Ink screen driving waveform debugging method, device and equipment and readable storage medium

Abstract

This application discloses a method, apparatus, device, and readable storage medium for debugging e-ink screen driving waveforms. The method includes: acquiring training data, which includes ambient temperature, each frame of the driving waveform corresponding to various display grayscale changes, and the corresponding e-ink screen display grayscale image; training a first preset neural network model based on the training data to obtain a display grayscale change prediction model, and pre-training and simulating environment reinforcement training a second preset neural network model based on the display grayscale change prediction model to obtain a first driving waveform prediction model; performing real-environment reinforcement training on the first driving waveform prediction model based on a target e-ink screen to obtain a second driving waveform prediction model; and determining the target driving waveform set for the target e-ink screen based on the second driving waveform prediction model, the target ambient temperature set, and the target display grayscale change set. This improves the efficiency and accuracy of e-ink screen driving waveform debugging.

Description

Technical Field

[0001] This application relates to the field of e-ink display technology, and in particular to an e-ink display driving waveform debugging method, apparatus, device and readable storage medium. Background Technology

[0002] E-ink displays, due to their advantages such as bistable operation, low power consumption, and paper-like appearance, have been widely used in e-book readers, digital signage, industrial labels, and wearable devices. Their display quality is highly dependent on the design of the driving waveform.

[0003] Currently, the debugging of the driving waveform for e-ink screens still mainly relies on engineers repeatedly adjusting the voltage amplitude, timing, and polarity combination based on batch samples using oscilloscopes and visual evaluation. Debugging a single screen often takes several hours to several days, resulting in low debugging efficiency. Furthermore, the subjectivity of manual debugging is relatively strong, which can affect the accuracy of the driving waveform. Summary of the Invention

[0004] In view of this, the purpose of this application is to overcome the shortcomings of the prior art and provide a method for debugging the driving waveform of an e-ink screen, the method comprising: Acquire training data, which includes: ambient temperature, each frame of the driving waveform corresponding to various display grayscale changes, and the e-ink screen display grayscale image corresponding to each frame of the waveform. Based on the training data, the first preset neural network model is trained to obtain a grayscale change prediction model, and the second preset neural network model is pre-trained and subjected to simulated environment reinforcement training based on the grayscale change prediction model to obtain a first driving waveform prediction model. Based on the target e-ink screen, the first driving waveform prediction model is subjected to real-environment reinforcement training to obtain the second driving waveform prediction model. Based on the second driving waveform prediction model, the target ambient temperature set, and the target display grayscale change set, the target driving waveform set of the target e-ink screen is determined.

[0005] In one embodiment, the step of acquiring training data includes: Each frame of the driving waveform corresponding to the grayscale change is sent to the e-ink screen in sequence to drive the e-ink screen to achieve the grayscale change; After sending each waveform segment for a preset duration, the corresponding grayscale image is captured and displayed on the e-ink screen. The ambient temperature is acquired, and the ambient temperature, each frame of the driving waveform corresponding to the grayscale change is bound together with the grayscale image of the e-ink screen corresponding to each frame of the waveform to obtain training data.

[0006] In one embodiment, the step of training a first preset neural network model based on the training data to obtain a display grayscale change prediction model, and then pre-training and performing simulated environment reinforcement training on a second preset neural network model based on the display grayscale change prediction model to obtain a first driving waveform prediction model, includes: Based on the training data, the first preset neural network model is trained to establish the neural representation between each frame of the driving waveform corresponding to the grayscale change and the corresponding grayscale image of the e-ink screen, so as to obtain the grayscale change prediction model. Based on the display grayscale change prediction model, a simulation training dataset and a simulation verification dataset are constructed, and a second preset neural network model is designed. The second preset neural network model is pre-trained based on the simulation training dataset, and the display grayscale change prediction model is used as a simulation environment to strengthen the training of the second preset neural network model until the output error of the second preset neural network model is verified to be less than a first preset threshold based on the simulation verification dataset, thus obtaining the first driving waveform prediction model.

[0007] In one embodiment, the step of performing real-world environment reinforcement training on the first driving waveform prediction model based on the target e-ink screen to obtain a second driving waveform prediction model includes: The first driving waveform prediction model responds to the target display grayscale change instruction for the target e-ink screen and outputs a reference driving waveform and a reference grayscale sequence corresponding to the reference driving waveform. The reference driving waveform is sent to the target e-ink screen to drive the target e-ink screen to achieve grayscale change, and the set of grayscale images of the target e-ink screen is collected; A target grayscale sequence is determined based on the set of displayed grayscale images, and the model parameters of the first driving waveform prediction model are adjusted based on the target grayscale sequence and the reference grayscale sequence to obtain a second driving waveform prediction model.

[0008] In one embodiment, the step of adjusting the model parameters of the first driving waveform prediction model based on the target grayscale sequence and the reference grayscale sequence to obtain the second driving waveform prediction model includes: Based on the target grayscale sequence and the reference grayscale sequence, a grayscale error sequence is calculated, and each grayscale error in the grayscale error sequence is compared with a second preset threshold. If there is a grayscale error greater than the second preset threshold in the grayscale error sequence, a reward vector is generated based on the grayscale error sequence, and the model parameters of the first driving waveform prediction model are adjusted based on the reward vector. The second driving waveform prediction model is obtained by determining that each grayscale error in the grayscale error sequence is less than the second preset threshold.

[0009] In one embodiment, the step of determining the target driving waveform set of the target e-ink screen based on the second driving waveform prediction model, the target ambient temperature set, and the target display grayscale change set includes: Based on the target ambient temperature set and the target display grayscale change set, input data containing the target ambient temperature and the target display grayscale change is constructed; The input data is fed into the second driving waveform prediction model. The second driving waveform prediction model determines each frame waveform segment of the target driving waveform based on the target ambient temperature and the target display grayscale change, and outputs the target driving waveform. The target driving waveform set of the target e-ink screen is obtained by predicting the target ambient temperature set and the target display grayscale change set based on the second driving waveform prediction model.

[0010] In one embodiment, the method further includes: Based on the target ambient temperature set, the target display grayscale change set, and the target driving waveform set, a driving waveform mapping table is constructed; The driving waveform mapping table is stored in the target e-ink screen so that the target e-ink screen can query and trigger the corresponding driving waveform based on the driving waveform mapping table to achieve grayscale changes.

[0011] This application also provides an e-ink screen driving waveform debugging device, the e-ink screen driving waveform debugging device comprising: The acquisition module is used to acquire training data, which includes: ambient temperature, each frame of the driving waveform corresponding to various display grayscale changes, and the e-ink screen display grayscale image corresponding to each frame of the waveform. The first training module is used to train a first preset neural network model based on the training data to obtain a display grayscale change prediction model, and to perform pre-training and simulated environment reinforcement training on a second preset neural network model based on the display grayscale change prediction model to obtain a first driving waveform prediction model; the second training module is used to perform real-environment reinforcement training on the first driving waveform prediction model based on the target e-ink screen to obtain a second driving waveform prediction model. The determination module is used to determine the target driving waveform set of the target e-ink screen based on the second driving waveform prediction model, the target ambient temperature set, and the target display grayscale change set.

[0012] This application also provides a computer device, which includes a processor and a memory, the memory storing a computer program, and the processor executing the computer program to implement the above-described e-ink screen driving waveform debugging method.

[0013] This application also provides a computer-readable storage medium storing a computer program that executes the above-described e-ink screen driving waveform debugging method when run on a processor.

[0014] The embodiments of this application have the following beneficial effects: This application embodiment obtains training data, which includes: ambient temperature, each frame of the driving waveform corresponding to various display grayscale changes, and the grayscale image of the e-ink screen display corresponding to each frame of the waveform. Based on the training data, a first preset neural network model is trained to obtain a display grayscale change prediction model, and a second preset neural network model is pre-trained and subjected to simulated environment reinforcement training based on the display grayscale change prediction model to obtain a first driving waveform prediction model. Based on the target e-ink screen, the first driving waveform prediction model is subjected to real environment reinforcement training to obtain a second driving waveform prediction model. Based on the second driving waveform prediction model, the target ambient temperature set, and the target display grayscale change set, the target driving waveform set of the target e-ink screen is determined. By training a second preset neural network model to output the driving waveform of the e-ink screen, the efficiency of driving waveform debugging for the e-ink screen is improved. Simultaneously, due to the significant time-varying, hysteretic, and nonlinear nature of ink particle motion, the same display grayscale change may be achieved by driving waveforms with multiple distinct intermediate evolution paths. Existing models trained using the complete driving waveform and the final grayscale image displayed on the e-ink screen after receiving the driving waveform lead to reduced prediction accuracy. This application trains a display grayscale change prediction model by using each frame of the driving waveform corresponding to various display grayscale changes and the corresponding grayscale image displayed on the e-ink screen for each frame. This further trains the second preset neural network model, enabling it to predict each frame of the driving waveform, thus improving the accuracy of e-ink screen driving waveform debugging. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of this application, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of this application and therefore should not be considered as a limitation on the scope of protection of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1A flowchart illustrating the first embodiment of the e-ink screen driving waveform debugging method provided in this application; Figure 2 A schematic diagram of the driving waveform provided in this application; Figure 3 A flowchart illustrating the second embodiment of the e-ink screen driving waveform debugging method provided in this application; Figure 4 A schematic diagram of the data acquisition platform provided in this application; Figure 5 A flowchart illustrating the third embodiment of the e-ink screen driving waveform debugging method provided in this application; Figure 6 A flowchart illustrating the fourth embodiment of the e-ink screen driving waveform debugging method provided in this application; Figure 7 A flowchart illustrating the fifth embodiment of the e-ink screen driving waveform debugging method provided in this application; Figure 8 This is a schematic diagram of the structure of the e-ink screen driving waveform debugging device provided in this application.

[0017] Explanation of icon numbers: Drive unit 11, high-speed camera device 12, controller 13, e-ink screen 14. Detailed Implementation

[0018] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0019] The components of the embodiments of this application described and illustrated in the accompanying drawings can be arranged and designed in a variety of different configurations. Therefore, the following detailed description of the embodiments of this application provided in the drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0020] In the following, the terms “comprising,” “having,” and their cognates, which may be used in various embodiments of this application, are intended only to indicate a particular feature, number, step, operation, element, component, or combination thereof, and should not be construed as excluding, firstly, the presence of one or more other features, numbers, steps, operations, elements, components, or combinations thereof, or adding the possibility of one or more features, numbers, steps, operations, elements, components, or combinations thereof.

[0021] Furthermore, the terms "first," "second," and "third" are used only to distinguish descriptions and should not be interpreted as indicating or implying relative importance.

[0022] Unless otherwise specified, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which the various embodiments of this application pertain. Terms (such as those defined in commonly used dictionaries) shall be interpreted as having the same meaning as in their contextual meaning in the relevant technical field and shall not be construed as having an idealized or overly formal meaning, unless clearly defined in the various embodiments of this application.

[0023] It is understood that the method of this application is applied to an e-ink screen driving waveform debugging device, which can be a smart terminal, PC device, etc. For ease of description, the following embodiments use an e-ink screen driving waveform debugging device as the execution subject for illustration.

[0024] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0025] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a first embodiment of the e-ink screen driving waveform debugging method provided in this application. The method includes: Step S101: Obtain training data, which includes: ambient temperature, each frame of the driving waveform corresponding to various display grayscale changes, and the e-ink screen display grayscale image corresponding to each frame of the waveform.

[0026] In this embodiment, before performing model training, the e-ink screen driving waveform debugging device needs to acquire training data, which includes ambient temperature, each frame of the driving waveform corresponding to various display grayscale changes, and the e-ink screen display grayscale image corresponding to each frame of the waveform.

[0027] It should be noted that, due to the characteristics of e-ink screens, the number of frames of the driving waveform corresponding to the grayscale change of the e-ink screen is different under different ambient temperatures. Therefore, ambient temperature is an important parameter that needs to be included in the training data.

[0028] It should be noted that grayscale change refers to the change of the grayscale value of the e-ink screen from one grayscale value to another. For example, when the e-ink screen switches from displaying white to displaying black, it is a change from the display grayscale value of white to the display grayscale value of black.

[0029] It should be noted that due to the physical characteristics of e-ink screens, they cannot instantly display grayscale changes. Instead, they follow a relaxation process similar to diffusion / electrophoresis, with typical response times ranging from tens to hundreds of milliseconds. Therefore, the driving waveform that drives the e-ink screen to display grayscale changes usually consists of multiple waveform segments. Each waveform segment of the driving waveform acts on the e-ink screen, causing a change in the displayed grayscale; that is, each waveform segment corresponds to a grayscale image displayed on the e-ink screen.

[0030] like Figure 2 As shown, Figure 2 This is a schematic diagram of the driving waveform provided in this application, where X represents time, Y represents the voltage number of the driving waveform, and the interval between every two loops in the curve of the driving waveform is a frame segment of the driving waveform.

[0031] Step S102: Based on the training data, train the first preset neural network model to obtain a display grayscale change prediction model, and design a second preset neural network model. Based on the display grayscale change prediction model, perform pre-training and simulated environment reinforcement training on the second preset neural network model to obtain a first driving waveform prediction model.

[0032] In this embodiment, after the e-ink screen driving waveform debugging device acquires the training data, it trains a first preset neural network model based on each frame of the driving waveform corresponding to various display grayscale changes in the training data, thereby obtaining a display grayscale change prediction model. The display grayscale change prediction model can output the corresponding display grayscale change based on each input frame of the waveform.

[0033] It is understandable that the first preset neural network model is trained based on each frame of the driving waveform corresponding to various display grayscale changes until the error between the display grayscale change input to the first preset neural network model and the actual display grayscale change is less than a threshold. At this point, the training is considered complete and a display grayscale change prediction model is obtained.

[0034] In this embodiment, the e-ink screen driving waveform debugging device constructs corresponding simulation training datasets and simulation verification datasets based on the display grayscale change prediction model. Based on these datasets, it performs pre-training and simulated environment reinforcement training on the second preset neural network model to obtain the first driving waveform prediction model. It can be understood that pre-training is to adapt the display grayscale change prediction model to the physical characteristics determined by the e-ink screen, while simulated environment reinforcement training is to train the first preset neural network model to output a driving waveform based on the input ambient temperature and display grayscale change requirements.

[0035] It should be noted that the first preset neural network model and the second preset neural network model can be the same or different. The first preset neural network model and the second preset neural network model can be common large models of architectures such as transformer architecture or MLP architecture, and there is no limitation here.

[0036] Step S103: Based on the target e-ink screen, perform real-environment reinforcement training on the first driving waveform prediction model to obtain the second driving waveform prediction model.

[0037] In this embodiment, although the first driving waveform prediction model can output a driving waveform, it has not yet encountered the noise and non-ideal characteristics of a real e-ink screen. Furthermore, since different batches of e-ink screens typically exhibit differences, the driving waveform output by the first driving waveform prediction model cannot accurately adapt to a real e-ink screen. The e-ink screen driving waveform debugging device performs real-world environment reinforcement training on the first driving waveform prediction model for the target e-ink screen requiring debugging. This allows the first driving waveform prediction model to learn the noise and non-ideal characteristics of the target e-ink screen, resulting in a second driving waveform prediction model.

[0038] Step S104: Based on the second driving waveform prediction model, the target ambient temperature set, and the target display grayscale change set, determine the target driving waveform set of the target e-ink screen.

[0039] In this embodiment, after obtaining the second driving waveform prediction model, the e-ink screen driving waveform debugging device can use the second driving waveform prediction model to debug the driving waveform of the same batch of target e-ink screens. Specifically, the e-ink screen driving waveform debugging device constructs multiple input commands that bind ambient temperature and display grayscale changes based on the target ambient temperature set and the target display grayscale change set. These commands are then input into the second driving waveform prediction model. The second driving waveform prediction model outputs the corresponding target driving waveform based on the input commands, thereby obtaining the target driving waveform set of the target e-ink screen.

[0040] In one embodiment, the method further includes: Step S1041: Based on the target ambient temperature set, the target display grayscale change set, and the target driving waveform set, construct a driving waveform mapping table.

[0041] In this embodiment, after obtaining the target driving waveform set of the target e-ink screen, the e-ink screen driving waveform debugging device constructs a driving waveform mapping table based on the target ambient temperature set, the target display grayscale change set, and the target driving waveform set. It can be understood that each target driving waveform output by the second driving waveform prediction model corresponds to a target ambient temperature and a target display grayscale change; that is, the driving waveform mapping table stores the correlation between a target ambient temperature, a target display grayscale change, and the corresponding target driving waveform.

[0042] Step S1042: Store the driving waveform mapping table in the target e-ink screen so that the target e-ink screen queries and triggers the corresponding driving waveform based on the driving waveform mapping table to achieve grayscale change.

[0043] In this embodiment, after obtaining the driving waveform mapping table, the e-ink screen driving waveform debugging device stores the driving waveform mapping table in the target e-ink screen, so that when the target e-ink screen is put into use, it can query and trigger the corresponding driving waveform based on the driving waveform mapping table to achieve grayscale changes.

[0044] The e-ink screen driving waveform debugging device of this embodiment acquires training data, which includes: ambient temperature, each frame of the driving waveform corresponding to various display grayscale changes, and the e-ink screen display grayscale image corresponding to each frame of the waveform. Based on the training data, a first preset neural network model is trained to obtain a display grayscale change prediction model. Based on the display grayscale change prediction model, a second preset neural network model is pre-trained and subjected to simulated environment reinforcement training to obtain a first driving waveform prediction model. Based on the target e-ink screen, the first driving waveform prediction model is subjected to real environment reinforcement training to obtain a second driving waveform prediction model. Based on the second driving waveform prediction model, the target ambient temperature set, and the target display grayscale change set, the target driving waveform set of the target e-ink screen is determined. By training a second preset neural network model to output the driving waveform of the e-ink screen, the efficiency of driving waveform debugging for the e-ink screen is improved. Meanwhile, because the motion of ink particles has significant time-varying, hysteretic, and nonlinear characteristics, the same display grayscale change may be achieved by driving waveforms with multiple distinct intermediate evolution paths. Existing models trained using the complete driving waveform and the final grayscale image displayed on the e-ink screen after receiving the driving waveform lead to reduced model prediction accuracy. This application trains the second preset neural network model by using each frame of the driving waveform corresponding to various display grayscale changes and the corresponding grayscale image displayed on the e-ink screen for each frame. This enables the second preset neural network model to predict each frame of the driving waveform, improving the accuracy of e-ink screen driving waveform debugging.

[0045] Please refer to Figure 3 , Figure 3 This is a flowchart illustrating a second embodiment of the e-ink screen driving waveform debugging method provided in this application. The difference between the second embodiment and the first embodiment is that the step of acquiring training data includes: Step S201: Send each frame of the driving waveform corresponding to the grayscale change to the e-ink screen in sequence to drive the e-ink screen to achieve the grayscale change.

[0046] Step S202: After sending the preset duration of each waveform segment, acquire the corresponding grayscale image displayed on the e-ink screen.

[0047] Step S203: Obtain the ambient temperature, and bind the ambient temperature, each frame of the driving waveform corresponding to the grayscale change, and the grayscale image of the e-ink screen corresponding to each frame of the waveform to obtain training data.

[0048] In this embodiment, the e-ink screen driver waveform debugging device is equipped with a data acquisition platform, such as... Figure 4 So it seems, Figure 3 This is a schematic diagram of the data acquisition platform provided in this application. The platform includes a controller 13, a drive device 11, a high-speed camera device 12, and an e-ink screen 14. The e-ink screen 14 is an e-ink screen whose drive waveform has been pre-adjusted manually. The e-ink screen drive waveform debugging device, through the controller 13, controls the drive device 11 to sequentially send each frame of the drive waveform corresponding to the grayscale change to the e-ink screen 14, thereby driving the e-ink screen 14 to display grayscale changes. After sending each frame of the waveform for a preset duration, the controller 13 sends an acquisition signal to the high-speed camera device 12. The phase of the acquisition signal remains fixed with the end signal of each frame of the waveform driving the e-ink screen 14 (accuracy at the nanosecond level), acquiring the corresponding grayscale image displayed on the e-ink screen 14. The e-ink screen drive waveform debugging device acquires the ambient temperature and binds the ambient temperature, each frame of the drive waveform corresponding to the grayscale change, and the corresponding grayscale image displayed on the e-ink screen to obtain training data.

[0049] It's important to note that the e-ink screen driver waveform debugging device doesn't wait until the entire driver waveform has completed the grayscale transformation before acquiring a grayscale image of the e-ink screen. Instead, it triggers high-speed imaging immediately after each frame of the driver waveform causes a grayscale change in the e-ink screen, acquiring the corresponding grayscale image of that frame. By leveraging millisecond-level precision acquisition signals (strictly synchronized with the end of the waveform frame) in conjunction with a high-speed camera, it achieves phase-locked frame-by-frame dynamic response capture, fundamentally solving the data acquisition bottleneck caused by the nonlinearity and time-varying nature of e-ink screen responses, and improving the accuracy of training data acquisition.

[0050] Please refer to Figure 5 , Figure 5 This is a flowchart illustrating a third embodiment of the e-ink screen driving waveform debugging method provided in this application. The difference between the third embodiment and the first to second embodiments is that the step of training a first preset neural network model based on the training data to obtain a display grayscale change prediction model, and then pre-training and performing simulated environment reinforcement training on the second preset neural network model based on the display grayscale change prediction model to obtain a first driving waveform prediction model, includes: Step S301: Based on the training data, train the first preset neural network model to establish neural representations between each frame of the driving waveform corresponding to the grayscale change and the corresponding grayscale image of the e-ink screen, thereby obtaining a grayscale change prediction model.

[0051] In this embodiment, the e-ink screen driving waveform debugging device constructs a pre-training dataset based on training data, and trains a first preset neural network model based on the pre-training dataset. This enables the first preset neural network model to establish neural representations between each frame of the driving waveform corresponding to the display grayscale change and the corresponding e-ink screen display grayscale image, thus obtaining a display grayscale change prediction model. Specifically, the first preset neural network model is trained based on each frame of the driving waveform corresponding to various display grayscale changes until the error between the input display grayscale change and the actual display grayscale change is less than a threshold. At this point, training is considered complete, and the display grayscale change prediction model is obtained.

[0052] Step S302: Based on the display grayscale change prediction model, construct a simulation training dataset and a simulation verification dataset, pre-train the second preset neural network model based on the simulation training dataset, and use the display grayscale change prediction model as a simulation environment to strengthen the training of the second preset neural network model until the output error of the second preset neural network model is verified to be less than the first preset threshold based on the simulation verification dataset, thereby obtaining the first driving waveform prediction model.

[0053] In this embodiment, the e-ink screen driving waveform debugging device constructs a simulated training dataset and a simulated verification dataset based on a display grayscale change prediction model, and pre-trains a second preset neural network model based on the simulated training dataset. Specifically, the e-ink screen driving waveform debugging device verifies the pre-trained second preset neural network model based on the simulated verification dataset. If the error between the driving waveform output by the pre-trained second preset neural network model and the driving waveform in the simulated verification dataset is less than a preset error threshold, then pre-training is considered complete, and simulated environment reinforcement training can proceed. If the error between the driving waveform output by the pre-trained second preset neural network model and the driving waveform in the simulated verification dataset is greater than a preset error threshold, then training is considered incomplete, and further pre-training is required.

[0054] In this embodiment, the e-ink screen driving waveform debugging device uses a grayscale change prediction model as a simulation environment. Based on a simulated training dataset, it performs simulated environment reinforcement training on a pre-trained second preset neural network model. Based on a simulated verification dataset, it verifies the second preset neural network model after simulated environment reinforcement training. If the error between the driving waveform output by the second preset neural network model and the driving waveform in the simulated verification dataset is less than a first preset threshold, then the simulated environment reinforcement training is considered complete, and a first driving waveform prediction model is obtained. If the error between the driving waveform output by the second preset neural network model and the driving waveform in the simulated verification dataset is greater than the first preset threshold, then the simulated environment reinforcement training is considered incomplete and needs to continue. The first preset threshold is less than a preset error threshold; that is, pre-training trains a relatively coarse driving second preset neural network model, while simulated environment reinforcement training further trains a more accurate driving second preset neural network model.

[0055] The e-ink screen driving waveform debugging device of this embodiment trains a second preset neural network model by using each frame of the driving waveform corresponding to various display grayscale changes and the corresponding e-ink screen display grayscale image. This enables the second preset neural network model to predict each frame of the driving waveform, which helps to improve the accuracy of e-ink screen driving waveform debugging.

[0056] Please refer to Figure 6 , Figure 6 This is a flowchart illustrating the fourth embodiment of the e-ink screen driving waveform debugging method provided in this application. The difference between the fourth embodiment and the first to third embodiments lies in the step of performing real-environment reinforcement training on the first driving waveform prediction model based on the target e-ink screen to obtain the second driving waveform prediction model, including: Step S401: In response to the target display grayscale change instruction for the target e-ink screen, the first driving waveform prediction model outputs a reference driving waveform and a reference grayscale sequence corresponding to the reference driving waveform.

[0057] In this embodiment, the e-ink screen driving waveform debugging device generates a response to a target display grayscale change command for the target e-ink screen, and the ambient temperature before acquisition. The ambient temperature and the target display grayscale change command are input into a first driving waveform prediction model to obtain an output reference driving waveform and a corresponding reference grayscale sequence. It should be noted that the reference grayscale sequence is the sequence of display grayscale values ​​that each frame segment of the reference driving waveform predicted by the first driving waveform prediction model will cause when applied to the target e-ink screen.

[0058] Step S402: Send the reference driving waveform to the target e-ink screen to drive the target e-ink screen to achieve grayscale change, and collect the grayscale image set of the target e-ink screen.

[0059] In this embodiment, the e-ink screen driving waveform debugging device sends a reference driving waveform to the target e-ink screen to drive the target e-ink screen to achieve grayscale changes, and acquires a set of grayscale images of the target e-ink screen. Specifically, the target e-ink screen is placed in a position such as... Figure 4 In the data acquisition platform shown, the e-ink screen driving waveform debugging device controls the driving device 11 via controller 13 to sequentially send each frame of the reference driving waveform to the target e-ink screen to drive the e-ink screen to achieve grayscale changes. After sending each frame of waveform for a preset duration, controller 13 sends an acquisition signal to high-speed camera device 12. The phase of the acquisition signal remains fixed with the end signal of each frame of waveform (accuracy at the nanosecond level), acquiring the grayscale image of the target e-ink screen. This process continues until all grayscale images of the target e-ink screens are acquired, resulting in a set of grayscale images.

[0060] Step S403: Determine the target grayscale sequence based on the set of displayed grayscale images, and adjust the model parameters of the first driving waveform prediction model based on the target grayscale sequence and the reference grayscale sequence to obtain the second driving waveform prediction model.

[0061] In this embodiment, the e-ink screen driving waveform debugging device determines the target grayscale sequence based on the set of displayed grayscale images, and adjusts the model parameters of the first driving waveform prediction model based on the target grayscale sequence and the reference grayscale sequence to obtain the second driving waveform prediction model.

[0062] In one embodiment, the step of adjusting the model parameters of the first driving waveform prediction model based on the target grayscale sequence and the reference grayscale sequence to obtain the second driving waveform prediction model includes: Step S4031: Based on the target grayscale sequence and the reference grayscale sequence, calculate the grayscale error sequence, and compare each grayscale error of the grayscale error sequence with a second preset threshold.

[0063] In this embodiment, the e-ink screen driving waveform debugging device calculates the grayscale error between each target grayscale value in the target grayscale sequence and the corresponding reference grayscale value in the reference grayscale sequence, obtaining a grayscale error sequence. Each grayscale error in the grayscale error sequence is then compared with a second preset threshold. It can be understood that the grayscale error is the difference between the displayed grayscale value of a frame of the reference driving waveform predicted by the first driving waveform prediction model applied to the target e-ink screen, and the displayed grayscale value of a frame of the reference driving waveform applied to the target e-ink screen.

[0064] Step S4032: If there is a grayscale error greater than the second preset threshold in the grayscale error sequence, a reward vector is generated based on the grayscale error sequence, and the model parameters of the first driving waveform prediction model are adjusted based on the reward vector.

[0065] In this embodiment, if the e-ink screen driving waveform debugging device determines that there is a grayscale error greater than the second preset threshold in the grayscale error sequence, it generates a reward vector based on the grayscale error sequence, adjusts the model parameters of the first driving waveform prediction model based on the reward vector, and continues the above-mentioned real-world reinforcement training based on the first driving waveform prediction model with the adjusted model parameters.

[0066] Step S4033: Continue until it is determined that each grayscale error in the grayscale error sequence is less than the second preset threshold, and obtain the second driving waveform prediction model.

[0067] In this embodiment, the e-ink screen driving waveform debugging device continuously reinforces the first driving waveform prediction model in the above-mentioned real environment until it is determined that each grayscale error in the grayscale error sequence is less than the second preset threshold, thereby obtaining the second driving waveform prediction model.

[0068] In this embodiment, the e-ink screen driving waveform debugging device performs real-world reinforcement training on the first driving waveform prediction model before debugging the target e-ink screen driving waveform using the first driving waveform prediction model. This allows the first driving waveform prediction model to learn the physical characteristics of the target e-ink screen. The real-world reinforcement training is completed when the error between each frame of the reference driving waveform output by the first driving waveform prediction model acting on the reference grayscale sequence behind the target e-ink screen and the error between each frame of the reference driving waveform actually acting on the target grayscale sequence behind the target e-ink screen is less than a threshold, thus obtaining the second driving waveform prediction model. This further improves the compatibility of driving the second preset neural network model with the target e-ink screen to be debugged, thereby helping to improve the accuracy of e-ink screen driving waveform debugging.

[0069] Please refer to Figure 7 , Figure 7 This is a flowchart illustrating the fifth embodiment of the e-ink screen driving waveform debugging method provided in this application. The difference between the fifth embodiment and the first to fourth embodiments is that the step of determining the target driving waveform set of the target e-ink screen based on the second driving waveform prediction model, the target ambient temperature set, and the target display grayscale change set includes: Step S501: Based on the target ambient temperature set and the target display grayscale change set, construct input data containing the target ambient temperature and the target display grayscale change.

[0070] In this embodiment, after obtaining the second driving waveform prediction model, the e-ink screen driving waveform debugging device constructs input data for the second driving waveform prediction model, which includes the target ambient temperature and the target display grayscale change, based on the target ambient temperature set and the target display grayscale change set. It can be understood that the target ambient temperature set includes all possible operating ambient temperatures of the target e-ink screen, and the target display grayscale change set includes all possible display grayscale changes of the target e-ink screen. A target ambient temperature paired with a target display grayscale change constitutes one set of input data.

[0071] Step S502: Input the input data into the second driving waveform prediction model, and determine each frame waveform segment of the target driving waveform based on the target ambient temperature and the target display grayscale change through the second driving waveform prediction model, and output the target driving waveform.

[0072] In this embodiment, the e-ink screen driving waveform debugging device determines the number of frames of the target driving waveform based on the target ambient temperature in the input data, inputs the number of frames and the target display grayscale change in the input data into the second driving waveform prediction model, and determines each frame waveform segment of the target driving waveform based on the target display grayscale change and the number of frames through the second driving waveform prediction model, and outputs the target driving waveform based on the number of frames and each frame waveform segment.

[0073] It should be noted that the number of frames of the driven waveform varies under different ambient temperatures. For example, the number of frames of the driven waveform is 120 at an ambient temperature of 0 degrees Celsius and 20 at an ambient temperature of 25 degrees Celsius.

[0074] Step S503, until the prediction of the target ambient temperature set and the target display grayscale change set is completed based on the second driving waveform prediction model, the target driving waveform set of the target e-ink screen is obtained.

[0075] In this embodiment, the e-ink screen driving waveform debugging device sequentially inputs all the input data that can be combined into the second driving waveform prediction model until the second driving waveform prediction model completes the prediction of the target driving waveform corresponding to all input data, thereby obtaining the target driving waveform set of the target e-ink screen.

[0076] It should be noted that existing driver second preset neural network models only focus on the relationship between the final display effect and the complete waveform after the driving is completed. This application, however, focuses on the display effect of the screen after each frame of driving. Furthermore, e-ink screens cannot instantly complete grayscale changes; instead, they follow a relaxation process similar to diffusion / electrophoresis, with typical response times ranging from tens to hundreds of milliseconds. Therefore, the driving waveform that drives the e-ink screen to complete grayscale changes usually contains multiple waveform segments. Based on existing driver second preset neural network models, each waveform segment in the predicted driving waveform may not be accurate. This could lead to overshoot (particles overshoot, causing flickering or afterimages), unevenness (edge ​​particles arrive first, center lags, resulting in "smudging"), or driving failure (excessive voltage breaks down microcapsules, causing permanent damage) after a certain waveform segment is sent to the e-ink screen. These issues can prevent the driving waveform from driving the e-ink screen to complete the corresponding grayscale changes and may even damage the e-ink screen. Based on the second preset neural network model of this application, since the training process considers both each waveform segment and the display effect of the screen after each waveform segment is driven, the second preset neural network model learns the physical logic of the display effect of the screen after each waveform segment is driven. As a result, the prediction of each waveform segment in the driving waveform is accurate, and sending it to the e-ink screen will not cause overshoot, unevenness, driving failure, etc., thus improving the accuracy of e-ink screen driving waveform debugging.

[0077] refer to Figure 8 , Figure 8 This is a schematic diagram of the structure of the e-ink screen driving waveform debugging device provided in this application. The e-ink screen driving waveform debugging device includes: The acquisition module 10 is used to acquire training data, which includes: ambient temperature, each frame of the driving waveform corresponding to various display grayscale changes, and the e-ink screen display grayscale image corresponding to each frame of the waveform.

[0078] The first training module 20 is used to train a first preset neural network model based on the training data to obtain a display grayscale change prediction model, and to pre-train and simulate environment reinforcement training a second preset neural network model based on the display grayscale change prediction model to obtain a first driving waveform prediction model.

[0079] The second training module 30 is used to perform real-world reinforcement training on the first driving waveform prediction model based on the target e-ink screen to obtain the second driving waveform prediction model.

[0080] The determination module 40 is used to determine the target driving waveform set of the target e-ink screen based on the second driving waveform prediction model, the target ambient temperature set, and the target display grayscale change set.

[0081] It is understood that the e-ink screen driving waveform debugging device in this embodiment corresponds to the e-ink screen driving waveform debugging method in the above embodiment. The options in the above embodiment are also applicable to this embodiment, so they will not be described again here.

[0082] This application also provides a computer device, exemplary of which includes a processor and a memory, wherein the memory stores a computer program, and the processor executes the above-described e-ink screen drive waveform debugging method by running the computer program.

[0083] The processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, including at least one of a Central Processing Unit (CPU), Graphics Processing Unit (GPU), Network Processor (NP), Digital Signal Processor (DSP), Application-Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this application.

[0084] The memory can be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc. The memory is used to store computer programs, and the processor can execute the computer programs accordingly after receiving execution instructions.

[0085] This application also provides a computer storage medium for storing the computer program used in the aforementioned computer device. The computer storage medium can be a readable storage medium, a non-volatile storage medium, or a volatile storage medium. For example, the computer storage medium may include, but is not limited to, various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0086] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that, in alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0087] In addition, the functional modules or units in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0088] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part 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 smartphone, 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 application.

[0089] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for debugging the driving waveform of an e-ink screen, characterized in that, The method includes: Acquire training data, which includes: ambient temperature, each frame of the driving waveform corresponding to various display grayscale changes, and the e-ink screen display grayscale image corresponding to each frame of the waveform. Based on the training data, a first preset neural network model is trained to obtain a grayscale change prediction model, and a second preset neural network model is pre-trained and subjected to simulated environment reinforcement training based on the grayscale change prediction model to obtain a first driving waveform prediction model. Based on the target e-ink screen, the first driving waveform prediction model is subjected to real-environment reinforcement training to obtain the second driving waveform prediction model. Based on the second driving waveform prediction model, the target ambient temperature set, and the target display grayscale change set, the target driving waveform set of the target e-ink screen is determined.

2. The e-ink screen driving waveform debugging method according to claim 1, characterized in that, The steps for obtaining training data include: Each frame of the driving waveform corresponding to the grayscale change is sent to the e-ink screen in sequence to drive the e-ink screen to achieve the grayscale change; After sending each waveform segment for a preset duration, the corresponding grayscale image is captured and displayed on the e-ink screen. The ambient temperature is acquired, and the ambient temperature, each frame of the driving waveform corresponding to the grayscale change is bound together with the grayscale image of the e-ink screen corresponding to each frame of the waveform to obtain training data.

3. The e-ink screen driving waveform debugging method according to claim 1, characterized in that, The steps of training a first preset neural network model based on the training data to obtain a display grayscale change prediction model, and pre-training and simulating environment reinforcement training a second preset neural network model based on the display grayscale change prediction model to obtain a first driving waveform prediction model, include: Based on the training data, the first preset neural network model is trained to establish the neural representation between each frame of the driving waveform corresponding to the grayscale change and the corresponding grayscale image of the e-ink screen, so as to obtain the grayscale change prediction model. Based on the display grayscale change prediction model, a simulation training dataset and a simulation verification dataset are constructed. The second preset neural network model is pre-trained based on the simulation training dataset. The second preset neural network model is reinforced by using the display grayscale change prediction model as a simulation environment until the output error of the second preset neural network model is verified to be less than a first preset threshold based on the simulation verification dataset, thus obtaining the first driving waveform prediction model.

4. The e-ink screen driving waveform debugging method according to claim 1, characterized in that, The steps for performing real-world environment-enhanced training on the first driving waveform prediction model based on the target e-ink screen to obtain the second driving waveform prediction model include: The first driving waveform prediction model responds to the target display grayscale change instruction for the target e-ink screen and outputs a reference driving waveform and a reference grayscale sequence corresponding to the reference driving waveform. The reference driving waveform is sent to the target e-ink screen to drive the target e-ink screen to achieve grayscale change, and the set of grayscale images of the target e-ink screen is collected; A target grayscale sequence is determined based on the set of displayed grayscale images, and the model parameters of the first driving waveform prediction model are adjusted based on the target grayscale sequence and the reference grayscale sequence to obtain a second driving waveform prediction model.

5. The e-ink screen driving waveform debugging method according to claim 4, characterized in that, The step of adjusting the model parameters of the first driving waveform prediction model based on the target grayscale sequence and the reference grayscale sequence to obtain the second driving waveform prediction model includes: Based on the target grayscale sequence and the reference grayscale sequence, a grayscale error sequence is calculated, and each grayscale error in the grayscale error sequence is compared with a second preset threshold. If there is a grayscale error greater than the second preset threshold in the grayscale error sequence, a reward vector is generated based on the grayscale error sequence, and the model parameters of the first driving waveform prediction model are adjusted based on the reward vector. The second driving waveform prediction model is obtained by determining that each grayscale error in the grayscale error sequence is less than the second preset threshold.

6. The e-ink screen driving waveform debugging method according to claim 1, characterized in that, The step of determining the target driving waveform set of the target e-ink screen based on the second driving waveform prediction model, the target ambient temperature set, and the target display grayscale change set includes: Based on the target ambient temperature set and the target display grayscale change set, input data containing the target ambient temperature and the target display grayscale change is constructed; The input data is fed into the second driving waveform prediction model. The second driving waveform prediction model determines each frame waveform segment of the target driving waveform based on the target ambient temperature and the target display grayscale change, and outputs the target driving waveform. The target driving waveform set of the target e-ink screen is obtained by predicting the target ambient temperature set and the target display grayscale change set based on the second driving waveform prediction model.

7. The e-ink screen driving waveform debugging method according to any one of claims 1-6, characterized in that, The method further includes: Based on the target ambient temperature set, the target display grayscale change set, and the target driving waveform set, a driving waveform mapping table is constructed; The driving waveform mapping table is stored in the target e-ink screen so that the target e-ink screen can query and trigger the corresponding driving waveform based on the driving waveform mapping table to achieve grayscale changes.

8. An e-ink screen driving waveform debugging device, characterized in that, The e-ink screen driving waveform debugging device includes: The acquisition module is used to acquire training data, which includes: ambient temperature, each frame of the driving waveform corresponding to various display grayscale changes, and the e-ink screen display grayscale image corresponding to each frame of the waveform. The first training module is used to train a first preset neural network model based on the training data to obtain a display grayscale change prediction model, and to pre-train and simulate environment reinforcement training a second preset neural network model based on the display grayscale change prediction model to obtain a first driving waveform prediction model. The second training module is used to perform real-world reinforcement training on the first driving waveform prediction model based on the target e-ink screen to obtain the second driving waveform prediction model. The determination module is used to determine the target driving waveform set of the target e-ink screen based on the second driving waveform prediction model, the target ambient temperature set, and the target display grayscale change set.

9. A computer device, characterized in that, The computer device includes a processor and a memory, the memory storing a computer program, and the processor executing the computer program to implement the e-ink screen driving waveform debugging method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when run on a processor, executes the e-ink screen driving waveform debugging method according to any one of claims 1-7.

Images