Display parameter repair optimization method and device based on deep learning, equipment, medium and program product

Abstract

The invention relates to the technical field of machine learning, in particular to a display parameter repair optimization method and device based on deep learning, computer equipment, a storage medium and a computer program product. The method comprises the steps of obtaining panel display data of a to-be-processed target object, wherein the panel display data is obtained based on preset environment conditions and set parameters of the target object; converting the panel display data to obtain display characteristic parameters meeting a preset format standard; the display characteristic parameters are processed based on a pre-trained display optimization model, display optimization parameters output by the display optimization model are obtained, the display optimization model is constructed based on a generative adversarial network model, and the display optimization model is obtained through sample training based on paired original panel display data and optimal panel display data; and driving the target object based on the display optimization parameter to obtain a repaired and optimized target object. By adopting the method, the display parameter optimization efficiency and the repair effect can be improved.

Description

Technical Field

[0001] The present application relates to the field of machine learning technology, and in particular to a display parameter repair and optimization method, apparatus, computer equipment, storage medium, and computer program product based on deep learning. Background Art

[0002] Display modules are components used to display visual information, typically composed of multiple parts. They are used in electronic devices such as televisions, computer monitors, mobile phones, and tablets. With the technological advancement of display modules, they have found widespread application in consumer electronics, industrial displays, automotive displays, advertising, and public information displays.

[0003] Due to process limitations, display modules require pre-shipment processing to adjust parameters for optimal display quality and achieve demura. Mura (Japanese for "spots" or "unevenness") refers to uneven brightness or color on a display. Mura typically manifests as a significant difference in brightness or color between a localized area of the screen and the surrounding area, causing visual discomfort to the user. Mura can occur in various situations, including manufacturing process issues, aging effects, and temperature fluctuations. Inconsistencies in materials or processes during display production can lead to uneven pixel response. Over time, the display characteristics of certain areas may change, exacerbating mura. When operating in different ambient temperatures, displays may exhibit varying response characteristics, resulting in uneven brightness or color. Demura parameters are technical parameters used to address and correct mura. Demura technology primarily aims to reduce or eliminate mura through post-production processing to improve display quality. Demura treatment typically involves detection, compensation, and verification. Specifically, first, through testing and measurement technology, the areas on the display screen with mura are identified; an algorithm is used to adjust the output signal of the display screen based on the detected mura characteristics, usually by dynamically compensating the brightness and color to balance the uneven parts; after the modification, the display effect of the screen is re-evaluated to ensure that the mura phenomenon is effectively improved.

[0004] However, the current display parameter repair and optimization methods have the following technical problems:

[0005] The existing parameter adjustment method has the problem of requiring multiple updates and confirmations of visual effects, and the optimization efficiency and effect need to be improved. Summary of the Invention

[0006] Based on this, it is necessary to provide a display parameter repair and optimization method, device, computer equipment, computer-readable storage medium and computer program product based on deep learning to address the above technical problems.

[0007] In a first aspect, the present application provides a method for display parameter repair and optimization based on deep learning. The method comprises:

[0008] Acquiring panel display data of a target object to be processed, wherein the panel display data is acquired based on preset environmental conditions and set parameters of the target object;

[0009] Converting the panel display data to obtain display characteristic parameters that meet preset format standards;

[0010] Processing the display feature parameters based on a pre-trained display optimization model to obtain display optimization parameters output by the display optimization model, wherein the display optimization model is constructed based on a generative adversarial network model and is trained based on paired original panel display data and optimal panel display data as samples;

[0011] The target object is driven based on the display optimization parameters to obtain the repaired and optimized target object.

[0012] In one embodiment, converting the panel display data to obtain display characteristic parameters that meet a preset format standard includes:

[0013] The panel display data is processed based on a preset pre-processing process to obtain the processed panel display data, wherein the pre-processing process includes noise removal and sub-pixel extraction.

[0014] In one embodiment, the method includes:

[0015] Acquire sample data, the sample data including the paired original panel display data and the matched optimal panel display data;

[0016] Constructing a data set for training a model based on the sample data, and training an initial generative adversarial network model;

[0017] The training of the initial generative adversarial network model is constrained based on a preset loss function to obtain network weight parameters.

[0018] In one embodiment, the preset loss function includes a content loss function and an adversarial loss function, wherein the content loss function is used to constrain the difference between the real image and the generated image, and the adversarial loss function is used to enhance the processing capability of the generation network in the display optimization model to enhance the processing performance of the display optimization model.

[0019] In one embodiment, after constraining the training of the initial generative adversarial network model based on a preset loss function to obtain network weight parameters, the method further includes:

[0020] Acquiring the network weight parameters, and determining the weight relationships of different computing units in the display optimization model based on the network weight parameters;

[0021] A skip connection channel is established between the computing units based on the weight relationship.

[0022] In one embodiment, driving the target object based on the display optimization parameter to obtain the repaired and optimized target object includes:

[0023] updating preset parameters of the repair formula in the display optimization model based on the display optimization parameters to obtain the updated repair formula;

[0024] The target object is subjected to display optimization processing based on the updated repair formula to obtain the repaired and optimized target object.

[0025] In a second aspect, the present application also provides a display parameter repair and optimization device based on deep learning. The device comprises:

[0026] A display data module, configured to obtain panel display data of a target object to be processed, wherein the panel display data is obtained based on preset environmental conditions and set parameters for the target object;

[0027] A characteristic parameter module, used for converting the panel display data to obtain display characteristic parameters that meet preset format standards;

[0028] a model processing module, configured to process the display feature parameters based on a pre-trained display optimization model to obtain display optimization parameters output by the display optimization model, wherein the display optimization model is constructed based on a generative adversarial network model and is trained based on paired original panel display data and optimal panel display data as samples;

[0029] A display optimization module is used to drive the target object based on the display optimization parameters to obtain the target object after repair and optimization.

[0030] In one embodiment, the characteristic parameter module includes:

[0031] The pre-processing module is used to process the panel display data based on a preset pre-processing process to obtain the processed panel display data, wherein the pre-processing process includes noise removal and sub-pixel extraction.

[0032] In one embodiment, the apparatus comprises:

[0033] A sample data module, configured to obtain sample data, wherein the sample data includes paired original panel display data and matched optimal panel display data;

[0034] A data set module, used to construct a data set for training a model based on the sample data, and train an initial generative adversarial network model;

[0035] The model training module is used to constrain the training of the initial generative adversarial network model based on a preset loss function to obtain network weight parameters.

[0036] In one embodiment, the preset loss function includes a content loss function and an adversarial loss function, wherein the content loss function is used to constrain the difference between the real image and the generated image, and the adversarial loss function is used to enhance the processing capability of the generation network in the display optimization model to enhance the processing performance of the display optimization model.

[0037] In one embodiment, after the model training module, the method further includes:

[0038] A weight parameter module, configured to obtain the network weight parameters and determine the weight relationship between different computing units in the display optimization model based on the network weight parameters;

[0039] A skip channel module is used to establish a skip connection channel between the computing units based on the weight relationship.

[0040] In one embodiment, the display optimization module includes:

[0041] A repair formula module is used to update the preset parameters of the repair formula in the display optimization model based on the display optimization parameters to obtain the updated repair formula; and to perform display optimization processing on the target object based on the updated repair formula to obtain the repaired and optimized target object.

[0042] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the display parameter repair and optimization method based on deep learning as described in any embodiment of the first aspect.

[0043] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the display parameter repair and optimization method based on deep learning as described in any embodiment of the first aspect.

[0044] In a fifth aspect, the present application further provides a computer program product, comprising a computer program that, when executed by a processor, implements the steps of the display parameter repair and optimization method based on deep learning as described in any embodiment of the first aspect.

[0045] The above-mentioned method, apparatus, computer device, storage medium, and computer program product for display parameter repair and optimization based on deep learning are derived through the technical features in the claims and can achieve the beneficial effects corresponding to the technical problems in the background technology:

[0046] The present application provides a display parameter repair and optimization method based on deep learning, comprising obtaining panel display data of a target object to be processed, wherein the panel display data is obtained based on preset environmental conditions and set parameters for the target object; converting the panel display data to obtain display feature parameters that meet preset format standards; processing the display feature parameters based on a pre-trained display optimization model to obtain display optimization parameters output by the display optimization model, wherein the display optimization model is constructed based on a generative adversarial network model and is trained based on paired original panel display data and optimal panel display data as samples; and driving the target object based on the display optimization parameters to obtain the repaired and optimized target object. In implementation, a terminal can obtain the original data of the display panel to be processed. When the panel data is obtained based on preset environmental conditions and set parameters, consistency in processing of different objects can be ensured, reducing interference from environmental factors. Subsequently, the terminal can convert the panel display data, and by establishing clear data standards, it helps to support compatibility between subsequent processing and input models. In this way, the terminal can obtain display optimization parameters through display optimization model processing. The model is trained using paired original panel display data and optimal panel display data, so that the model can learn the mapping relationship from original data to optimal data. Ultimately, the resulting display optimization parameters are applied to the target object (such as a display or display module). These optimizations significantly improve the target object's display performance, reduce visual defects such as mura, and deliver a better user experience. This overall process improves the efficiency and effectiveness of display optimization compared to repeated manual adjustments. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments of the present application or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying any creative work.

[0048] Figure 1 This is a diagram of an application environment of a display parameter repair and optimization method based on deep learning in one embodiment;

[0049] Figure 2 1 is a schematic diagram of a first flow chart of a display parameter repair and optimization method based on deep learning in one embodiment;

[0050] Figure 3 Schematic diagram of a second flow chart of a display parameter repair and optimization method based on deep learning in another embodiment;

[0051] Figure 4 3 is a schematic diagram of a third flow chart of a display parameter repair and optimization method based on deep learning in another embodiment;

[0052] Figure 5 4 is a schematic diagram of a fourth flow chart of a display parameter repair and optimization method based on deep learning in another embodiment;

[0053] Figure 6 4 is a fifth flow chart of a method for display parameter repair and optimization based on deep learning in another embodiment;

[0054] Figure 7 This is a structural block diagram of a display parameter repair and optimization device based on deep learning in one embodiment;

[0055] Figure 8 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0057] Due to process limitations, display modules require pre-shipment processing to adjust parameters for optimal display quality and achieve demura. Mura (Japanese for "spots" or "unevenness") refers to uneven brightness or color on a display. Mura typically manifests as a significant difference in brightness or color between a localized area of the screen and the surrounding area, causing visual discomfort to the user. Mura can occur in various situations, including manufacturing process issues, aging effects, and temperature fluctuations. Inconsistencies in materials or processes during display production can lead to uneven pixel response. Over time, the display characteristics of certain areas may change, exacerbating mura. When operating in different ambient temperatures, displays may exhibit varying response characteristics, resulting in uneven brightness or color. Demura parameters are technical parameters used to address and correct mura. Demura technology primarily aims to reduce or eliminate mura through post-production processing to improve display quality. Demura treatment typically involves detection, compensation, and verification. Specifically, first, through testing and measurement technology, the areas on the display screen with mura are identified; an algorithm is used to adjust the output signal of the display screen based on the detected mura characteristics, usually by dynamically compensating the brightness and color to balance the uneven parts; after the modification, the display effect of the screen is re-evaluated to ensure that the mura phenomenon is effectively improved.

[0058] However, the current display parameter repair and optimization methods have the following technical problems:

[0059] The existing parameter adjustment method has the problem of requiring multiple updates and confirmations of visual effects, and the optimization efficiency and effect need to be improved.

[0060] Based on this, the embodiments of the present application provide a display parameter repair and optimization method, apparatus, computer equipment, storage medium and computer program product based on deep learning.

[0061] The embodiment of the present application provides a display parameter repair optimization method based on deep learning, which can be applied to Figure 1The display module demura pre- and post-processing parameter optimization system shown in the figure. The display module mura repair SOC system primarily comprises an input / output RGB signal interface conversion module, an FPGA / CPU core processor module (the mura repair IP and pre- and post-processing parameter optimization network are implemented in the FPGA or CPU), memory, and peripheral circuits. Specifically, the system converts the display module input signal (such as MIPI or V-BY-ONE signals) into an RGB signal. The core processor calculates the compensated RGB signal of the current display signal based on display module mura compensation data (data model parameter table) pre-stored in memory. After passing the pre- and post-processing parameter optimization network through the system, the new RGB signal is restored to its original signal type and output, thereby repairing display module mura defects.

[0062] In one embodiment, Figure 2 As shown in the figure, a display parameter repair optimization method based on deep learning is provided. Figure 1 The following steps are used as an example to illustrate the processor terminal in the example:

[0063] Step 202: Acquire panel display data of a target object to be processed, wherein the panel display data is acquired based on preset environmental conditions and set parameters for the target object.

[0064] Step 204: converting the panel display data to obtain display characteristic parameters that meet preset format standards.

[0065] Step 206: Process the display feature parameters based on a pre-trained display optimization model to obtain display optimization parameters output by the display optimization model, wherein the display optimization model is constructed based on a generative adversarial network model, and the display optimization model is obtained by sample training based on paired original panel display data and optimal panel display data.

[0066] Step 208: driving the target object based on the display optimization parameters to obtain the repaired and optimized target object.

[0067] In the above-mentioned display parameter repair and optimization method based on deep learning, reasonable deduction is performed in combination with the technical features in the embodiment to achieve the beneficial effect of solving the technical problems raised in the background technology:

[0068] The present application provides a display parameter repair and optimization method based on deep learning, comprising obtaining panel display data of a target object to be processed, wherein the panel display data is obtained based on preset environmental conditions and set parameters for the target object; converting the panel display data to obtain display feature parameters that meet preset format standards; processing the display feature parameters based on a pre-trained display optimization model to obtain display optimization parameters output by the display optimization model, wherein the display optimization model is constructed based on a generative adversarial network model and is trained based on paired original panel display data and optimal panel display data as samples; and driving the target object based on the display optimization parameters to obtain the repaired and optimized target object. In implementation, a terminal can obtain the original data of the display panel to be processed. When the panel data is obtained based on preset environmental conditions and set parameters, consistency in processing of different objects can be ensured, reducing interference from environmental factors. Subsequently, the terminal can convert the panel display data, and by establishing clear data standards, it helps to support compatibility between subsequent processing and input models. In this way, the terminal can obtain display optimization parameters through display optimization model processing. The model is trained using paired original panel display data and optimal panel display data, so that the model can learn the mapping relationship from original data to optimal data. Ultimately, the resulting display optimization parameters are applied to the target object (such as a display or display module). These optimizations significantly improve the target object's display performance, reduce visual defects such as mura, and deliver a better user experience. This overall process improves the efficiency and effectiveness of display optimization compared to repeated manual adjustments.

[0069] In one embodiment, Figure 3 As shown, step 204 includes:

[0070] Step 302: Processing the panel display data based on a preset pre-processing process to obtain processed panel display data, wherein the pre-processing process includes noise removal and sub-pixel extraction.

[0071] In this embodiment, processing the panel display data through a pre-processing flow helps to enhance the validity and availability of the panel display data and improve the stability of subsequent processing.

[0072] In one embodiment, Figure 4 As shown, the method includes:

[0073] Step 402: Acquire sample data, where the sample data includes a pair of the original panel display data and the matched optimal panel display data.

[0074] Step 404: Construct a data set for training a model based on the sample data, and train the initial generative adversarial network model.

[0075] Step 406: Constrain the training of the initial generative adversarial network model based on a preset loss function to obtain network weight parameters.

[0076] For example, the generative adversarial network model provided in this application can be composed of a generator and a discriminator. Both the generator and the discriminator models are convolutional neural networks. The generator inputs the original image and outputs the generated image. The discriminator inputs the image samples generated by the generator and the real samples. The discriminator judges whether the generated samples are true or false based on the real samples. The GAN network is a game between the generator and the discriminator. During the network iteration process, the generator outputs more realistic data samples, and the discriminator continuously improves its recognition ability. The loss function implements reward or punishment measures based on the goals of the generator and the discriminator. After training is completed, the generator can generate samples that are closer to the distribution of real samples.

[0077] In this embodiment, the display optimization model is obtained based on generative adversarial network training, which helps to improve the stability and optimization performance of the display optimization model.

[0078] In one embodiment, the preset loss function includes a content loss function and an adversarial loss function, wherein the content loss function is used to constrain the difference between the real image and the generated image, and the adversarial loss function is used to enhance the processing capability of the generation network in the display optimization model to enhance the processing performance of the display optimization model.

[0079] For example, the loss function can be composed of a content loss function and an adversarial loss function. The content loss function uses L1 loss to constrain the difference between the real image and the generated image. The formula is as follows:

[0080]

[0081] Among them, yi represents the pixel value of a certain position in the real image, and y'i represents the pixel value of the corresponding position in the generator's predicted image.

[0082] A single content loss function will cause the generated image to be smooth and blurry. By introducing an adversarial loss function, better texture details can be generated. The loss function formula is as follows:

[0083]

[0084] Where Dra is the difference between the generated image and the real image, and E represents the expectation of the data.

[0085] In this embodiment, setting the content loss function and the adversarial loss function in the loss function helps to enhance the processing performance of the model while showing the difference between the real image and the generated image, which helps to improve the overall processing performance of the network.

[0086] In one embodiment, Figure 5 As shown, after step 406, the following steps are further included:

[0087] Step 502: Acquire the network weight parameters, and determine the weight relationship between different computing units in the display optimization model based on the network weight parameters.

[0088] Step 504: Establish a jump connection channel between the computing units based on the weight relationship.

[0089] Among them, jump connection can refer to the structure set in the neural network model, which can enable direct input and output between different computing units, allowing information to flow across layers, and can be used to alleviate gradient disappearance, accelerate convergence and improve network performance.

[0090] For example, the skip connection can be implemented by direct addition of the skip connection, concatenation of the skip connection, weighted skip connection, etc. For example, the skip connection relationship determined by the network weight parameter in the embodiment of the present application can refer to calculating the feature similarity or weight relationship (such as dot product, correlation coefficient, attention weight, etc.) between the features output by two computing units.

[0091] In this embodiment, the attention mechanism helps improve model performance and reduce the loss of redundant information, ultimately enhancing the model's optimization performance. Furthermore, the skip connections set in the optimization model can reduce the amount of detail data that might be lost during decoder processing, thereby improving the model's sensitivity and prediction accuracy.

[0092] In one embodiment, Figure 6 As shown, step 208 includes:

[0093] Step 602: updating preset parameters of the repair formula in the display optimization model based on the display optimization parameters to obtain the updated repair formula;

[0094] Step 604: performing display optimization processing on the target object based on the updated repair formula to obtain the repaired and optimized target object.

[0095] For example, the repair formula may be: Out(i,j) =G(P(In(i,j)))

[0096] Wherein, I and J are the pixel position coordinates of the display module of the target object; In is the grayscale value of the input RGB image pixel of different grayscales, and Out is the output grayscale value; P(*) is the pre-processing mura repair function; G(*) is the post-processing mura repair function.

[0097] In this embodiment, the specific architecture of the repair formula is used to optimize the display of the target object through the repair formula, which helps to improve the optimization effect.

[0098] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0099] Based on the same inventive concept, the embodiments of the present application also provide a deep learning-based display parameter repair and optimization device for implementing the aforementioned deep learning-based display parameter repair and optimization method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more embodiments of the deep learning-based display parameter repair and optimization device provided below can be found in the above-mentioned limitations of the deep learning-based display parameter repair and optimization method, and will not be repeated here.

[0100] In one embodiment, Figure 7 As shown, a display parameter repair and optimization device based on deep learning is provided, including: a display data module, a feature parameter module, a model processing module and a display optimization module, wherein:

[0101] A display data module, configured to obtain panel display data of a target object to be processed, wherein the panel display data is obtained based on preset environmental conditions and set parameters for the target object;

[0102] A characteristic parameter module, used for converting the panel display data to obtain display characteristic parameters that meet preset format standards;

[0103] a model processing module, configured to process the display feature parameters based on a pre-trained display optimization model to obtain display optimization parameters output by the display optimization model, wherein the display optimization model is constructed based on a generative adversarial network model and is trained based on paired original panel display data and optimal panel display data as samples;

[0104] A display optimization module is used to drive the target object based on the display optimization parameters to obtain the target object after repair and optimization.

[0105] In one embodiment, the characteristic parameter module includes:

[0106] The pre-processing module is used to process the panel display data based on a preset pre-processing process to obtain the processed panel display data, wherein the pre-processing process includes noise removal and sub-pixel extraction.

[0107] In one embodiment, the apparatus comprises:

[0108] A sample data module, configured to obtain sample data, wherein the sample data includes paired original panel display data and matched optimal panel display data;

[0109] A data set module, used to construct a data set for training a model based on the sample data, and train an initial generative adversarial network model;

[0110] The model training module is used to constrain the training of the initial generative adversarial network model based on a preset loss function to obtain network weight parameters.

[0111] In one embodiment, the preset loss function includes a content loss function and an adversarial loss function, wherein the content loss function is used to constrain the difference between the real image and the generated image, and the adversarial loss function is used to enhance the processing capability of the generation network in the display optimization model to enhance the processing performance of the display optimization model.

[0112] In one embodiment, after the model training module, the method further includes:

[0113] A weight parameter module, configured to obtain the network weight parameters and determine the weight relationship between different computing units in the display optimization model based on the network weight parameters;

[0114] A skip channel module is used to establish a skip connection channel between the computing units based on the weight relationship.

[0115] In one embodiment, the display optimization module includes:

[0116] A repair formula module is used to update the preset parameters of the repair formula in the display optimization model based on the display optimization parameters to obtain the updated repair formula; and to perform display optimization processing on the target object based on the updated repair formula to obtain the repaired and optimized target object.

[0117] Each module in the above-mentioned deep learning-based display parameter repair and optimization device can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in hardware form, or can be stored in a computer device memory in software form, so that the processor can call and execute the corresponding operations of each module.

[0118] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 8 As shown. The computer device includes a processor, memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals via wired or wireless means, and the wireless means can be achieved via Wi-Fi, mobile cellular networks, NFC (near-field communication), or other technologies. When executed by the processor, the computer program implements a display parameter repair and optimization method based on deep learning. The display unit of the computer device is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse.

[0119] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0120] In one embodiment, a computer device is further provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0121] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0122] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0123] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0124] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), programmable logic devices (PLDs), data processing logic devices based on quantum computing, and the like.

[0125] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0126] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A display parameter repair and optimization method based on deep learning, characterized in that: The method comprises: Acquiring panel display data of a target object to be processed, wherein the panel display data is acquired based on preset environmental conditions and set parameters of the target object; Converting the panel display data to obtain display characteristic parameters that meet preset format standards; Processing the display feature parameters based on a pre-trained display optimization model to obtain display optimization parameters output by the display optimization model, wherein the display optimization model is constructed based on a generative adversarial network model and is trained based on paired original panel display data and optimal panel display data as samples; The target object is driven based on the display optimization parameters to obtain the repaired and optimized target object.

2. The method according to claim 1, characterized in that The converting of the panel display data to obtain display characteristic parameters that meet the preset format standard includes: The panel display data is processed based on a preset pre-processing process to obtain the processed panel display data, wherein the pre-processing process includes noise removal and sub-pixel extraction.

3. The method according to claim 1, characterized in that The method comprises: Acquire sample data, the sample data including the paired original panel display data and the matched optimal panel display data; Constructing a data set for training a model based on the sample data, and training an initial generative adversarial network model; The training of the initial generative adversarial network model is constrained based on a preset loss function to obtain network weight parameters.

4. The method according to claim 3, characterized in that The preset loss function includes a content loss function and an adversarial loss function. The content loss function is used to constrain the difference between the real image and the generated image, and the adversarial loss function is used to enhance the processing capability of the generation network in the display optimization model to enhance the processing performance of the display optimization model.

5. The method according to claim 3, characterized in that After constraining the training of the initial generative adversarial network model based on a preset loss function to obtain network weight parameters, the method further includes: Acquiring the network weight parameters, and determining the weight relationships of different computing units in the display optimization model based on the network weight parameters; A skip connection channel is established between the computing units based on the weight relationship.

6. The method according to claim 1, characterized in that The driving the target object based on the display optimization parameter to obtain the repaired and optimized target object includes: updating preset parameters of the repair formula in the display optimization model based on the display optimization parameters to obtain the updated repair formula; The target object is subjected to display optimization processing based on the updated repair formula to obtain the repaired and optimized target object.

7. A display parameter repair and optimization device based on deep learning, characterized in that: The device comprises: A display data module, configured to obtain panel display data of a target object to be processed, wherein the panel display data is obtained based on preset environmental conditions and set parameters for the target object; A characteristic parameter module, used for converting the panel display data to obtain display characteristic parameters that meet preset format standards; a model processing module, configured to process the display feature parameters based on a pre-trained display optimization model to obtain display optimization parameters output by the display optimization model, wherein the display optimization model is constructed based on a generative adversarial network model and is trained based on paired original panel display data and optimal panel display data as samples; A display optimization module is used to drive the target object based on the display optimization parameters to obtain the target object after repair and optimization.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images