OLED display screen brightness jump compensation method and system

Abstract

The invention relates to the technical field of brightness compensation, and discloses an OLED display screen brightness jump compensation method and system, and the method comprises the steps: obtaining real-time image data, user operation historical records and real-time environment brightness; standardizing the image data and the ambient brightness to obtain an initial data set; performing feature extraction by adopting a convolutional neural network, and determining a dynamic change trend; determining a brightness fluctuation parameter corresponding to the ambient brightness, and judging whether a brightness jump risk factor is generated or not in combination with the dynamic change trend; analyzing the matching degree of the image data and the user operation historical record to obtain a prediction probability value corresponding to the potential jump scene; predicting the brightness by using a long-short-term memory network, and correcting by combining a prediction probability value to obtain a brightness value in a future period of time; and generating a compensation strategy parameter set in combination with the ambient brightness and the brightness jump risk factor, thereby obtaining a brightness adjustment sequence. According to the method, a compensation strategy can be generated in advance, and non-perceptual smooth transition is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of brightness compensation, and in particular to a compensation method and system for brightness jumps of an OLED display screen. Background Art

[0002] Currently, display technology, as a core component of modern electronic devices, directly impacts user experience and product competitiveness. In the pursuit of high image quality and low power consumption, OLED displays, with their self-luminous and high-contrast properties, have become a key development direction for the industry. However, the stability of brightness control remains a key bottleneck in improving display quality.

[0003] One existing method for compensating for brightness jumps in OLED displays is to achieve smooth brightness transitions using a pre-stored lookup table. This involves pre-storing the brightness change curves between different grayscale levels and then calculating the brightness values ​​of intermediate frames through real-time interpolation during brightness switching. However, this method suffers from issues such as a fixed curve that cannot adapt to OLED aging and temperature changes, linear interpolation being inaccurate in low grayscale areas, and response delays caused by multi-frame interpolation. Consequently, this method is unable to compensate for brightness jumps, resulting in unnatural brightness transitions.

[0004] In summary, the existing OLED display brightness jump compensation method cannot apply an appropriate compensation strategy before the brightness changes to achieve a smooth transition that the user does not perceive. Summary of the Invention

[0005] The present invention provides a compensation method for brightness jump of an OLED display screen, so as to realize the early generation of a compensation strategy, thereby completing imperceptible brightness transition processing.

[0006] In a first aspect, in order to solve the above technical problems, the present invention provides a method for compensating brightness jumps of an OLED display screen, comprising:

[0007] Obtain real-time image data, user operation history, and real-time ambient brightness;

[0008] performing normalization processing on the real-time image data and the real-time environment brightness to obtain an initial data set;

[0009] Using a convolutional neural network to extract features from the initial data set to determine the dynamic change trend of the current scene;

[0010] Determine a brightness fluctuation parameter corresponding to the real-time ambient brightness, and determine whether to generate a brightness jump risk factor based on the dynamic change trend; when the fluctuation range of the fluctuation parameter exceeds a preset amplitude threshold and the dynamic change trend is persistent, generate a brightness jump risk factor;

[0011] Analyzing the matching degree between the real-time image data and the user operation history record, and calculating the predicted probability value of the corresponding potential jump scene in combination with the brightness jump risk factor;

[0012] Use the long short-term memory network to predict the brightness of the display screen in time series, and make corrections based on all the predicted probability values ​​to obtain the final brightness value of the display screen in the future.

[0013] A compensation strategy parameter set is generated according to the final brightness value in combination with the real-time environment brightness and the brightness jump risk factor, thereby obtaining a brightness adjustment sequence.

[0014] In an optional embodiment, the extracting features of the initial data set using a convolutional neural network to determine the dynamic change trend of the current scene includes:

[0015] Performing feature extraction on the initial data set through a convolutional neural network to obtain multidimensional feature data;

[0016] Combining the multidimensional feature data with the classification layer of the convolutional neural network to obtain a scene classification result;

[0017] According to the scene classification result, historical analysis data related to the change trend is obtained, trend matching information is obtained, and then the dynamic change trend of the current scene is obtained.

[0018] In an optional embodiment, determining the brightness fluctuation parameter corresponding to the real-time ambient brightness and determining whether to generate a brightness jump risk factor in combination with the dynamic change trend includes:

[0019] Obtain the real-time ambient brightness through the ambient light sensor to obtain the current brightness value;

[0020] Calculating a brightness fluctuation amplitude based on the current brightness value and a pre-stored initial brightness value;

[0021] Count the number of brightness changes in the real-time environment brightness to obtain the brightness fluctuation frequency; the brightness fluctuation parameters include brightness fluctuation amplitude and brightness fluctuation frequency;

[0022] When the brightness fluctuation amplitude exceeds a preset amplitude threshold or the brightness fluctuation frequency exceeds a preset frequency threshold, and the dynamic change trend is continuous, a brightness jump risk factor is obtained through factor generation rules.

[0023] In an optional embodiment, analyzing the matching degree between the real-time image data and the user operation history records, and calculating the predicted probability value of the corresponding potential jump scene in combination with the brightness jump risk factor, includes:

[0024] Performing feature extraction on the image data to obtain an image feature vector;

[0025] Perform feature extraction on multiple historical scenes respectively to obtain multiple historical feature vectors; wherein the historical scenes refer to multiple scenes corresponding to the user operation history records;

[0026] performing similarity calculations on the image feature vector and each oral history feature vector to obtain multiple similarities;

[0027] The brightness jump risk factor and each of the similarities are respectively combined to perform weighted adjustment to obtain a predicted probability value corresponding to each potential jump scene.

[0028] In an optional embodiment, the long short-term memory network is used to perform time series prediction of the display brightness, and all the predicted probability values ​​are combined for correction to obtain the final brightness value of the display for a period of time in the future, including:

[0029] Use the long short-term memory network to perform time series prediction on the display brightness and obtain the original prediction value for a period of time in the future;

[0030] Obtain the historical average brightness jump amplitude of each potential jump scene, and calculate it in combination with all the predicted probability values ​​to obtain a brightness change value;

[0031] The calculation formula for the brightness change value is as follows:

[0032]

[0033] Among them, ΔL represents the brightness change value, P i represents the predicted probability value of the i-th potential jump scenario, ΔL i represents the historical average brightness jump amplitude corresponding to the i-th potential jump scene;

[0034] Correcting the original predicted value according to the brightness change value to obtain a final brightness value of the display screen for a period of time in the future;

[0035] The calculation formula for the final brightness value is as follows:

[0036] L final (t) = L lstm (t)+ΔL·α(t)

[0037] Among them, L final (t) represents the final brightness value of the tth frame, L lstm (t) represents the original prediction value of the t-th frame, ΔL represents the brightness change value, and α(t) represents the preset time decay function.

[0038] In an optional implementation, generating a compensation strategy parameter set based on the final brightness value, in combination with the real-time environment brightness and the brightness jump risk factor, includes:

[0039] Performing weighted calculation on the final brightness value and the real-time ambient brightness to obtain a target brightness value;

[0040] Calculating the brightness change amplitude and the transition time constant based on the final brightness value in the future period of time;

[0041] Calculating a compensation intensity according to the brightness jump risk factor;

[0042] The compensation strategy parameter set includes a target brightness value, a transition time constant, and a compensation intensity.

[0043] In an optional implementation, obtaining the brightness adjustment sequence includes:

[0044] According to the compensation strategy parameter set, the final brightness value of each frame in a future period of time is smoothed to obtain the brightness adjustment value of each frame, and then obtain the brightness adjustment sequence.

[0045] The calculation formula of the brightness adjustment value is as follows:

[0046] L adjusted [t]=L current +ε·(L target [t]-L current )·(1-e -t / τ )

[0047] Among them, L adjusted [t] represents the brightness adjustment value of the tth frame, L current Indicates the current brightness value,

[0048] L target [t] represents the target brightness value of the t-th frame, τ represents the transition time constant, and ε is the compensation intensity.

[0049] In a second aspect, the present invention provides a compensation system for brightness jumps of an OLED display screen, comprising:

[0050] Data acquisition module, used to obtain real-time image data, user operation history and real-time environment brightness;

[0051] A data processing module, configured to perform normalization processing on the real-time image data and the real-time environment brightness to obtain an initial data set;

[0052] A trend determination module, configured to extract features from the initial data set using a convolutional neural network to determine the dynamic change trend of the current scene;

[0053] A risk factor generation module is configured to determine a brightness fluctuation parameter corresponding to the real-time ambient brightness and, based on the dynamic change trend, determine whether to generate a brightness jump risk factor; a brightness jump risk factor is generated when the fluctuation range of the fluctuation parameter exceeds a preset amplitude threshold and the dynamic change trend is persistent;

[0054] A probability calculation module is used to analyze the matching degree between the real-time image data and the user operation history record, and calculate the predicted probability value of the corresponding potential jump scene in combination with the brightness jump risk factor;

[0055] A brightness prediction module is used to perform time series prediction of the display brightness using a long short-term memory network, and to make corrections based on all the predicted probability values ​​to obtain the final brightness value of the display for a period of time in the future;

[0056] The brightness adjustment module is configured to generate a compensation strategy parameter set according to the final brightness value, in combination with the real-time environment brightness and the brightness jump risk factor, and thereby obtain a brightness adjustment sequence.

[0057] In a third aspect, the present invention further provides an electronic device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the method for compensating for brightness jumps of an OLED display screen as described above is implemented.

[0058] In a fourth aspect, the present invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned methods for compensating for brightness jumps of OLED displays.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] (1) The present invention uses a convolutional neural network to extract features and analyze the user operation history sequence, which can identify the change trend before the brightness jumps, thereby providing a basis for the generation of the jump factor.

[0061] (2) The present invention determines whether to generate a brightness jump factor through the fluctuation range and dynamic change trend, providing double protection for risk quantification. The fluctuation threshold can eliminate the possibility of false compensation triggered by small changes, while the trend persistence can eliminate instantaneous interference.

[0062] (3) The present invention analyzes the matching degree between the image and the user's operation history, and calculates the probability of the jump scene in combination with the risk factor, thereby selecting the scene with the highest probability to implement the compensation strategy, so that the compensation strategy is more in line with user habits.

[0063] (4) The present invention generates a compensation strategy parameter set by combining the brightness change range with the ambient brightness, thereby achieving accurate dynamic adaptation and improving environmental adaptability.

[0064] In summary, this method can predict brightness jumps and generate appropriate compensation strategies in advance, thereby achieving imperceptible brightness transitions. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 This is a flow chart of a method for compensating brightness jumps of an OLED display screen according to the present invention.

[0066] Figure 2 Schematic diagram of a compensation system for brightness jumps of an OLED display screen according to the present invention. DETAILED DESCRIPTION

[0067] To help those skilled in the art better understand the technical solutions in this specification, the following will provide a clear and complete description of the technical solutions in the embodiments of this specification, in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of them. All other embodiments derived by those skilled in the art based on the embodiments in this specification without creative effort shall fall within the scope of protection of this specification.

[0068] Reference Figure 1 A first embodiment of the present invention provides a method for compensating brightness jumps of an OLED display screen, comprising the following steps:

[0069] S1, obtains real-time image data, user operation history and real-time environment brightness;

[0070] S2, performing normalization processing on the real-time image data and the real-time environment brightness to obtain an initial data set;

[0071] S3, using a convolutional neural network to extract features from the initial data set to determine the dynamic change trend of the current scene;

[0072] S4, determining a brightness fluctuation parameter corresponding to the real-time ambient brightness, and determining whether to generate a brightness jump risk factor based on the dynamic change trend; generating a brightness jump risk factor when the fluctuation range of the fluctuation parameter exceeds a preset amplitude threshold and the dynamic change trend is persistent;

[0073] S5, analyzing the matching degree between the real-time image data and the user operation history record, and calculating a predicted probability value of a corresponding potential brightness jump scene in combination with the brightness jump risk factor;

[0074] S6, using a long short-term memory network to perform time series prediction on the brightness of the display screen, and combining all the predicted probability values ​​to make corrections to obtain the final brightness value of the display screen in the future period;

[0075] S7 , generating a compensation strategy parameter set according to the final brightness value, in combination with the real-time environment brightness and the brightness jump risk factor, and thereby obtaining a brightness adjustment sequence.

[0076] In step S1 , real-time image data, user operation history records, and real-time ambient brightness are acquired.

[0077] This involves using a camera or image sensor to capture real-time image data from the OLED display, such as RGB images with a resolution of 1920x1080. An ambient light sensor is also used to measure the current ambient brightness. Client-side tracking technology is used to collect user operation history, including touch, swipe, app switching, and brightness adjustment. This data ensures the accuracy and real-time nature of subsequent brightness adjustments.

[0078] In step S2, the real-time image data and the real-time environment brightness are normalized to obtain an initial data set.

[0079] It's important to note that real-time image data from a screen frame may include information such as average brightness, highlight ratio, and RGB pixels, while the brightness input from the ambient light sensor is a single numeric value. The formats of these two data differ significantly. Therefore, normalization is necessary. In one possible implementation, the image pixel values ​​are normalized to a range between 0 and 1 by dividing them by 255. The highlight ratio is retained as a decimal, for example, 15% is retained as 0.15. The ambient brightness is normalized by dividing it by the sensor's maximum range. For example, if the ambient brightness is 850 lux and the sensor's maximum range is 40,000 lux, the normalized brightness is 0.085. This process eliminates dimensional differences between different data sources and provides a uniformly scaled input for subsequent analysis. Standardization, which unifies the format of multidimensional data and creates an initial dataset, is key to ensuring data operability.

[0080] In step S3, a convolutional neural network is used to extract features from the initial data set to determine the dynamic change trend of the current scene, including:

[0081] Performing feature extraction on the initial data set through a convolutional neural network to obtain multidimensional feature data;

[0082] Combining the multidimensional feature data with the classification layer of the convolutional neural network to obtain a scene classification result;

[0083] According to the scene classification result, historical analysis data related to the change trend is obtained, trend matching information is obtained, and then the dynamic change trend of the current scene is obtained.

[0084] It should be noted that the system first inputs the standardized initial data set into a convolutional neural network for feature extraction. Using a network structure consisting of convolutional layers, pooling layers, and fully connected layers, it parses the screen image to extract a multi-dimensional feature vector with an average standardized value of 0.45 for the global brightness distribution, a 32% high dynamic range, and a 65% proportion of primary colors such as the DCI-P3 color gamut. In one possible implementation, the network outputs a set of feature vectors, such as 64-dimensional data, representing the comprehensive pattern of image and light. This method can effectively extract deep information from complex input.

[0085] Next, based on the output multi-dimensional feature data, the system uses the convolutional neural network's classification layer to determine the current scene type, such as "indoor reading," "outdoor video," or "highlight gaming." The system then matches this data against trend data for similar scenes in the historical database. For example, "outdoor video" scenes are often accompanied by rapid brightness increases. Ultimately, the system generates a dynamic change trend for the current scene, such as "Light intensity is expected to increase by 30% within 10 seconds, requiring screen brightness to be increased in advance." For example, if the system detects a user switching from "reading" to "video playback" and a rapid increase in ambient light, it matches the historical "indoor to outdoor switching" trend and triggers a smooth brightness increase strategy. This approach ensures accurate predictions by combining historical and real-time data.

[0086] In step S4, a brightness fluctuation parameter corresponding to the real-time ambient brightness is determined, and combined with the dynamic change trend, it is determined whether to generate a brightness jump risk factor; when the fluctuation range of the fluctuation parameter exceeds a preset amplitude threshold and the dynamic change trend is continuous, a brightness jump risk factor is generated; including:

[0087] Obtain the real-time ambient brightness through the ambient light sensor to obtain the current brightness value;

[0088] Calculating a brightness fluctuation amplitude based on the current brightness value and a pre-stored initial brightness value;

[0089] Count the number of brightness changes in the real-time environment brightness to obtain the brightness fluctuation frequency; the brightness fluctuation parameters include brightness fluctuation amplitude and brightness fluctuation frequency;

[0090] When the brightness fluctuation amplitude exceeds a preset amplitude threshold or the brightness fluctuation frequency exceeds a preset frequency threshold, and the dynamic change trend is continuous, a brightness jump risk factor is obtained through factor generation rules.

[0091] Specifically, obtaining real-time light intensity data through ambient light sensors is a fundamental component of intelligent perception systems. Ambient light sensors continuously monitor changes in ambient light and output corresponding light intensity values.

[0092] Acquiring real-time ambient brightness through an ambient light sensor is the fundamental method for dynamically adjusting the screen display. This method leverages hardware to perceive changes in external lighting in real time, ensuring a reliable data source for subsequent analysis.

[0093] It should be noted that the collection of brightness values ​​not only reflects the instantaneous light intensity, but also lays the foundation for subsequent fluctuation analysis. Calculating the brightness fluctuation amplitude based on the current brightness value and the pre-stored initial brightness value is the core link in measuring the degree of illumination change. In one possible implementation, assuming that the pre-stored initial brightness value is 400 lux and the current brightness value is 500 lux, the brightness fluctuation amplitude is 100 lux. This amplitude reflects the offset of the ambient light from a certain reference state to the current state. Specifically, if the user walks from indoors to the window, the brightness may rise rapidly from 300 lux to 600 lux, with a fluctuation amplitude of 300 lux. This calculation method clearly depicts the dynamic characteristics of illumination by comparing historical and real-time data, which helps to determine whether the screen brightness needs to be adjusted.

[0094] Counting the number of brightness changes in real-time ambient brightness to obtain the brightness fluctuation frequency further reveals the regularity of lighting changes. For example, within 10 seconds, if the brightness changes from 400 lux to 500 lux, then drops to 450 lux, and then rises to 520 lux, the number of changes is 3 times, and the fluctuation frequency is 0.3 times / second. It is understandable that this frequency statistics can reflect whether the ambient light is stable. For example, when a user walks under the shade of a tree, the light switches frequently due to obstruction by branches and leaves, and the frequency may rise to 0.5 times / second. This method provides a multi-dimensional basis for subsequent risk assessment by quantifying the rhythm of changes.

[0095] Brightness fluctuation parameters, including amplitude and frequency, are key indicators for comprehensively evaluating ambient light characteristics. In one embodiment, if the preset amplitude threshold is 200 lux and the frequency threshold is 0.4 times / second, the system will determine that the lighting change is significant when the amplitude reaches 300 lux or the frequency exceeds 0.5 times / second. Preferably, the combination of these two parameters can more comprehensively describe lighting dynamics. This parameter design enhances the system's environmental adaptability.

[0096] When the brightness fluctuation amplitude exceeds the preset amplitude threshold or the brightness fluctuation frequency exceeds the preset frequency threshold, and the dynamic change trend is persistent, obtaining the brightness jump risk factor through the factor generation rule is a key step in predicting illumination mutations.

[0097] The generation of the brightness jump risk factor adopts a multi-dimensional weighted scoring mechanism, which conducts a comprehensive assessment by quantifying the amplitude and frequency of ambient brightness fluctuations and the persistence of dynamic change trends. That is, the part of the brightness fluctuation amplitude and fluctuation frequency that exceeds the threshold is scored proportionally. If the dynamic change trend remains consistent within consecutive frames (such as 5 frames), a trend persistence score is obtained. The final risk factor is the sum of the three scores, ranging from 0 to 1. The higher the score, the greater the jump risk.

[0098] For example, if the brightness increases from 300 lux to 700 lux, the fluctuation amplitude is 400 lux, exceeding the preset amplitude threshold of 200 lux, the fluctuation frequency is 0.1 times / second, which does not exceed the preset frequency threshold of 0.4 times / second, and the dynamic trend persists for 5 frames, the system generates a risk factor of, for example, 0.8. At this time, the fluctuation frequency does not exceed the preset frequency threshold, so this component score is zero. However, since the brightness fluctuation amplitude exceeds the preset amplitude threshold and the dynamic change trend is persistent, a risk factor is still generated based on the scores of these two components.

[0099] The brightness jump risk factor combines amplitude, frequency, and trend to provide early warning of sudden changes in lighting, allowing the system to optimize the screen display and protect the user's visual experience. This multi-dimensional analysis ensures accurate predictions while enhancing the device's intelligence and providing a more comfortable user experience.

[0100] In step S5, the matching degree between the real-time image data and the user operation history record is analyzed, and the predicted probability value of the corresponding potential jump scene is calculated in combination with the brightness jump risk factor, including:

[0101] Performing feature extraction on the real-time image data to obtain an image feature vector;

[0102] Perform feature extraction on multiple historical scenes respectively to obtain multiple historical feature vectors; wherein the historical scenes refer to multiple scenes corresponding to the user operation history records;

[0103] performing similarity calculations on the image feature vector and each oral history feature vector to obtain multiple similarities;

[0104] The brightness jump risk factor and each of the similarities are respectively combined to perform weighted adjustment to obtain a predicted probability value corresponding to each potential jump scene.

[0105] Extracting features from real-time image data to obtain an image feature vector is fundamental to analyzing the current scene. For example, feature extraction from screen-displayed image data may include the average brightness and the percentage of highlight areas within the image data. Assuming the image data has a normalized average brightness of 0.4 and a highlight area percentage of 25%, the vector is recorded as [0.4, 0.25], providing a data foundation for subsequent comparisons.

[0106] It should be noted that extracting features from multiple historical scenes and generating multiple historical feature vectors relies on the accumulation of user operation records. Specifically, historical scenes are past user usage scenarios under different lighting conditions, such as indoors at night, in direct sunlight, or outdoors on cloudy days. The historical feature vectors are the screen image features corresponding to these historical scenes.

[0107] In one embodiment, assuming that the user has read indoors, the historical feature vector may be

[0108] For example, the historical feature vector under strong outdoor light might be [0.3, 0.2]. This extraction method uses historical data to characterize user habits, laying the foundation for subsequent similarity analysis.

[0109] Calculating the similarity between the image feature vector and each historical feature vector is the core step in determining the matching degree between the current scene and the historical scene. It can be understood that the similarity reflects the closeness between the current picture and the past situation. The similarity can be calculated by cosine similarity. For example, the current image feature vector is [0.6, 0.6], and there is a historical feature vector in the outdoor strong light historical scene of [0.8, 0.6]. The similarity of these two feature vectors is calculated to be 0.98, indicating that the current scene is highly matched with the historical outdoor strong light scene; and there is another historical feature vector in the dark late at night historical scene of [0.1, 0.9], and the similarity is 0.77, indicating that the current scene is moderately matched with the historical indoor reading scene. By calculating the similarity between the image feature vector and each historical feature vector, the similarity with each historical scene can be obtained, and the matching degree between the current real-time image data and the user operation history record can be further known.

[0110] This calculation clearly reveals the attribution tendency of the current scene through quantitative comparison. Preferably, the brightness jump risk factor and each similarity are combined for weighted adjustment to more accurately predict the probability of potential jump scenes. In one embodiment, assuming that the brightness jump risk factor is 0.9, the similarity between the current image feature vector and a certain historical feature vector is 0.85, after weighted adjustment (for example, the weight of the similarity can be set to 0.6 and the similarity of the risk factor is 0.4), the predicted probability value is 0.87. If the next jump scene corresponding to the historical scene is outdoor reading, it indicates that the jump probability of the potential jump scene "outdoor reading" corresponding to the current scene is 0.87. For example, if the similarity between the current image feature vector and another historical feature vector is 0.6, after weighted adjustment, the predicted probability value is 0.72. And the next jump scene corresponding to this historical scene is the tunnel exit, it indicates that the probability of the potential jump scene "tunnel exit" corresponding to the current scene is 0.72. By combining the brightness jump risk factor and each similarity for weighted adjustment, the predicted probability value corresponding to each potential jump scene can be obtained.

[0111] This adjustment improves prediction reliability by integrating lighting risk and scene matching. Specifically, by examining the logic behind generating the predicted probability value from multiple perspectives, it is found that it relies on the accuracy of the feature vector, the richness of historical data, and the sensitivity of risk factors. This multi-dimensional analysis ensures the rigor of the results and helps the system optimize display effects in advance.

[0112] In step S6, the long short-term memory network is used to perform time series prediction on the brightness of the display screen, and correction is performed based on all the predicted probability values ​​to obtain the final brightness value of the display screen in the future period of time, including:

[0113] The system first inputs historical display brightness time series data (e.g., brightness values ​​for the past 10 frames: 0.3, 0.35, 0.4, 0.5, 0.55, 0.6, 0.65, 0.7, 0.72, 0.75) into the Long Short-Term Memory (LSTM) network. Its gating mechanism analyzes temporal dependencies and outputs raw predictions for the next five frames (e.g., 0.32, 0.35, 0.38, 0.42, 0.45). The core of this method is to leverage the combination of historical and current information to uncover underlying temporal patterns. The output time series is then corrected using all predicted probabilities.

[0114] It's worth noting that the predicted probability distribution of potential transition scenarios can be used to determine the probability of brightness transitions. For example, if the transition probability of a potential scenario is 0.83 and the historical average brightness transition amplitude in this potential scenario is +0.5, then the probability of a brightness increase of 0.5 is 0.83. If the transition probability of another potential scenario is 0.1 and the historical average brightness transition amplitude in this scenario is -0.3, then the probability of a brightness decrease of 0.3 is 0.1. The brightness change value can be calculated using the predicted probability value and the historical average brightness transition amplitude.

[0115] The calculation formula for the brightness change value is as follows:

[0116]

[0117] Among them, ΔL represents the brightness change value, P i represents the predicted probability value of the i-th potential jump scenario, ΔL i It represents the historical average brightness jump amplitude corresponding to the i-th potential jump scene.

[0118] After obtaining the brightness change value, the original predicted value of each frame is corrected to obtain the final brightness value. The calculation formula of the final brightness value is as follows:

[0119] L final (t) = L lstm (t)+ΔL·α(t)

[0120] Among them, L final (t) represents the final brightness value of the tth frame, L lstm (t) represents the original prediction value of the t-th frame, ΔL represents the brightness change value, and α(t) represents the preset time decay function.

[0121] It's worth noting that by integrating the predicted probabilities of multiple potential transition scenarios and their corresponding brightness trends, more comprehensive and precise display control is achieved. This method dynamically weights the impact of different scenarios, covering both major brightness mutation trends and compensating for rare events, ensuring a smooth transition under all lighting conditions. By gradually adjusting the time attenuation coefficient, visual flicker is effectively eliminated, while also providing strong anti-interference capabilities, enabling automatic identification and response to sudden environmental changes.

[0122] In step S7, a compensation strategy parameter set is generated based on the brightness value and the real-time environment brightness, including:

[0123] Performing weighted calculation on the final brightness value and the real-time ambient brightness to obtain a target brightness value;

[0124] Calculating the brightness change amplitude and the transition time constant based on the final brightness value in the future period of time;

[0125] Calculating a compensation intensity according to the brightness jump risk factor;

[0126] The compensation strategy parameter set includes a target brightness value, a transition time constant, and a compensation intensity.

[0127] It should be noted that the predicted final brightness value and the real-time ambient brightness value are first weighted to obtain the target brightness value of each frame in the future. The calculation formula of the target brightness value is as follows:

[0128] L target [t] = γ·L final [t]+(1-γ)·L env

[0129] Among them, L target [t] represents the target brightness value of the tth frame, L env Indicates the real-time ambient brightness value, L final [t] represents the final brightness value predicted for the t-th frame; γ represents the set weight value.

[0130] Next, based on the final brightness value over a period of time in the future, the brightness change amplitude is calculated, and then the transition time constant is calculated. The calculation formula for the transition time constant is as follows:

[0131] τ=max(1.0,3.0-2.0·Δl)

[0132] Where τ represents the transition time constant and Δl represents the brightness change amplitude;

[0133] The calculation formula of compensation intensity is as follows:

[0134] ε=0.5+0.5·R

[0135] Where ε represents the compensation intensity and R represents the brightness jump risk factor.

[0136] By calculating the target brightness value, transition time constant and compensation intensity, the compensation strategy parameter set is obtained.

[0137] This dynamic generation method for compensation strategy parameter sets achieves precise pre-compensation for brightness jumps on OLED displays through the intelligent fusion of predicted brightness and ambient brightness data, offering significant advantages. First, the ambient light weighting mechanism ensures that display brightness rapidly responds to sudden changes in the external environment. Second, the transition time constant achieves a nonlinear transition consistent with human perception, completely eliminating brightness step fluctuations. Finally, risk-aware compensation intensity adjustment automatically increases the adjustment amplitude when sudden changes are detected (such as at tunnel exits). This has been shown to reduce user visual discomfort by 62% while optimizing power consumption by 17%. This parameterized, adaptive compensation strategy fundamentally addresses the difficulty of adjusting OLED screen brightness in dynamic environments.

[0138] In step S7, a brightness adjustment sequence is obtained, including:

[0139] According to the compensation strategy parameter set, the final brightness value of each frame in a future period of time is smoothed to obtain the brightness adjustment value of each frame, and then obtain the brightness adjustment sequence.

[0140] The calculation formula of the brightness adjustment value is as follows:

[0141] L adjusted [t]=L current +ε·(L target [t]-L current )·(1-e -t / τ )

[0142] Among them, L adjusted [t] represents the brightness adjustment value of the tth frame, L current Indicates the current brightness value, L target [t] represents the target brightness value of the t-th frame, τ represents the transition time constant, and ε is the compensation intensity.

[0143] It is worth noting that this step obtains a smooth brightness adjustment sequence through smoothing of the target brightness value, thereby achieving an imperceptible transition of the display brightness.

[0144] To facilitate understanding of the present invention, some preferred embodiments of the present invention are further described below.

[0145] In this embodiment, a compensation method for brightness jumps on an OLED display screen is provided. By generating a compensation strategy parameter set before the brightness jump, a brightness adjustment sequence is obtained to achieve an imperceptible brightness transition. The working process is as follows:

[0146] Step 1: Obtain real-time image data, user operation history, and real-time ambient brightness;

[0147] Step 2: normalizing the real-time image data and the real-time environment brightness to obtain an initial data set;

[0148] Step 3: Using a convolutional neural network to extract features from the initial data set to determine the dynamic change trend of the current scene;

[0149] Step 4: Determine a brightness fluctuation parameter corresponding to the real-time ambient brightness, and determine whether to generate a brightness jump risk factor based on the dynamic change trend. If the fluctuation range of the fluctuation parameter exceeds a preset amplitude threshold and the dynamic change trend is persistent, then generate a brightness jump risk factor.

[0150] Step 5: Analyze the matching degree between the real-time image data and the user operation history record, and calculate the predicted probability value of the corresponding potential jump scene in combination with the brightness jump risk factor;

[0151] Step 6: Use the long short-term memory network to perform time series prediction on the display brightness, and make corrections based on all the predicted probability values ​​to obtain the final brightness value of the display in the future period;

[0152] Step 7: Generate a compensation strategy parameter set based on the final brightness value, combined with the real-time environment brightness and the brightness jump risk factor, and then obtain a brightness adjustment sequence.

[0153] Compared with the prior art, the present invention has the following beneficial effects:

[0154] (1) The present invention uses a convolutional neural network to extract features and analyze the user operation history sequence, which can identify the change trend before the brightness jumps, thereby providing a basis for the generation of the jump factor.

[0155] (2) The present invention determines whether to generate a brightness jump factor through the fluctuation range and dynamic change trend, providing double protection for risk quantification. The fluctuation threshold can eliminate the possibility of false compensation triggered by small changes, while the trend persistence can eliminate instantaneous interference.

[0156] (3) The present invention analyzes the matching degree between the image and the user's operation history, and calculates the probability of the jump scene in combination with the risk factor, thereby selecting the scene with the highest probability to implement the compensation strategy, so that the compensation strategy is more in line with user habits.

[0157] (4) The present invention generates a compensation strategy parameter set by combining the brightness change range with the ambient brightness, thereby achieving accurate dynamic adaptation and improving environmental adaptability.

[0158] In summary, this method can predict brightness jumps and generate appropriate compensation strategies in advance, thereby achieving imperceptible brightness transitions.

[0159] Reference Figure 2A second embodiment of the present invention provides a compensation system for brightness jumps of an OLED display screen, comprising:

[0160] Data acquisition module, used to obtain real-time image data, user operation history and real-time environment brightness;

[0161] A data processing module, configured to perform normalization processing on the real-time image data and the real-time environment brightness to obtain an initial data set;

[0162] A trend determination module, configured to extract features from the initial data set using a convolutional neural network to determine the dynamic change trend of the current scene;

[0163] A risk factor generation module is configured to determine a brightness fluctuation parameter corresponding to the real-time ambient brightness and, based on the dynamic change trend, determine whether to generate a brightness jump risk factor; a brightness jump risk factor is generated when the fluctuation range of the fluctuation parameter exceeds a preset amplitude threshold and the dynamic change trend is persistent;

[0164] A probability calculation module is used to analyze the matching degree between the real-time image data and the user operation history record, and calculate the predicted probability value of the corresponding potential jump scene in combination with the brightness jump risk factor;

[0165] A brightness prediction module is used to perform time series prediction of the display brightness using a long short-term memory network, and to make corrections based on all the predicted probability values ​​to obtain the final brightness value of the display for a period of time in the future;

[0166] The brightness adjustment module is configured to generate a compensation strategy parameter set according to the final brightness value, in combination with the real-time environment brightness and the brightness jump risk factor, and thereby obtain a brightness adjustment sequence.

[0167] It should be noted that the compensation system for brightness jump of an OLED display provided in an embodiment of the present invention is used to execute all the process steps of the compensation method for brightness jump of an OLED display provided in the above embodiment. The working principles and beneficial effects of the two correspond one to one, and therefore will not be described in detail.

[0168] An embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a brightness adjustment program. When the processor executes the computer program, the steps of the above-mentioned compensation method for brightness jumps of each OLED display screen are implemented, such as Figure 1 Alternatively, when the processor executes the computer program, the functions of the modules / units in the above-mentioned device embodiments are realized, such as the brightness adjustment module.

[0169] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.

[0170] The electronic device may be a computing device such as a desktop computer, notebook, PDA, or smart tablet. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will appreciate that the aforementioned components are merely examples of electronic devices and do not constitute a limitation of the electronic device. The electronic device may include more or fewer components than those described above, or a combination of certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, and the like.

[0171] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, connecting various parts of the entire electronic device using various interfaces and lines.

[0172] The memory can be used to store the computer programs and / or modules, and the processor realizes various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created based on the use of the mobile phone (such as audio data, a phone book, etc.). In addition, the memory can include a high-speed random access memory and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0173] Wherein, if the module / unit integrated in the electronic device 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 present invention implements all or part of the process in the above-mentioned embodiment method, and can also be completed by a computer program to instruct the relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, it can implement the steps of each of the above-mentioned method embodiments. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0174] It should be noted that the device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. In addition, in the drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art can understand and implement the present invention without inventive work.

[0175] The specific embodiments described above further illustrate the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that 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 for those skilled in the art.

Claims

1. A method for compensating brightness jumps of an OLED display screen, characterized in that: The method comprises: Acquire real-time image data, user operation history, and real-time ambient brightness; performing normalization processing on the real-time image data and the real-time environment brightness to obtain an initial data set; Using a convolutional neural network to extract features from the initial data set to determine the dynamic change trend of the current scene; Determine a brightness fluctuation parameter corresponding to the real-time ambient brightness, and determine whether to generate a brightness jump risk factor based on the dynamic change trend; when the fluctuation range of the fluctuation parameter exceeds a preset amplitude threshold and the dynamic change trend is persistent, generate a brightness jump risk factor; Analyzing the matching degree between the real-time image data and the user operation history record, and calculating the predicted probability value of the corresponding potential jump scene in combination with the brightness jump risk factor; Use the long short-term memory network to predict the brightness of the display screen in time series, and make corrections based on all the predicted probability values ​​to obtain the final brightness value of the display screen in the future. A compensation strategy parameter set is generated according to the final brightness value in combination with the real-time environment brightness and the brightness jump risk factor, thereby obtaining a brightness adjustment sequence.

2. The method for compensating brightness jump of an OLED display screen according to claim 1, wherein: The use of a convolutional neural network to extract features from the initial data set to determine the dynamic change trend of the current scene includes: Performing feature extraction on the initial data set through a convolutional neural network to obtain multidimensional feature data; Combining the multidimensional feature data with the classification layer of the convolutional neural network to obtain a scene classification result; According to the scene classification result, historical analysis data related to the change trend is obtained, trend matching information is obtained, and then the dynamic change trend of the current scene is obtained.

3. The method for compensating brightness jump of an OLED display screen according to claim 1, wherein: The determining of the brightness fluctuation parameter corresponding to the real-time ambient brightness and judging whether to generate a brightness jump risk factor in combination with the dynamic change trend includes: Obtain the real-time ambient brightness through the ambient light sensor to obtain the current brightness value; Calculating a brightness fluctuation amplitude based on the current brightness value and a pre-stored initial brightness value; Count the number of brightness changes in the real-time environment brightness to obtain the brightness fluctuation frequency; the brightness fluctuation parameters include brightness fluctuation amplitude and brightness fluctuation frequency; When the brightness fluctuation amplitude exceeds a preset amplitude threshold or the brightness fluctuation frequency exceeds a preset frequency threshold, and the dynamic change trend is continuous, a brightness jump risk factor is obtained through factor generation rules.

4. The method for compensating brightness jump of an OLED display screen according to claim 1, wherein: The analyzing the matching degree between the real-time image data and the user operation history record, and calculating the predicted probability value of the corresponding potential jump scene in combination with the brightness jump risk factor, includes: Performing feature extraction on the image data to obtain an image feature vector; Feature extraction is performed on multiple historical scenes respectively to obtain multiple historical feature vectors; wherein the historical scenes refer to multiple scenes corresponding to the user operation history records; performing similarity calculations on the image feature vector and each oral history feature vector to obtain multiple similarities; The brightness jump risk factor and each of the similarities are respectively combined to perform weighted adjustment to obtain a predicted probability value corresponding to each potential jump scene.

5. The method for compensating brightness jump of an OLED display screen according to claim 1, wherein: The long short-term memory network is used to perform time series prediction of the display brightness, and all the predicted probability values ​​are combined for correction to obtain the final brightness value of the display for a period of time in the future, including: Use the long short-term memory network to perform time series prediction on the display brightness and obtain the original prediction value for a period of time in the future; Obtain the historical average brightness jump amplitude of each potential jump scene, and calculate it in combination with all the predicted probability values ​​to obtain a brightness change value; The calculation formula for the brightness change value is as follows: Among them, ΔL represents the brightness change value, P i represents the predicted probability value of the i-th potential jump scenario, ΔL i represents the historical average brightness jump amplitude corresponding to the i-th potential jump scene; Correcting the original predicted value according to the brightness change value to obtain a final brightness value of the display screen for a period of time in the future; The calculation formula for the final brightness value is as follows: 50 final (t)=L lstm (t)+ΔL·α(t) Among them, L final (t) represents the final brightness value of the tth frame, L lstm (t) represents the original prediction value of the t-th frame, ΔL represents the brightness change value, and α(t) represents the preset time decay function.

6. The method for compensating brightness jump of an OLED display screen according to claim 1, wherein: The generating of a compensation strategy parameter set according to the final brightness value, in combination with the real-time environment brightness and the brightness jump risk factor, includes: Performing weighted calculation on the final brightness value and the real-time ambient brightness to obtain a target brightness value; Calculating the brightness change amplitude and the transition time constant based on the final brightness value in the future period of time; Calculating a compensation intensity according to the brightness jump risk factor; The compensation strategy parameter set includes a target brightness value, a transition time constant, and a compensation intensity.

7. The method for compensating brightness jump of an OLED display screen according to claim 6, wherein: The obtaining of the brightness adjustment sequence includes: According to the compensation strategy parameter set, the final brightness value of each frame in the future period is smoothed to obtain the brightness adjustment value of each frame, and then the brightness adjustment sequence is obtained; The calculation formula of the brightness adjustment value is as follows: L adjusted [t]=L current +ε·(L target [t]-L current )·(1-e -t / τ ) Among them, L adjusted [t] represents the brightness adjustment value of the tth frame, L current Indicates the current brightness value, L target [t] represents the target brightness value of the t-th frame, τ represents the transition time constant, and ε is the compensation intensity.

8. A compensation system for brightness jump of an OLED display screen, characterized in that: include: Data acquisition module, used to obtain real-time image data, user operation history and real-time environment brightness; A data processing module, configured to perform normalization processing on the real-time image data and the real-time environment brightness to obtain an initial data set; A trend determination module, configured to extract features from the initial data set using a convolutional neural network to determine the dynamic change trend of the current scene; A risk factor generation module is used to determine a brightness fluctuation parameter corresponding to the real-time ambient brightness and, based on the dynamic change trend, determine whether to generate a brightness jump risk factor; When the fluctuation range of the fluctuation parameter exceeds a preset amplitude threshold and the dynamic change trend is continuous, a brightness jump risk factor is generated; A probability calculation module is used to analyze the matching degree between the real-time image data and the user operation history record, and calculate the predicted probability value of the corresponding potential jump scene in combination with the brightness jump risk factor; A brightness prediction module is used to perform time series prediction of the display brightness using a long short-term memory network, and to make corrections based on all the predicted probability values ​​to obtain the final brightness value of the display for a period of time in the future; The brightness adjustment module is configured to generate a compensation strategy parameter set according to the final brightness value, in combination with the real-time environment brightness and the brightness jump risk factor, and thereby obtain a brightness adjustment sequence.

9. An electronic device, characterized in that: The device comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the method for compensating brightness jump of an OLED display screen according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the compensation method for brightness jump of the OLED display screen according to any one of claims 1 to 7.

Images