the output ΔL(t) of the dynamic correction layer belongs to [0.95L]. o 1.05L o If the value falls within the [0.95L] interval, proceed to step 240 for output constraint layer processing; otherwise, return to step 220 for dynamic correction layer, recalculate the α and β coefficients by attenuating them by 10% until ΔL(t) belongs to the [0.95L] interval. o 1.05L o ] interval;

[0072] Step 240, Output constraint layer processing: Processing the data after range judgment... ΔL(t) Limited to [0.95L] o 1.05L o The interval is used to generate the final correction coefficient.

[0073] It should be noted that in this embodiment, the basic mapping layer processing refers to the process of establishing an initial calibration reference based on the standard photoelectric characteristics of the LED at the time of manufacture; the photoelectric characteristic curve c=f(I o, T o This refers to the LED chip operating at its rated current I. o and rated junction temperature T o Below, standard brightness T o The functional relationship between the two; dynamic correction layer processing refers to the process of adjusting the initial correction benchmark by combining real-time characteristics; bivariate error compensation function refers to the function that simultaneously considers the dynamic influence of current, temperature and display content on brightness to calculate the brightness correction amount; α and β are self-learning coefficients, which are coefficients that can be dynamically adjusted according to the system operation; brightness correction amount ΔL(t) refers to the amount of brightness adjustment required for the LED at time t; range judgment and processing refers to the process of checking whether the brightness correction amount output by the dynamic correction layer is within a reasonable range; [0.95L o 1.05L o The "range" refers to the reasonable range of brightness correction, ensuring that the corrected brightness does not deviate too much from the standard brightness. The output constraint layer processing refers to limiting the brightness correction within this range, ultimately generating correction coefficients that can be directly used for control. By refining the sub-steps of dynamically coupled correction coefficient generation, from the initial reference to dynamic correction and then to range constraints, each step ensures that the generated correction coefficients are accurate and reasonable, adapting to changes in the real-time status of the LEDs and the displayed content, thus improving the consistency of the spliced ​​display.

[0074] In one possible implementation, the base mapping layer can store multiple sets of photoelectric characteristic curves for different LED bead models to adapt to mixed splicing scenarios; in another possible implementation, [0.95L o 1.05L o The range can be dynamically adjusted according to the degree of aging of the LED beads. For LED beads with more severe aging, the range can be appropriately widened to [0.9L]. o 1.1L o ].

[0075] For example, in one feasible implementation, when the system performs dynamic coupling correction coefficient generation, the base mapping layer is based on the LED chip's factory photoelectric characteristic curve L. o =f(20mA,45℃)=500cd / m² establishes the initial baseline; the dynamic correction layer is established through the function...

[0076] ΔL(t)=0.3·(22mA-20mA)·(50℃-45℃)+0.2·0.6·∫ΔL(t-1)dt,

[0077] The calculated value is ΔL(t) = 3.6 + 0.12·∫ΔL(t-1)dt = 4.8cd / m 2 If 4.8 cd / m² is determined to be within the range of [475 cd / m², 525 cd / m²] (0.95×500 to 1.05×500), it will be processed in the output constraint layer to generate the final correction coefficient.

[0078] The update mechanisms for α and β include:

[0079] When the pixel activity change ΔR > 0.5 or the temperature change rate ΔT / Δt > 2℃ / s, update immediately; otherwise, the update cycle is adaptively adjusted according to the cumulative error. The cycle is 2 hours when the error is < 0.05% and 10 minutes when the error is > 0.1%. The initial value is 0.5 for both.

[0080] It should be noted that in this embodiment, the update mechanism of α and β refers to the rule that the self-learning coefficients are adjusted according to the changes in the system's operating state; the pixel activity change ΔR refers to the absolute value of the difference between the pixel value change rates at adjacent times, i.e., |R(t)-R(t-1)|; the temperature change rate ΔT / Δt>2℃ / s means that the temperature change exceeds 2℃ per second; the error accumulation refers to the cumulative deviation between the actual display brightness and the standard brightness; and the update cycle refers to the time interval for the automatic update of the self-learning coefficients. By clearly defining the update conditions and cycle of the self-learning coefficients, α and β can be dynamically adjusted in real time or at regular intervals according to the display content and the temperature changes of the LED beads, ensuring that the bivariate error compensation function always adapts to the current system state and improving the adaptability and accuracy of the correction coefficients.

[0081] In one possible implementation, for outdoor displays (where temperature changes drastically), the trigger threshold for the rate of temperature change can be adjusted to 1.5℃ / s to update the coefficients more sensitively; in another possible implementation, the calculation of the cumulative error can be performed using a weighted average method, giving higher weight to recent errors to make the update more in line with the latest state.

[0082] For example, in one feasible implementation, the system detects a pixel activity change ΔR = 0.6 (>0.5) and immediately updates the values ​​of α and β; when ΔR = 0.3 and the temperature change rate ΔT / Δt = 1℃ / s, if the cumulative error is 0.03% (<0.05%), the update cycle of α and β is set to 2 hours; if the cumulative error is 0.12% (>0.1%), the update cycle is set to 10 minutes, with initial values ​​of α = 0.5 and β = 0.5.

[0083] Step 300, hardware-level real-time execution: Real-time control is performed based on the correction coefficient, and during the execution process, it is determined whether the consistency of the correction coefficients in adjacent areas meets the preset threshold. If so, a basic adjustment operation is performed; otherwise, a global synchronization adjustment operation is performed.

[0084] It should be noted that, in this embodiment, real-time control refers to the immediate adjustment and management of the display state of the LED display screen; adjacent areas refer to mutually adjacent display areas in the LED display screen; correction coefficient consistency refers to the degree of agreement between the correction coefficients of adjacent areas; preset threshold refers to the standard value used to determine whether the consistency of the correction coefficients of adjacent areas meets the requirements; basic adjustment operation refers to the routine adjustment operation performed when the consistency of the correction coefficients of adjacent areas meets the preset threshold; global synchronous adjustment operation refers to the operation performed to synchronize the correction coefficients of each area when the consistency of the correction coefficients of adjacent areas does not meet the preset threshold.

[0085] The purpose of this step is to perform real-time control of the LED display based on the generated correction coefficients. By judging the consistency of the correction coefficients in adjacent areas, it determines whether to perform basic adjustment operations or global synchronous adjustment operations, thus ensuring the display consistency and stability of the LED display. Real-time control can respond promptly to changes in display requirements, while judging the consistency of correction coefficients in adjacent areas can avoid display splicing problems caused by differences in correction between areas, thereby improving the overall display effect.

[0086] In one possible implementation, the preset threshold can be set according to the splicing accuracy requirements of the display screen. For scenarios with high splicing accuracy requirements, the preset threshold can be set to a smaller value. In another possible implementation, the basic adjustment operation can include adjusting parameters such as the PWM duty cycle of the drive current and the refresh frequency.

[0087] Specifically, in the hardware-level real-time execution process of step 300:

[0088] The basic adjustment operation includes: acquiring I(t), T(t) data and R(t) characteristics every 10ms, adjusting the PWM duty cycle of the drive current based on the correction coefficient, with an adjustment accuracy of 1‰, and maintaining an end-to-end delay of <15ms;

[0089] The global synchronization adjustment operation includes: triggering a global synchronization command, re-executing the dynamic coupling correction coefficient generation step based on a unified time reference to generate synchronization correction coefficients, and then performing the basic adjustment operation based on the synchronization correction coefficients.

[0090] It should be noted that in this embodiment, the basic adjustment operation refers to the routine real-time control operation performed by the system when the consistency of the correction coefficients in adjacent areas meets a preset threshold; the PWM duty cycle refers to the ratio of the high-level duration to the period in the pulse width modulation signal, used to adjust the effective value of the drive current; the adjustment accuracy of 1‰ means that the minimum adjustment unit of the PWM duty cycle is one-thousandth; the end-to-end delay refers to the time interval from acquiring feature data to completing the adjustment; the global synchronization adjustment operation refers to the operation of synchronously updating the correction coefficients performed by the system when the consistency of the correction coefficients in adjacent areas does not meet a preset threshold; the global synchronization command refers to the signal that triggers each area to simultaneously regenerate the correction coefficients; and the unified time reference refers to the common time reference for the operation performed by each area. By clarifying the specific operations under different consistency conditions, stable display is maintained when the correction coefficients are consistent, and rapid synchronization is achieved when they are inconsistent, avoiding display differences in spliced ​​areas and ensuring the continuity of the overall display effect.

[0091] In one possible implementation, for scenarios with high refresh rate requirements (such as e-sports displays), the period for acquiring feature data can be shortened to 5ms, and the end-to-end latency can be controlled within 10ms; in another possible implementation, the global synchronization command can be sent via wireless synchronization signal to adapt to the wiring limitations of large splicing screens.

[0092] For example, in one feasible implementation, when the system performs hardware-level real-time control, the basic adjustment operation acquires I(t)=21mA, T(t)=51℃ and R(t)=0.7 every 10ms. Based on the correction coefficient, the PWM duty cycle of the drive current is adjusted from 30% to 32%, with an adjustment accuracy of 1‰ and an end-to-end delay of 12ms. When the difference in correction coefficient between adjacent regions is detected to be ΔC=8% (exceeding the 5% preset threshold), a global synchronization adjustment operation is performed, triggering a global synchronization command. After each region regenerates the synchronization correction coefficient based on a unified time base, the adjustment is performed according to the basic adjustment operation procedure.

[0093] In this embodiment, the present invention comprehensively captures the real-time photoelectric characteristics of LED beads and the dynamic characteristics of the displayed content through dual-dimensional feature extraction, and generates accurate correction coefficients by combining a nonlinear mapping engine to achieve hardware-level real-time control; by judging the consistency of correction coefficients in adjacent areas, it dynamically executes basic adjustment or global synchronous adjustment to ensure display consistency; the self-learning coefficients α and β can be adaptively updated to improve adaptability to complex scenarios, reduce splicing area deviation, improve display stability and smoothness, and have low end-to-end latency and high adjustment accuracy, making it suitable for various LED display splicing scenarios.

[0094] Based on the above embodiments, such as Figure 2As shown, the present invention also provides an intelligent control system for LED display splicing, used to support the intelligent control method for LED display splicing described in the above embodiments. The intelligent control system for LED display splicing includes:

[0095] Feature extraction module 11 is used to extract raw data through dual-dimensional feature capture, and generate feature vectors that reflect the real-time photoelectric characteristics of the LED beads and content feature matrix that reflects the dynamic characteristics of the display content;

[0096] The correction coefficient generation module 12 is used to generate correction coefficients based on the feature vector and the content feature matrix through a nonlinear mapping engine. The nonlinear mapping engine includes a basic mapping layer, a dynamic correction layer and an output constraint layer. After processing by the dynamic correction layer, it is determined whether the output result belongs to the limited range. If so, the output constraint layer is executed; otherwise, the process is returned to the dynamic correction layer for recalculation.

[0097] The execution control module 13 is used to perform real-time control based on the correction coefficient, and during the execution process, it determines whether the consistency of the correction coefficients in adjacent areas meets the preset threshold. If so, it performs basic adjustment operation; otherwise, it performs global synchronization adjustment operation.

[0098] In an optional embodiment, the feature extraction module 11 is further configured to collect the operating current I(t) and junction temperature T(t) of each LED in real time, calculate the current change rate ΔI / Δt and the temperature change rate ΔT / Δt, generate the feature vector [I(t), T(t), ΔI / Δt, ΔT / Δt], calculate the pixel value change rate R(t) in each display frame, generate the content feature matrix, and associate and store the feature vector with the content feature matrix.

[0099] In an optional embodiment, the correction coefficient generation module 12 is further configured to generate a correction coefficient based on the photoelectric characteristic curve L of the LED chip at the time of manufacture. o =f(I o, T o Establish an initial calibration reference, where L o For standard brightness, I o T is the rated operating current of the LED chip. o The rated junction temperature of the LED chip is given; the brightness correction amount ΔL(t) is calculated using a bivariate error compensation function, which is:

[0100] ΔL(t)=α·(I(t)-I o )·(T(t)-T o )+β·R(t)·∫ΔL(t-1)dt

[0101] In the formula, ΔL(t) is the brightness correction amount at time t, α and β are self-learning coefficients, I(t) is the operating current of the LED at time t, T(t) is the junction temperature of the LED at time t, R(t) is the pixel value change rate at time t, and ∫ΔL(t-1)dt is the cumulative value of the brightness correction amount before time t-1; determine whether the ΔL(t) output by the dynamic correction layer belongs to [0.95L o 1.05L o If the value falls within the range [0.95L], then proceed to the output constraint layer for processing; otherwise, return to the dynamic correction layer, attenuate the α and β coefficients by 10%, and recalculate until ΔL(t) belongs to the range [0.95L]. o 1.05L o The interval is defined; the range-determined ΔL(t) is limited to [0.95L]. o 1.05L o The interval is used to generate the final correction coefficient.

[0102] In an optional embodiment, the execution control module 13 is further configured to acquire I(t), T(t) data and R(t) characteristics every 10ms, adjust the PWM duty cycle of the drive current based on the correction coefficient, with an adjustment accuracy of 1‰, and maintain an end-to-end delay of <15ms; trigger a global synchronization instruction, re-execute the dynamic coupling correction coefficient generation step based on a unified time base to generate a synchronization correction coefficient, and then execute the basic adjustment operation based on the synchronization correction coefficient.

[0103] In this embodiment, the present invention comprehensively captures the real-time photoelectric characteristics of LED beads and the dynamic characteristics of display content through dual-dimensional feature extraction, and generates accurate correction coefficients by combining a nonlinear mapping engine to achieve hardware-level real-time control; by judging the consistency of correction coefficients in adjacent areas, it dynamically executes basic adjustment or global synchronous adjustment to ensure display consistency; the self-learning coefficients α and β can be adaptively updated to improve adaptability to complex scenarios, reduce splicing area deviation, improve display stability and smoothness, and have low end-to-end latency and high adjustment accuracy, making it suitable for various LED display splicing scenarios.

[0104] Furthermore, the LED display splicing intelligent control system can run the above-mentioned LED display splicing intelligent control method. For specific implementation details, please refer to the method embodiment, which will not be repeated here.

[0105] Based on the above embodiments, such as Figure 3 As shown, the present invention also provides an electronic device, the electronic device comprising:

[0106] The processor 22 includes at least one processor 22, at least one memory 21, a communication interface 23, and a communication bus 24, wherein the processor 22 is communicatively connected to the memory 21.

[0107] In this embodiment, the memory 21 can be implemented in any suitable manner, for example, the memory 21 can be a read-only memory, a hard disk drive, a solid-state drive, or a USB flash drive, etc.; the memory 21 is used to store at least one executable instruction executed by the processor;

[0108] In this embodiment, the processor 22 can be implemented in any suitable manner. For example, the processor 22 can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) that can be executed by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers, etc.; the processor is used to execute the executable instructions to implement the LED display splicing intelligent control method as described above.

[0109] Based on the above embodiments, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the LED display splicing intelligent control method as described above.

[0110] Those skilled in the art will recognize that the modules and method steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0111] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, equipment, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0112] In the several embodiments provided in this application, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or units may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or equipment, and may be electrical, mechanical, or other forms.

[0113] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0114] In addition, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0115] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program instructions, such as USB flash drives, portable hard drives, read-only storage servers, random access storage servers, magnetic disks, or optical disks.

[0116] In addition, it should be noted that the combination of the technical features in this case is not limited to the combination methods described in this case or the combination methods described in the specific embodiments. All technical features described in this case can be freely combined or combined in any way, unless there is a contradiction between them.

[0117] It should be noted that the above examples are merely specific embodiments of the present invention, and the present invention is obviously not limited to the above embodiments, with many similar variations. All modifications that can be directly derived or conceived by those skilled in the art from the content disclosed in this invention should fall within the protection scope of this invention.

[0118] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for intelligent control of LED display splicing, characterized in that, Includes the following steps: Native feature extraction: Extract raw data through dual-dimensional feature capture to generate feature vectors that reflect the real-time photoelectric characteristics of LED beads and content feature matrices that reflect the dynamic characteristics of the displayed content; Dynamic coupling correction coefficient generation: Based on the feature vector and the content feature matrix, correction coefficients are generated through a nonlinear mapping engine. The nonlinear mapping engine includes a base mapping layer, a dynamic correction layer, and an output constraint layer. The nonlinear mapping engine first establishes an initial correction benchmark through the base mapping layer, then calculates the brightness correction amount through the dynamic correction layer and determines whether it belongs to the defined range. Finally, the output constraint layer generates the correction coefficients. The dynamic correction layer processes the data and determines whether its output result belongs to the defined range. If it does, the output constraint layer processes the data; otherwise, it returns to the dynamic correction layer to recalculate. Hardware-level real-time execution: Real-time control is performed based on the correction coefficients, and during the execution process, it is determined whether the consistency of the correction coefficients in adjacent regions meets the preset threshold. If so, basic adjustment operations are performed; otherwise, global synchronization adjustment operations are performed.

2. The intelligent control method for LED display splicing according to claim 1, characterized in that, The native feature extraction process includes: Physical dimension feature extraction: Real-time acquisition of the operating current I(t) and junction temperature T(t) of each LED, calculation of the current change rate ΔI / Δt and temperature change rate ΔT / Δt, and generation of the feature vector [I(t), T(t), ΔI / Δt, ΔT / Δt]; Content dimension feature extraction: Calculate the pixel value change rate R(t) in each display frame to generate the content feature matrix; Feature association: The feature vector is associated with and stored in relation to the content feature matrix.

3. The intelligent control method for LED display splicing according to claim 2, characterized in that, The current change rate ΔI / Δt is the ratio of the current difference between adjacent sampling periods to the sampling period, and the temperature change rate ΔT / Δt is the ratio of the temperature difference between adjacent sampling periods to the sampling period. The current sampling accuracy is 1μA, and the temperature sampling accuracy is ±0.5℃.

4. The intelligent control method for LED display splicing according to claim 2, characterized in that, The pixel value change rate R(t) is calculated by the variance of the pixel difference of three consecutive frames. The content feature matrix includes high-speed motion area identifiers and static area identifiers, where R(t) > 0.8 for high-speed motion areas and R(t) < 0.2 for static areas. The pixel activity change ΔR is the difference in pixel value change rate between adjacent frames, i.e., ΔR = |R(t) - R(t-1)|.

5. The intelligent control method for LED display splicing according to claim 1, characterized in that, The dynamic coupling correction coefficient generation process includes: Basic mapping layer processing: Based on the photoelectric characteristic curve L of the LED at the time of manufacture. o =f(I o, T o Establish an initial calibration reference, where L o For standard brightness, I o T is the rated operating current of the LED chip. o The rated junction temperature of the LED chip; Dynamic correction layer processing: The brightness correction amount ΔL(t) is calculated using a bivariate error compensation function, which is: ΔL(t)=α·(I(t)-I o )·(T(t)-T o )+β·R(t)·∫ΔL(t-1)dt where ΔL(t) is the brightness correction amount at time t, α and β are self-learning coefficients, I(t) is the working current of the LED at time t, T(t) is the junction temperature of the LED at time t, R(t) is the pixel value change rate at time t, and ∫ΔL(t-1)dt is the cumulative value of the brightness correction amount before time t-1; Range determination and processing: Determine whether the output ΔL(t) of the dynamic correction layer belongs to [0.95L]. o 1.05L o If the value falls within the range [0.95L], then proceed to the output constraint layer for processing; otherwise, return to the dynamic correction layer, attenuate the α and β coefficients by 10%, and recalculate until ΔL(t) belongs to the range [0.95L]. o 1.05L o ] interval; Output constraint layer processing: Limits ΔL(t) after range judgment to [0.95L]. o 1.05L o The interval is used to generate the final correction coefficient.

6. The intelligent control method for LED display splicing according to claim 1, characterized in that, In the hardware-level real-time execution process: The basic adjustment operation includes: acquiring I(t), T(t) data and R(t) characteristics every 10ms, adjusting the PWM duty cycle of the drive current based on the correction coefficient, with an adjustment accuracy of 1‰, and maintaining an end-to-end delay of <15ms; The global synchronization adjustment operation includes: triggering a global synchronization command, re-executing the dynamic coupling correction coefficient generation step based on a unified time reference to generate synchronization correction coefficients, and then performing the basic adjustment operation based on the synchronization correction coefficients.

7. The intelligent control method for LED display splicing according to claim 5, characterized in that, The update mechanisms for α and β include: When the pixel activity change ΔR > 0.5 or the temperature change rate ΔT / Δt > 2°C / s, update immediately; otherwise, the update cycle is adaptively adjusted according to the cumulative error. The cycle is 2 hours when the error is < 0.05% and 10 minutes when the error is > 0.1%. The initial value is 0.5 for both.

8. An intelligent control system for LED display splicing, characterized in that, include: The feature extraction module is used to extract raw data through two-dimensional feature capture, and generate feature vectors that reflect the real-time photoelectric characteristics of the LED beads and content feature matrices that reflect the dynamic characteristics of the displayed content. The correction coefficient generation module is used to generate correction coefficients based on the feature vector and the content feature matrix through a nonlinear mapping engine. The nonlinear mapping engine includes a basic mapping layer, a dynamic correction layer and an output constraint layer. After processing by the dynamic correction layer, it is determined whether the output result belongs to the limited range. If so, the output constraint layer processing is executed. Conversely, if the result is not satisfactory, the calculation is returned to the dynamic correction layer for recalculation. The execution control module is used to perform real-time control based on the correction coefficient, and during the execution process, it determines whether the consistency of the correction coefficients in adjacent areas meets a preset threshold. If so, it performs basic adjustment operations. Conversely, a global synchronization adjustment operation is performed.

9. An electronic device, characterized in that, The electronic device includes: The processor and the memory are communicatively connected. The memory is used to store at least one executable instruction executed by the processor, the processor being used to execute the executable instruction to implement the LED display splicing intelligent control method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the LED display splicing intelligent control method as described in any one of claims 1 to 7.

Images