Screen image display method and device, storage medium and electronic equipment

By preprocessing the original screen image of the image display terminal and identifying the afterimage recognition model, and automatically determining and applying compensation parameters, the problems of low efficiency and accuracy of afterimage compensation in the prior art are solved, and efficient and accurate image compensation is achieved.

CN118968899APending Publication Date: 2024-11-15BOE TECHNOLOGY GROUP CO LTD
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202410915509.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-09
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

In the prior art, the image compensation method cannot automatically perform afterimage compensation, resulting in low compensation efficiency and low accuracy of the target screen image.

Method used

By acquiring the original screen image of the image display terminal for image preprocessing, afterimage recognition is performed based on the preset afterimage recognition model, compensation parameters are determined and compensation is compensated in real time, and automatic recognition and compensation of the original screen image is realized.

Benefits of technology

The efficiency of afterimage compensation and the accuracy of the target screen image are improved, manual intervention is reduced, and display quality is significantly improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118968899A_ABST
    Figure CN118968899A_ABST
Patent Text Reader

Abstract

The invention relates to a screen image display method and device, a storage medium and electronic equipment, and relates to the technical field of image processing, and the method comprises the steps: obtaining an original screen image currently displayed by an image display terminal, and carrying out the image preprocessing of the original screen image, and obtaining a standard screen image; performing afterimage identification on the standard screen image based on a preset afterimage identification model to obtain a first afterimage identification result, and determining a first afterimage compensation parameter of the original screen image according to the first afterimage identification result; performing residual image real-time compensation on the original screen image based on the residual image compensation parameter to obtain a target screen image; and transmitting the target screen image to the image display terminal, so that the image display terminal performs replacement display on the original screen image based on the target screen image. The residual image compensation efficiency is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Background Art

[0002] In the existing image compensation method, the screen image displayed by the image display terminal is compared with the original screen image to determine the image afterimage, and then the afterimage compensation is performed manually.

[0003] However, the above method has the following defects: on the one hand, it is not possible to automatically perform afterimage compensation, which results in low efficiency of afterimage compensation; on the other hand, performing afterimage compensation manually results in low accuracy of the compensated target screen image.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] The purpose of the present disclosure is to provide a method for displaying a screen image, a device for displaying a screen image, a computer-readable storage medium and an electronic device, thereby at least to a certain extent overcoming the problems of low afterimage compensation efficiency and low accuracy of the target screen image caused by the limitations and defects of the relevant technology.

[0006] According to one aspect of the present disclosure, a method for displaying a screen image is provided, comprising:

[0007] Acquire the original screen image currently displayed by the image display terminal, and perform image preprocessing on the original screen image to obtain a standard screen image;

[0008] Performing afterimage recognition on the standard screen image based on a preset afterimage recognition model to obtain a first afterimage recognition result, and determining a first afterimage compensation parameter of the original screen image according to the first afterimage recognition result;

[0009] Performing real-time afterimage compensation on the original screen image based on the afterimage compensation parameters to obtain a target screen image;

[0010] The target screen image is transmitted to the image display terminal, so that the image display terminal replaces the original screen image based on the target screen image.

[0011] In an exemplary embodiment of the present disclosure, performing image preprocessing on the original screen image to obtain a standard screen image includes:

[0012] Processing the original image brightness and the original image contrast of the original screen image to obtain a standard image brightness and a standard image contrast, and processing the original color space of the original screen image to obtain a standard color space;

[0013] Based on the standard image brightness, standard image contrast and standard color space, the original image brightness, original image contrast and original color space are replaced to obtain a standard screen image.

[0014] In an exemplary embodiment of the present disclosure, processing the original image brightness and the original image contrast of the original screen image to obtain the standard image brightness and the standard image contrast includes:

[0015] Obtaining preset brightness adjustment parameters and preset contrast adjustment parameters;

[0016] Adjusting the original image brightness of the original screen image according to the preset brightness adjustment parameter to obtain a standard image brightness;

[0017] The contrast of the original image is adjusted according to the preset contrast adjustment parameter to obtain a standard image contrast.

[0018] In an exemplary embodiment of the present disclosure, the original color space of the original screen image is processed to obtain a standard color space, including:

[0019] Obtaining a preset color value conversion matrix; wherein the preset color value conversion matrix includes a first color value conversion matrix, a second color value conversion matrix, and a third color value conversion matrix;

[0020] Calculating a first standard color value of an X dimension in the standard color space based on the first color value conversion matrix and the original color value in the original color space;

[0021] Calculating a second standard color value of a Y dimension in the standard color space based on the second color value conversion matrix and the original color value in the original color space;

[0022] Calculating a third standard color value of a Z dimension in the standard color space based on a third color value conversion matrix and the original color value in the original color space;

[0023] The standard color space is obtained based on the first standard color value, the second standard color value and the third standard color value.

[0024] In an exemplary embodiment of the present disclosure, performing afterimage recognition on the standard screen image based on a preset afterimage recognition model to obtain a first afterimage recognition result includes:

[0025] Performing convolution processing on the standard screen image based on the convolution layer in the preset afterimage recognition model to obtain local image features of the standard screen image;

[0026] Based on the nonlinear layer in the preset residual image recognition model, the local image features are subjected to nonlinear processing to obtain target image features;

[0027] Based on the pooling layer in the preset residual image recognition model, the target image processing features are down-sampled to obtain the first residual image recognition result.

[0028] In an exemplary embodiment of the present disclosure, the preset afterimage recognition model is obtained in the following manner:

[0029] The preset reference weight parameters are read from the weight storage component, and the parameters of the convolutional neural network are updated based on the preset reference weight parameters to obtain the preset afterimage recognition model.

[0030] In an exemplary embodiment of the present disclosure, the preset reference weight parameter is obtained in the following manner:

[0031] Acquire a historical screen image and an image afterimage label corresponding to the historical screen image, and input the historical screen image into a convolutional neural network to obtain an afterimage prediction result;

[0032] Constructing a loss function according to the residual image prediction result and the image residual image label, and training the convolutional neural network based on the loss function;

[0033] Extract the parameters in the trained convolutional neural network to obtain the preset reference weight parameters, and store the preset reference weight parameters in the weight storage component.

[0034] In an exemplary embodiment of the present disclosure, determining a first afterimage compensation parameter of the original screen image according to the first afterimage recognition result includes:

[0035] A compensation parameter query condition is generated according to the first afterimage recognition result, and based on the compensation parameter query condition, a first afterimage compensation parameter corresponding to the original screen image is matched from a preset color comparison table.

[0036] In an exemplary embodiment of the present disclosure, the first afterimage compensation parameter includes at least one of a brightness compensation coefficient, an R channel compensation coefficient, a G channel compensation coefficient, and a B channel compensation coefficient;

[0037] The method of performing real-time afterimage compensation on the original screen image based on the afterimage compensation parameter to obtain a target screen image includes:

[0038] Compensating the original image brightness of the original screen image based on the brightness compensation coefficient to obtain a target image brightness, and compensating the original R pixel channel value of the original screen image based on the R channel compensation coefficient to obtain a target R pixel channel value;

[0039] Compensating the original G pixel channel value of the original screen image based on the G channel compensation coefficient to obtain a target G pixel channel value, and compensating the original B pixel channel value of the original screen image based on the B channel compensation coefficient to obtain a target B pixel channel value;

[0040] Based on the target image brightness, target R pixel channel value, target G pixel channel value and target B pixel channel value, the original image brightness, original R pixel channel value, original G pixel channel value and original B pixel channel value are replaced to obtain a target screen image.

[0041] In an exemplary embodiment of the present disclosure, the first afterimage compensation parameter further includes an image filter coefficient;

[0042] The method further comprises: performing real-time afterimage compensation on the original screen image based on the afterimage compensation parameter to obtain a target screen image;

[0043] The original screen image is filtered based on the image filtering coefficient to obtain a target screen image.

[0044] In an exemplary embodiment of the present disclosure, the method for displaying a screen image further includes:

[0045] Acquire a target screen image currently displayed by the image display terminal, and perform afterimage recognition on the target screen image based on a preset afterimage recognition model to obtain a second afterimage recognition result;

[0046] Determining whether the target screen image needs to be compensated again according to the second residual image recognition result;

[0047] If image compensation is required again, a second afterimage compensation parameter of the target screen image is determined based on the second afterimage recognition result, and the target screen image is compensated based on the second afterimage compensation parameter.

[0048] According to one aspect of the present disclosure, there is provided a device for displaying a screen image, comprising:

[0049] An image preprocessing module, used to obtain the original screen image currently displayed by the image display terminal, and perform image preprocessing on the original screen image to obtain a standard screen image;

[0050] A first compensation parameter determination module, configured to perform afterimage recognition on the standard screen image based on a preset afterimage recognition model to obtain a first afterimage recognition result, and determine a first afterimage compensation parameter of the original screen image according to the first afterimage recognition result;

[0051] An afterimage real-time compensation module, used for performing afterimage real-time compensation on the original screen image based on the afterimage compensation parameters to obtain a target screen image;

[0052] The screen image display module is used to transmit the target screen image to the image display terminal, so that the image display terminal replaces the original screen image based on the target screen image.

[0053] According to one aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method for displaying a screen image described in any one of the above is implemented.

[0054] According to one aspect of the present disclosure, there is provided an electronic device, including:

[0055] Processor; and

[0056] A memory, configured to store executable instructions of the processor;

[0057] Wherein, the processor is configured to execute any one of the above-mentioned methods for displaying a screen image by executing the executable instructions.

[0058] A method for displaying a screen image provided by an embodiment of the present disclosure comprises the following steps: on the one hand, obtaining an original screen image currently displayed by an image display terminal and performing image preprocessing on the original screen image to obtain a standard screen image; then performing afterimage recognition on the standard screen image based on a preset afterimage recognition model to obtain a first afterimage recognition result, and determining a first afterimage compensation parameter of the original screen image according to the first afterimage recognition result; then performing real-time afterimage compensation on the original screen image based on the afterimage compensation parameter to obtain a target screen image; finally, transmitting the target screen image to the image display terminal so that the image display terminal replaces the original screen image based on the target screen image for display, thereby realizing automatic recognition and automatic compensation of the afterimage of the original screen image, solving the problem in the prior art that afterimage compensation cannot be automatically performed, resulting in low efficiency of afterimage compensation; on the other hand, since afterimage recognition can be performed on the standard screen image based on a preset afterimage recognition model to obtain a first afterimage recognition result, and determining a first afterimage compensation parameter of the original screen image according to the first afterimage recognition result; then performing real-time afterimage compensation on the original screen image based on the afterimage compensation parameter to obtain the target screen image, thereby improving the accuracy of the obtained target screen image.

[0059] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0061] Figure 1 The flowchart schematically shows a method for displaying a screen image according to an exemplary embodiment of the present disclosure.

[0062] Figure 2 A diagram schematically shows an example structure of a system for displaying a screen image according to an example embodiment of the present disclosure.

[0063] Figure 3 A diagram schematically shows an example scene of image display based on a spliced ​​screen according to an example embodiment of the present disclosure.

[0064] Figure 4 The following schematically shows an example functional structure diagram of an FPGA chip according to an example embodiment of the present disclosure.

[0065] Figure 5 A schematic diagram shows a structural example of a preset afterimage recognition model according to an exemplary embodiment of the present disclosure.

[0066] Figure 6 A flowchart schematically illustrates a process for generating a preset reference weight parameter according to an exemplary embodiment of the present disclosure.

[0067] Figure 7 A diagram schematically shows an example structure of a device for displaying a screen image according to an exemplary embodiment of the present disclosure.

[0068] Figure 8 An electronic device for implementing a method for displaying a screen image according to an exemplary embodiment of the present disclosure is schematically illustrated. DETAILED DESCRIPTION

[0069] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as being limited to the examples set forth herein; on the contrary, these embodiments are provided so that the present disclosure will be more comprehensive and complete, and the concepts of the example embodiments are fully conveyed to those skilled in the art. The described features, structures, or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known technical solutions are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0070] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0071] In this exemplary embodiment, a method for displaying a screen image is first provided, which can be run on a terminal device, a server, a server cluster or a cloud server, etc. Of course, those skilled in the art can also run the method disclosed in this disclosure on other platforms as required, which is not specifically limited in this exemplary embodiment. Figure 1 As shown, the method for displaying the screen image may include the following steps:

[0072] Step S110. Acquire the original screen image currently displayed by the image display terminal, and perform image preprocessing on the original screen image to obtain a standard screen image;

[0073] Step S120. Perform afterimage recognition on the standard screen image based on a preset afterimage recognition model to obtain a first afterimage recognition result, and determine a first afterimage compensation parameter of the original screen image according to the first afterimage recognition result;

[0074] Step S130. Performing real-time afterimage compensation on the original screen image based on the afterimage compensation parameters to obtain a target screen image;

[0075] Step S140: delivering the target screen image to the image display terminal, so that the image display terminal replaces the original screen image based on the target screen image.

[0076] In the above-mentioned screen image display method, on the one hand, the original screen image currently displayed by the image display terminal is obtained, and the original screen image is preprocessed to obtain a standard screen image; then, the standard screen image is subjected to afterimage recognition based on a preset afterimage recognition model to obtain a first afterimage recognition result, and a first afterimage compensation parameter of the original screen image is determined according to the first afterimage recognition result; then, the original screen image is subjected to real-time afterimage compensation based on the afterimage compensation parameter to obtain a target screen image; finally, the target screen image is transmitted to the image display terminal, so that the image display terminal replaces the original screen image based on the target screen image for display, thereby realizing automatic recognition and automatic compensation of the afterimage of the original screen image, solving the problem in the prior art that afterimage compensation cannot be automatically performed, resulting in low efficiency of afterimage compensation; on the other hand, since afterimage recognition can be performed on the standard screen image based on a preset afterimage recognition model to obtain a first afterimage recognition result, and a first afterimage compensation parameter of the original screen image is determined according to the first afterimage recognition result; then, the original screen image is subjected to real-time afterimage compensation based on the afterimage compensation parameter to obtain a target screen image, thereby improving the accuracy of the obtained target screen image.

[0077] Hereinafter, the method for displaying a screen image according to an exemplary embodiment of the present disclosure will be explained and illustrated in detail with reference to the accompanying drawings.

[0078] First, the technical implementation principle and application scenario of the exemplary embodiment of the present disclosure are explained and illustrated. Specifically, the screen image display method recorded in the exemplary embodiment of the present disclosure is an automatic parameter adjustment method of MLED_COG_LTPS afterimage compensation based on FPGA (Field Programmable Gate Array), which aims to automatically solve the afterimage problem of MLED (Micro Light Emitting Diode Display) screen when using COG (Chip on Glass) packaging technology and LTPS (Low Temperature Poly-Silicon) backplane technology; in the actual application process, the method realizes high-speed parallel processing through FPGA, and automatically adjusts the afterimage compensation parameters in combination with machine learning algorithms to achieve fast and accurate afterimage compensation. Among them, the COG recorded here is a technology that directly connects the bare chip on the glass circuit board of the liquid crystal display by wire connection and gluing; at the same time, the technology can be applied to the liquid crystal display of electronic products.

[0079] Secondly, the screen image display method recorded in the example embodiments of the present disclosure can at least solve the following technical problems: on the one hand, how to achieve rapid detection and real-time compensation of MLED_COG_LTPS screen afterimages, and how to reduce manual intervention through automatic parameter adjustment to improve the efficiency and accuracy of afterimage compensation; on the other hand, how to design a hardware architecture that can efficiently execute the afterimage compensation algorithm on the FPGA, improve the stability and processing speed of the system, and make the most of the parallel computing capabilities of the FPGA and how to optimize the design of data communication and storage; on the other hand, how to combine machine learning algorithms to achieve intelligent optimization of afterimage compensation parameters and further improve display quality; at the same time, the algorithm automatically adjusts the afterimage compensation parameters by analyzing a large amount of actual data to adapt to changes in different scenarios, thereby achieving more accurate and effective compensation and improving the display quality.

[0080] Furthermore, the screen image display method provided by the exemplary embodiment of the present disclosure realizes fast and accurate compensation of residual images through the combination of hardware acceleration and intelligent algorithms; at the same time, the uniqueness of this method lies in the use of the flexibility and high performance of FPGA to hardwareize the residual image compensation process, thereby significantly improving the compensation efficiency and accuracy; compared with the traditional software implementation, this method greatly shortens the compensation response time, effectively eliminates the residual image phenomenon, and significantly improves the display quality of the MLED_COG_LTPS screen; not only that, the present disclosure also reduces the cost and time of product development and debugging; at the same time, due to the automatic adjustment of the residual image compensation process, the need for manual intervention is greatly reduced, saving a lot of human resources and time costs. In addition, the FPGA-based hardware implementation method ensures the stability and efficiency of the system and further improves the user experience.

[0081] Furthermore, with the continuous advancement of display technology, the screen image display method recorded in the example embodiment of the present disclosure has broad application prospects and market value; at the same time, the method has great application potential in fields such as MLED display screens, and has contributed important technological breakthroughs and innovative ideas as well as user experience to the development of display technology; and, the method has high practicality and application prospects, and has broad application value in fields such as MLED display screens.

[0082] Hereinafter, the screen image display system involved in the exemplary embodiment of the present disclosure will be explained and described. Figure 2 As shown, the screen image display system may include an image display terminal 210, an image acquisition device 220 and a computer 230; wherein the computer is connected to the image acquisition device and the image display terminal via an external interface; the external interface described herein includes an HDMI (High Definition Multimedia Interface) and a USB (Universal Serial Bus) interface, or may be other interfaces, such as LVDS (Low-Voltage Differential Signaling) or other dedicated interfaces, and this example does not impose any special restrictions on this. In actual application, the image display terminal may be used to display screen images; the computer may be used to implement the screen image display method described in the exemplary embodiment of the present disclosure; the image acquisition device may be used to acquire the screen image displayed by the image display terminal.

[0083] In an exemplary embodiment, the image display terminal described above may be a spliced ​​display screen implemented based on MLED; wherein the specific display scene diagram of the spliced ​​display screen may refer to Figure 3As shown; the image acquisition device recorded above can be a camera or a camera group, which can be used to capture high-resolution image data; the computer recorded above is provided with an FPGA chip, and the specific image processing process is implemented based on the FPGA chip; that is, the specific image processing process can be integrated into the FPGA chip, and then the FPGA chip can be installed on the corresponding computer in a pluggable manner to achieve specific image processing.

[0084] In an exemplary embodiment, the hardware design and programming of the above modules are implemented on an FPGA chip, including image processing, parameter adjustment, afterimage compensation and other functions, and the parallel computing capability of FPGA is used to realize efficient execution of the algorithm; further, in the actual application process, reference Figure 4 As shown, the FPGA chip may include an image acquisition component 410, an AI residual image analysis component 420, and an output control component 430; wherein each component may be connected to each other through a high-speed data bus for communication, and each component is responsible for different functions; and, if specific image compensation processing is required, the AI ​​analysis model suitable for the MLED splicing screen may be integrated and trained after the development of the relevant algorithm is completed, and then the FPGA-based hardware system architecture may be designed and implemented, and the above algorithm may be implemented on the FPGA to achieve the purpose of processing the residual image of the screen image. Further, the data acquisition component recorded herein is provided with an interface processing component 411, a buffer synchronization component 412, and a data preprocessing component 413; the AI ​​residual image analysis component is provided with a weight storage component 421, a data calculation component 422, and a parameter recommendation component 423; and the output control component is provided with a real-time compensation component 431, a parameter adjustment component 432, and a feedback iteration component 433. At the same time, the data preprocessing component is used to perform the preprocessing process of the original screen image, the data calculation component is used to perform the afterimage recognition process, the parameter recommendation component is used to perform the afterimage compensation parameter determination process; the real-time compensation component is used to perform the afterimage compensation process; the specific execution functions of other components will be described in detail later and will not be repeated here.

[0085] In an exemplary embodiment, the image acquisition component is the starting link in the entire system. The image acquisition component is responsible for capturing the display image (original screen image) of the MLED_COG_LTPS screen through a camera, and transmitting the image data of the original screen image to the data preprocessing component through an interface; after receiving the image data, the data preprocessing component first performs grayscale processing and filtering operations on the image to reduce noise interference and improve the accuracy of afterimage detection; at the same time, these preprocessing steps ensure that the system can accurately identify and locate the afterimage from the original image data.

[0086] In an exemplary embodiment, the interface processing component 411 described above can be used to process the communication interface with the MLED splicing screen, and the use of these high-speed serial communication interfaces can meet the high-speed data transmission requirements; further, in the actual application process, the interface processing component also needs to be able to manage the communication interface with the image source (that is, to process the communication interface with the camera), and receive the original image data from the image source (data collected by the camera) and then process these data streams at high speed and transmit them to the FPGA. It should be noted here that the implementation of the interface processing component needs to ensure the integrity of the interface signal, including signal matching, filtering out noise interference, checking and error detection of received data to ensure data integrity and accuracy, etc.; afterwards, the clock management unit PLL (PhaseLocked Loop) inside the FPGA is also used to provide a clock signal synchronized with the image source data. Furthermore, in the actual application process, it is also necessary to set the FPGA's clock signal, reset signal and image frame start signal, use the image data of the external device as input, output the data to the buffer synchronization submodule, and define a data output valid signal; at the same time, when the start signal of a frame of image is received, the data is transmitted from the input port to the output port, and it is determined whether the data is valid and marked; the synchronization mechanism based on these clocks ensures the synchronization of the data.

[0087] In an exemplary embodiment, the buffer synchronization component 412 described above is responsible for buffering the image data received from the interface processing component and facilitates subsequent processing by ensuring the synchronization of the data; in actual application, the buffer synchronization component can use the FIFO (first-in-first-out) buffer inside the FPGA to temporarily store the image data, so as to solve the loss or overflow problem caused by data rate mismatch; the specific buffer mechanism can adopt a double buffer or triple buffer mechanism to reduce interruptions during the image acquisition process; and, pointers can also be used to manage the position of data in the FIFO to ensure the order and integrity of the data, and finally use handshake signals (ready and valid) to coordinate the reception and transmission of data to ensure data synchronization; based on the data management and scheduling method described above, timely transmission and processing of data are ensured.

[0088] In an example embodiment, the AI ​​afterimage analysis component is a core component of the system, which uses a machine learning algorithm to identify afterimages of preprocessed image data (i.e., standard screen images); through learning and training of a large amount of image data, the module can accurately identify different types of afterimages and extract key features for analysis; at the same time, once an afterimage is detected, the parameter adjustment component will immediately intervene to automatically calculate and adjust the afterimage compensation parameters according to a preset algorithm; at the same time, the afterimage compensation parameters may include but are not limited to brightness compensation parameters, RGB compensation parameters, 3x3 filter coefficients, etc., aiming to eliminate afterimages and restore image quality.

[0089] In an example embodiment, in the weight storage component, the size and precision of the reference weight parameters must first be determined, and the storage resources of the FPGA must not be excessively wasted while maintaining sufficient precision; for example, the reference weight parameters can be quantized into a 16-bit fixed-point format for storage to balance the requirements of precision and storage space; at the same time, the reference weight parameters can be loaded through the FPGA's Bitstream configuration file, and then the BRAM is used to store the weights to improve data reading speed and efficiency. Furthermore, in order to achieve the storage and reading of reference weight parameters, it is necessary to create a BRAM of appropriate size to store weight data, and design an address interface for other sub-modules to read weights; during the initialization of the module, the weight data trained by the AI ​​algorithm model is loaded into the BRAM for use in subsequent calculations and inference processes.

[0090] Furthermore, in the process of storing the reference weight parameters, some parameters can be defined and the address width of the weight storage can be set to 10 bits; under this premise, the address can store 1024 weights, set the weight data width to 16 bits, and define a weight initialization file, due to the addition of BRAM declaration and initialization, as well as the weight initialization process; at the same time, the weight reading logic needs to be set, due to resetting the weight output and reading the weight according to the address. In the actual application process, the reference weight parameters are obtained by training the neural network. The weight file obtained by training needs to be converted into a 16-bit fixed-point format and stored in BRAM to facilitate the use of the weight file generated by the training process; among them, the initialization file containing the quantized weights can be as follows:

[0091] 000000000000000010000000000000011 / / The weight value is quantized to a 16-bit fixed-point number 1;

[0092] 000000000000000001000000000000000 / / The weight value is quantized to a 16-bit fixed point number 2; ....

[0094] In the initialization file recorded above, each weight value is a binary number, arranged in a specific byte order, and these weights are loaded in the initial block of the module using the $readmemb function; based on this method, the weight value can be represented as a 16-bit fixed-point number for storage and calculation in FPGA; at the same time, in neural networks, floating-point numbers are usually used to represent weight values, but on hardware platforms such as FPGA, due to limited resources and complex floating-point calculations, weight values ​​are usually quantized to fixed-point numbers to reduce resource consumption and improve calculation efficiency; further, the 16-bit fixed-point number recorded here refers to the use of 16 bits to represent fixed-point numbers, in which a part of the bits are fixed to represent the integer part, and the remaining bits represent the decimal part; and based on this method, efficient multiplication and addition operations can be performed on FPGA, while also reducing storage overhead. In the given example, each weight value is quantized into a 16-bit fixed-point number, in which the first 8 bits are used to represent the integer part and the last 8 bits are used to represent the decimal part. Based on this method, certain accuracy requirements can be met while maintaining a small storage space requirement, which is suitable for calculation and storage on FPGA.

[0095] In an example embodiment, the output control component can be responsible for applying the adjusted parameters to the MLED_COG_LTPS screen in real time to perform real-time compensation for afterimages; at the same time, the system continuously monitors the compensation effect through a feedback mechanism, and further adjusts the parameters based on the feedback iteration to achieve the best compensation effect; this closed-loop feedback mechanism ensures that the system can continue to work stably under changing environmental conditions, and ensures continuous automated optimization of the compensation effect.

[0096] The following will explain and illustrate the preset residual image recognition model involved in the exemplary embodiment of the present disclosure. Figure 5As shown, the preset residual image recognition model may include an input layer 501, a convolution layer 502, a nonlinearization layer 503, a pooling layer 504 and an output layer 505; wherein the input layer 501, the convolution layer 502, the nonlinearization layer 503, the pooling layer 504 and the output layer 505 are connected in sequence; wherein the specific functions of the different layers will be further described later and will not be repeated here. Furthermore, the preset residual image recognition model recorded here is obtained in the following manner: the preset reference weight parameters are read from the weight storage component, and the parameters of the convolutional neural network are updated based on the preset reference weight parameters to obtain the preset residual image recognition model. That is, the trained reference weight parameters can be stored in the weight storage component. When the reference weight parameters need to be used, the reference weight parameters can be directly obtained and the convolutional neural network can be initialized based on the reference weight parameters to obtain the preset residual image recognition model. It should be noted here that the weight storage component described above can store reference weight parameters for multiple different scenarios, which can be selected according to actual conditions during actual application; for example, reference weight parameters corresponding to multiple spliced ​​screens, reference weight parameters corresponding to a single display screen, etc. This example does not impose any special restrictions on this.

[0097] The following will explain and illustrate the specific generation process of the preset reference weight parameters. Figure 6 As shown, the specific generation process of the preset reference weight parameters may include the following steps:

[0098] Step S610, obtaining a historical screen image and an image afterimage label corresponding to the historical screen image, and inputting the historical screen image into a convolutional neural network to obtain an afterimage prediction result;

[0099] Step S620, constructing a loss function according to the residual image prediction result and the image residual image label, and training the convolutional neural network based on the loss function;

[0100] Step S630, extracting the parameters in the trained convolutional neural network, obtaining the preset reference weight parameters, and storing the preset reference weight parameters in the weight storage component.

[0101] In the following, step S610 to step S630 will be explained and illustrated. Specifically, the historical screen image recorded above refers to the screen image that has been displayed by the image display terminal, and the image residual image label refers to the residual image comparison result obtained by comparing the historical screen image and the original image corresponding to the historical screen image; the loss function recorded here can be a mean square error loss function or other loss functions, and this example does not impose special restrictions on this; at the same time, in the process of determining specific parameters, image display terminals with different numbers of spliced ​​screens can correspond to different reference weight parameters; therefore, in the process of model training, it can be achieved by sequential training; in this way, the accuracy of the reference weight parameters obtained can be further improved, thereby achieving the purpose of improving the accuracy of the residual image recognition result.

[0102] The following will be combined Figure 2-Figure 6 right Figure 1 The display method of the screen image shown in is further explained and illustrated. Specifically:

[0103] In step S110, the original screen image currently displayed by the image display terminal is acquired, and image preprocessing is performed on the original screen image to obtain a standard screen image.

[0104] In this example embodiment, first, the original screen image is acquired; specifically, the original screen image can be acquired from the interface processing component described above; that is, the image acquisition device transmits the acquired original screen image to the interface processing component via the USB interface, and the interface processing component then buffers the original screen image to the buffer synchronization component; finally, when the data preprocessing component needs to preprocess the original screen image, the original screen image can be read from the buffer synchronization component and the image preprocessing process can be executed.

[0105] Secondly, the original screen image is preprocessed; specifically, it can be achieved in the following ways: first, the original image brightness and original image contrast of the original screen image are processed to obtain standard image brightness and standard image contrast, and the original color space of the original screen image is processed to obtain standard color space; secondly, based on the standard image brightness, standard image contrast and standard color space, the original image brightness, original image contrast and original color space are replaced to obtain a standard screen image. That is, in the process of preprocessing the original screen image, it can be achieved from multiple different dimensions such as image brightness, image contrast and image color space; of course, in order to further remove image noise in the original screen image, filtering processing and the like can also be performed, and this example does not impose any special restrictions on this.

[0106] In an exemplary embodiment, the original image brightness and original image contrast of the original screen image are processed to obtain the standard image brightness and standard image contrast, which can be achieved in the following manner: first, a preset brightness adjustment parameter and a preset contrast adjustment parameter are obtained; second, the original image brightness of the original screen image is adjusted according to the preset brightness adjustment parameter to obtain the standard image brightness; then, the original image contrast is adjusted according to the preset contrast adjustment parameter to obtain the standard image contrast. Specifically, in the actual application process, it is first necessary to set the brightness adjustment parameter and the contrast adjustment parameter; second, the updated standard image brightness is calculated according to a simple addition formula; specifically, the specific calculation process of the standard image brightness can be as follows:

[0107] processed_data=raw_data+BRIGHTNESS;

[0108] Among them, processed_data is the standard image brightness, raw_data is the original image brightness; BRIGHTNESS is the brightness adjustment parameter;

[0109] Secondly, the updated contrast data (that is, the standard image contrast pair) is calculated according to the contrast adjustment formula; wherein the specific calculation process of the standard image contrast can be as follows:

[0110] Standard image contrast = (original image contrast - 128) * contrast adjustment parameter / 128 + 128;

[0111] Furthermore, in the actual application process, a module named image_processing_pipeline is defined, and then based on the module, raw_data (original image brightness) of 8-bit original image data is received, and the image data is enhanced by brightness and contrast adjustment. At the same time, brightness adjustment is achieved by simply adding the original data to a fixed value; contrast adjustment is achieved through a simple arithmetic formula, which multiplies the original data by the contrast parameter, and then divides it by a fixed value of 128, and shifts the result to the right by 7 bits to achieve the division by 128 operation, and the result is limited to 8 bits.

[0112] In an exemplary embodiment, the original color space of the original screen image is processed to obtain a standard color space, which can be achieved in the following manner: obtaining a preset color value conversion matrix; wherein the preset color value conversion matrix includes a first color value conversion matrix, a second color value conversion matrix, and a third color value conversion matrix; based on the first color value conversion matrix and the original color values ​​in the original color space, calculating the first standard color value of the X dimension in the standard color space; based on the second color value conversion matrix and the original color values ​​in the original color space, calculating the second standard color value of the Y dimension in the standard color space; based on the third color value conversion matrix and the original color values ​​in the original color space, calculating the third standard color value of the Z dimension in the standard color space; based on the first standard color value, the second standard color value, and the third standard color value, obtaining the standard color space. Specifically, the original color space recorded here is the RGB color space, and the standard color space is the XYZ color space; wherein, the color space conversion from RGB to XYZ is a complex process involving color coordinate transformation; the RGB color space is based on human perception of the three colors of red, green, and blue, while the XYZ color space is a linear color space based on human visual response, and is often used for precise color management and conversion; therefore, in the process of converting the RGB color space to the XYZ color space, each RGB color value is converted into a corresponding XYZ color value through a fixed 3x3 color value conversion matrix; wherein, the coefficients in the first color value conversion matrix, the second color value conversion matrix, and the third color value conversion matrix recorded here are obtained based on experimental data of human average color response.

[0113] Furthermore, the specific conversion process of the first standard color value X', the second standard color value Y' and the third standard color value Z' can be as follows:

[0114] X'=R*0.412453*255+G*0.357580*255+B*0.180423*255

[0115] Y'=R*0.212671*255+G*0.715160*255+B*0.072169*255

[0116] Z'=R*0.019334*255+G*0.119194*255+B*0.950227*255

[0117] Among them, R, G, B represent the input RGB values ​​(usually normalized to the range of 0-1, or converted to the range of 0-255), and X', Y', and Z' represent the output XYZ values. In the actual application process, this conversion can be implemented in FPGA; during the conversion process, each RGB value needs to be weighted and summed, and the result is formatted into the required number of bits, by receiving 8-bit RGB input and outputting 16-bit XYZ color values; at the same time, fixed coefficients can be used to represent the weights in the conversion matrix, and these coefficients are multiplied by the RGB values, and then the results are added to obtain the XYZ components; based on this, the grayscale processing of the image can be realized, the color image is converted to a grayscale image, and edge detection and smoothing are performed according to the image features to extract useful information and prepare for subsequent residual image analysis and compensation.

[0118] In step S120, afterimage recognition is performed on the standard screen image based on a preset afterimage recognition model to obtain a first afterimage recognition result, and a first afterimage compensation parameter of the original screen image is determined according to the first afterimage recognition result.

[0119] In this example embodiment, first, afterimage recognition is performed on the standard screen image; specifically, it can be implemented in the following manner: based on the convolution layer in the preset afterimage recognition model, the standard screen image is convoluted to obtain the local image features of the standard screen image; based on the nonlinear layer in the preset afterimage recognition model, the local image features are nonlinearly processed to obtain the target image features; based on the pooling layer in the preset afterimage recognition model, the target image processing features are downsampled to obtain the first afterimage recognition result. Among them, the specific implementation process of afterimage recognition recorded here can be implemented based on a data calculation component; the data calculation component can include a convolution layer, an activation function (i.e., a nonlinear layer) and a pooling layer.

[0120] Furthermore, in the actual application process, the data calculation subcomponent is the core part of the neural network, which is responsible for performing the calculation of the convolution layer, applying the activation function, and performing the pooling operation; among them, the convolution calculation can be implemented by sliding the window, using the dot product of the weight and the input data; the activation function implements the nonlinear activation function ReLU function; the pooling operation can be the maximum pooling, which is implemented by sliding the window and applying the corresponding aggregation function. Specifically:

[0121] For the pooling layer, first define the parameter data, including the convolution kernel size, input and output image size, step size, padding, etc.; at the same time, use loops to implement point-by-point multiplication and accumulation of weights and input images, and then implement the corresponding logic according to the activation function; finally, implement the corresponding logic according to the pooling method. At the same time, for the convolution calculation logic, it can be implemented by sliding windows; for example, the reset logic can be executed at the rising edge of the clock and the low level of the reset signal to clear the output data; when the start signal is received, the convolution calculation is started. In the specific calculation process, the accumulator is initialized to zero, and each weight of the convolution kernel is traversed through two layers of nested loops, the input data is multiplied by the corresponding weight, and then the product is accumulated to the result. Since the multiplication result may exceed 8 bits, a 16-bit accumulator is used to store the intermediate result; finally, the stage operation can be used to retain the main part of the multiplication result, and the high 8 bits of the accumulated result are assigned to the output data.

[0122] For the nonlinear layer, a dedicated hardware module can be used to apply the activation function. For the ReLU activation function: ReLU(x)=max(0,x), a separate always block can be defined to implement the function logic. When the reset signal is low, the output of the activation function is cleared to zero. The function is applied at the rising edge of each clock. If the convolution result is negative, the output is 0, otherwise the lower 8 bits of the convolution result are output. In addition, in order to improve computational efficiency, pipeline operations can be used to calculate the results of multiple convolution kernels in parallel, and stage operations can be used to optimize the storage and processing of multiplication results.

[0123] For the pooling layer, the max pooling operation selects the maximum value from the pooling window as the output, defines the size of the pooling window and the max pooling output length, initializes the maximum value to negative infinity, uses an 8-bit wide register to represent it, applies the ReLU function and finds the maximum value within the pooling window, and then assigns the maximum value to the output. Here, the ReLU and max pooling operations do not require additional multiplier or adder resources, only registers are needed to store intermediate data.

[0124] Secondly, after obtaining the first afterimage recognition result, the first afterimage compensation parameter of the original screen image can be determined according to the first afterimage recognition result; specifically, it can be achieved in the following manner: generating a compensation parameter query condition according to the first afterimage recognition result, and matching the first afterimage compensation parameter corresponding to the original screen image from a preset color comparison table based on the compensation parameter query condition. The specific determination process of the afterimage compensation parameter can be performed in the parameter recommendation component. Furthermore, in the process of determining the first afterimage compensation parameter, it is first necessary to maintain a color comparison table LUT, which includes the afterimage recognition results and the afterimage compensation parameters corresponding to the afterimage recognition results; that is, the afterimage recognition results and the afterimage compensation parameters can exist in the color comparison table in a one-to-one mapping manner; secondly, when querying the first afterimage compensation parameter, it can be determined directly based on the mapping relationship; and, in the query process, it is also necessary to define the output data width of 16 bits and the recommended compensation parameter width of 8 bits, and the address width of the LUT is 8 bits, so the size of the lookup table is 256; further, initialize and store the LUT, use `$readmemb` to read the initial value of the LUT from the file, search the LUT according to the input address and output the parameters; finally, in order to speed up the recommendation process, the LUT can be pre-stored and initialized to quickly find and obtain parameter values ​​during the recommendation process.

[0125] In step S130, the original screen image is compensated for afterimages in real time based on the afterimage compensation parameters to obtain a target screen image; wherein the first afterimage compensation parameters recorded here include brightness compensation coefficients, R channel compensation coefficients, G channel compensation coefficients and B channel compensation coefficients, and may also include image filtering coefficients, etc.

[0126] Under the premise of the afterimage compensation parameters recorded above, the specific implementation process of real-time afterimage compensation for the original screen image can be implemented in the following ways: one implementation method is: based on the brightness compensation coefficient, the original image brightness of the original screen image is compensated to obtain the target image brightness, and based on the R channel compensation coefficient, the original R pixel channel value of the original screen image is compensated to obtain the target R pixel channel value; based on the G channel compensation coefficient, the original G pixel channel value of the original screen image is compensated to obtain the target G pixel channel value, and based on the B channel compensation coefficient, the original B pixel channel value of the original screen image is compensated to obtain the target B pixel channel value; based on the target image brightness, target R pixel channel value, target G pixel channel value and target B pixel channel value, the original image brightness, original R pixel channel value, original G pixel channel value and original B pixel channel value are replaced to obtain the target screen image. Another implementation method is: based on the image filter coefficient, the original screen image is filtered to obtain the target screen image. Among them, the real-time afterimage compensation process recorded above can be implemented in the real-time compensation component.

[0127] In an exemplary embodiment, in the process of compensating the image brightness, it can be achieved in the following manner: first, configure the brightness compensation coefficient testlv, the R channel compensation coefficient, the G channel compensation coefficient and the B channel compensation coefficient; wherein, the brightness compensation coefficient recorded here can adjust the brightness of the original screen image as a whole, which is usually applied to each pixel value of each channel; the R channel compensation coefficient testR recorded here can adjust the red channel of the image, thereby correcting the color deviation or enhancing the red component; the G channel compensation coefficient testG recorded here can adjust the green channel of the image; the B channel compensation coefficient testB recorded here can adjust the blue channel of the image. Further, for the RGB value of each pixel (original R pixel channel value, original G pixel channel value, original B pixel channel value), the compensated RGB value (target R pixel channel value Rcomp, target G pixel channel value Gcomp, target B pixel channel value Bcomp) can be calculated by the following formula:

[0128] Rcomp=min(max(Rorig·testR·testlv,0),255)

[0129] Gcomp=min(max(Gorig·testG·testlv,0),255)

[0130] Bcomp=min(max(Borig·testB·testlv,0),255)

[0131] Among them, the min() function and max() function involved in the above formula can be used to ensure that the compensated RGB value does not exceed the legal pixel value range, that is, 0 to 255; Rcomp, Gcomp and Bcomp are the compensated RGB values, which represent the pixel values ​​of the red, green and blue channels after compensation. Rorig, Gorig and Borig are the original R pixel channel value, the original G pixel channel value and the original B pixel channel value, which represent the pixel values ​​of the red, green and blue channels without any processing; testR, testG and testB are the R channel compensation coefficient, the G channel compensation coefficient and the B channel compensation coefficient, which control the brightness and contrast of the red, green and blue channels respectively; testlv: This is the maximum brightness adjustment coefficient used to adjust the overall image brightness; The purpose of this formula is to ensure that the compensated RGB value is within the legal pixel value range and takes into account the need for dynamic adjustment of the brightness and color of the image.

[0132] Furthermore, since the product may exceed the legal pixel value range of 0-255 during the calculation process, this may cause the compensated RGB value to be out of range; therefore, by using the min function and the max function, it can be ensured that the final compensated RGB value is limited to the legal pixel value range, that is, between 0-255; by using these compensation coefficients in the calculation, the brightness and color of each pixel can be dynamically adjusted to achieve image optimization and correction; further, by limiting the compensated RGB value to the legal pixel value range, it helps to ensure the stability and rationality of image processing; if it exceeds the legal range, it may cause image distortion or unnatural color. Therefore, this writing method aims to ensure that the compensated image has both good visual effects and maintains the stability and rationality of the image.

[0133] In step S140, the target screen image is transmitted to the image display terminal, so that the image display terminal replaces the original screen image based on the target screen image.

[0134] Specifically, after obtaining the target screen image, the target screen image can be transmitted to the image display terminal based on the HDMI interface; after the image display terminal receives the target screen image, the target screen image can be used to replace the currently displayed original screen image. Among them, transmitting the target screen image to the image display terminal can be achieved based on the output control component; that is, the processed image data (target screen image) can be applied to the MLED splicing screen based on the output control component.

[0135] Furthermore, after the image display terminal displays the target screen image, the method for displaying the screen image also includes: acquiring the target screen image currently displayed by the image display terminal, and performing afterimage recognition on the target screen image based on a preset afterimage recognition model to obtain a second afterimage recognition result; judging whether the target screen image needs to be compensated again according to the second afterimage recognition result; if image compensation is required again, determining the second afterimage compensation parameter of the target screen image based on the second afterimage recognition result, and compensating the target screen image based on the second afterimage compensation parameter. That is, in order to improve the accuracy of the screen image displayed by the image display terminal, secondary afterimage compensation or multiple afterimage compensation can also be performed; the specific implementation process is similar to the first afterimage compensation process described above, and will not be further described here. Among them, the specific implementation process of the secondary afterimage compensation (or the second afterimage compensation parameter) can be implemented in the parameter adjustment component and the feedback iteration component; for example, the parameter adjustment component can adjust the compensation parameters according to the real-time feedback image data taken by the camera, and use BRAM to store the adjustment parameters, update the parameters according to external input and internal calculation results, and provide an interface for the feedback iteration submodule to access and obtain the current parameters; at the same time, the parameter update logic includes initializing all parameters, updating the parameter storage, and updating the parameters according to specific conditions, and setting the new parameters as the current parameters; further, in the feedback iteration component, system feedback is collected, and iterative adjustments are made based on the feedback, the feedback signal collection and processing logic is implemented, the parameters are adjusted according to the results of the feedback signal, and sent to the parameter adjustment component; the specific implementation logic is feedback evaluation and iteration, the performance is evaluated according to the feedback signal, an adjustment signal is sent, and the parameter adjustment submodule is notified to make adjustments.

[0136] At this point, the screen image display method recorded in the exemplary embodiment of the present disclosure has been fully realized. Based on the above-mentioned contents, it can be known that the screen image display method recorded in the exemplary embodiment of the present disclosure has at least the following advantages: (1) High-speed parallel processing: By utilizing the powerful parallel processing capability of FPGA, real-time detection and compensation of MLED_COG_LTPS screen afterimages can be realized; at the same time, through parallel processing, the system can process multiple image data at the same time, realize efficient afterimage detection and compensation, thereby improving the response speed and processing efficiency of the system; (2) Automatic parameter adjustment: With the help of machine learning algorithms, automatic adjustment of afterimage compensation parameters is realized, thereby reducing manual intervention and debugging time. The system can automatically optimize and adjust compensation parameters based on real-time collected image data and preset optimization targets, thus improving the intelligence and automation level of the system; (3) High-precision compensation: Through precise image processing and fine adjustment of compensation parameters, high-precision compensation of residual images is achieved, thereby improving display quality; and the system can accurately identify and compensate for different types of residual images, and achieve optimization effects while ensuring image clarity; (4) System stability: A complete MLED residual image compensation automatic parameter adjustment process is implemented on the FPGA chip, improving the stability and efficiency of the system; and by implementing the residual image compensation algorithm and parameter adjustment function at the hardware level, the system can run stably and ensure long-term stability and reliability; (5) Give full play to the advantages of FPGA: Utilize the flexibility and high Performance, achieving fast and accurate parameter adjustment, optimizing screen display effects, and reducing afterimage problems; further, as a programmable hardware platform, FPGA can implement customized afterimage compensation algorithms and parameter adjustment logic according to specific needs, giving full play to its advantages and potential; (6) Real-time feedback: Real-time adjustment of parameters according to the compensated image effects improves the stability and adaptability of the system; moreover, the system can respond to image changes and environmental changes in a timely manner, dynamically adjust parameters, and ensure the stability and performance of the system in different scenarios; (7) Intelligent optimization: The system can combine machine learning algorithms to achieve intelligent optimization and evolution of parameters, further improving system performance; through continuous learning and optimization, the system can continuously improve its own compensation effects and performance, adapt to the ever-changing environment and needs, and has a high level of intelligence.

[0137] The following are embodiments of the device disclosed herein, which can be used to execute the method embodiments disclosed herein. For details not disclosed in the device embodiments disclosed herein, please refer to the method embodiments disclosed herein.

[0138] The exemplary embodiment of the present disclosure also provides a display device for a screen image. Specifically, refer to Figure 7As shown, the screen image display device may include an image preprocessing module 710 , a first compensation parameter determination module 720 , an afterimage real-time compensation module 730 and a screen image display module 740 .

[0139] in:

[0140] The image preprocessing module 710 may be used to obtain the original screen image currently displayed by the image display terminal, and perform image preprocessing on the original screen image to obtain a standard screen image;

[0141] The first compensation parameter determination module 720 may be configured to perform afterimage recognition on the standard screen image based on a preset afterimage recognition model to obtain a first afterimage recognition result, and determine a first afterimage compensation parameter of the original screen image according to the first afterimage recognition result;

[0142] The afterimage real-time compensation module 730 may be used to perform afterimage real-time compensation on the original screen image based on the afterimage compensation parameters to obtain a target screen image;

[0143] The screen image display module 740 may be configured to transmit the target screen image to the image display terminal, so that the image display terminal replaces the original screen image with the target screen image.

[0144] In an exemplary embodiment of the present disclosure, the original screen image is subjected to image preprocessing to obtain a standard screen image, including: processing the original image brightness and the original image contrast of the original screen image to obtain a standard image brightness and a standard image contrast, and processing the original color space of the original screen image to obtain a standard color space; based on the standard image brightness, the standard image contrast and the standard color space, replacing the original image brightness, the original image contrast and the original color space to obtain a standard screen image.

[0145] In an exemplary embodiment of the present disclosure, the original image brightness and the original image contrast of the original screen image are processed to obtain a standard image brightness and a standard image contrast, including: obtaining a preset brightness adjustment parameter and a preset contrast adjustment parameter; adjusting the original image brightness of the original screen image according to the preset brightness adjustment parameter to obtain a standard image brightness; adjusting the original image contrast according to the preset contrast adjustment parameter to obtain a standard image contrast.

[0146] In an exemplary embodiment of the present disclosure, the original color space of the original screen image is processed to obtain a standard color space, including: obtaining a preset color value conversion matrix; wherein the preset color value conversion matrix includes a first color value conversion matrix, a second color value conversion matrix, and a third color value conversion matrix; based on the first color value conversion matrix and the original color values ​​in the original color space, calculating the first standard color value of the X dimension in the standard color space; based on the second color value conversion matrix and the original color values ​​in the original color space, calculating the second standard color value of the Y dimension in the standard color space; based on the third color value conversion matrix and the original color values ​​in the original color space, calculating the third standard color value of the Z dimension in the standard color space; and obtaining the standard color space based on the first standard color value, the second standard color value, and the third standard color value.

[0147] In an exemplary embodiment of the present disclosure, afterimage recognition is performed on the standard screen image based on a preset afterimage recognition model to obtain a first afterimage recognition result, including: performing convolution processing on the standard screen image based on a convolution layer in the preset afterimage recognition model to obtain local image features of the standard screen image; performing nonlinear processing on the local image features based on a nonlinear layer in the preset afterimage recognition model to obtain target image features; and performing downsampling processing on the target image processing features based on a pooling layer in the preset afterimage recognition model to obtain the first afterimage recognition result.

[0148] In an exemplary embodiment of the present disclosure, the preset afterimage recognition model is obtained in the following manner: preset reference weight parameters are read from a weight storage component, and the convolutional neural network parameters are updated based on the preset reference weight parameters to obtain the preset afterimage recognition model.

[0149] In an exemplary embodiment of the present disclosure, the preset reference weight parameters are obtained in the following manner: obtaining a historical screen image and an image afterimage label corresponding to the historical screen image, and inputting the historical screen image into a convolutional neural network to obtain an afterimage prediction result; constructing a loss function according to the afterimage prediction result and the image afterimage label, and training the convolutional neural network based on the loss function; extracting parameters in the trained convolutional neural network to obtain the preset reference weight parameters, and storing the preset reference weight parameters in the weight storage component.

[0150] In an exemplary embodiment of the present disclosure, a first afterimage compensation parameter of the original screen image is determined according to the first afterimage recognition result, including: generating a compensation parameter query condition according to the first afterimage recognition result, and matching the first afterimage compensation parameter corresponding to the original screen image from a preset color comparison table based on the compensation parameter query condition.

[0151] In an exemplary embodiment of the present disclosure, the first afterimage compensation parameter includes at least one of a brightness compensation coefficient, an R channel compensation coefficient, a G channel compensation coefficient and a B channel compensation coefficient; wherein, based on the afterimage compensation parameters, the original screen image is compensated for afterimages in real time to obtain a target screen image, including: compensating the original image brightness of the original screen image based on the brightness compensation coefficient to obtain a target image brightness, and compensating the original R pixel channel value of the original screen image based on the R channel compensation coefficient to obtain a target R pixel channel value; compensating the original G pixel channel value of the original screen image based on the G channel compensation coefficient to obtain a target G pixel channel value, and compensating the original B pixel channel value of the original screen image based on the B channel compensation coefficient to obtain a target B pixel channel value; replacing the original image brightness, the original R pixel channel value, the original G pixel channel value and the original B pixel channel value based on the target image brightness, the target R pixel channel value, the target G pixel channel value and the target B pixel channel value to obtain a target screen image.

[0152] In an exemplary embodiment of the present disclosure, the first afterimage compensation parameter also includes an image filtering coefficient; wherein, performing real-time afterimage compensation on the original screen image based on the afterimage compensation parameter to obtain a target screen image further includes: filtering the original screen image based on the image filtering coefficient to obtain a target screen image.

[0153] In an exemplary embodiment of the present disclosure, the display device of the screen image further includes:

[0154] The second afterimage recognition result determination module may be used to obtain a target screen image currently displayed by the image display terminal, and perform afterimage recognition on the target screen image based on a preset afterimage recognition model to obtain a second afterimage recognition result;

[0155] An image determination module, which can be used to determine whether the target screen image needs to be compensated again according to the second residual image recognition result;

[0156] The target screen image compensation module can be used to determine a second afterimage compensation parameter of the target screen image based on the second afterimage recognition result if further image compensation is required, and compensate the target screen image based on the second afterimage compensation parameter.

[0157] The specific details of each module in the above-mentioned screen image display device have been described in detail in the corresponding screen image display method, so they will not be repeated here.

[0158] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0159] In addition, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0160] In an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided.

[0161] Those skilled in the art will appreciate that various aspects of the present disclosure may be implemented as systems, methods or program products. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: complete hardware implementation, complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to herein as "circuits", "modules" or "systems".

[0162] Refer to the following Figure 8 The electronic device 800 according to this embodiment of the present disclosure is described. Figure 8 The electronic device 800 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0163] like Figure 8 As shown, the electronic device 800 is in the form of a general computing device. The components of the electronic device 800 may include, but are not limited to: the at least one processing unit 810, the at least one storage unit 820, a bus 830 connecting different system components (including the storage unit 820 and the processing unit 810), and a display unit 840.

[0164] The storage unit stores program codes, which can be executed by the processing unit 810, so that the processing unit 810 performs the steps according to various exemplary embodiments of the present disclosure described in the above “Exemplary Method” section of this specification. For example, the processing unit 810 can perform the following steps: Figure 1 The step S110 shown in the figure is as follows: acquiring the original screen image currently displayed by the image display terminal, and performing image preprocessing on the original screen image to obtain a standard screen image; step S120: performing afterimage recognition on the standard screen image based on a preset afterimage recognition model to obtain a first afterimage recognition result, and determining a first afterimage compensation parameter of the original screen image according to the first afterimage recognition result; step S130: performing real-time afterimage compensation on the original screen image based on the afterimage compensation parameter to obtain a target screen image; step S140: transmitting the target screen image to the image display terminal, so that the image display terminal replaces the original screen image based on the target screen image.

[0165] The storage unit 820 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 8201 and / or a cache memory unit 8202 , and may further include a read-only memory unit (ROM) 8203 .

[0166] The storage unit 820 may also include a program / utility 8204 having a set (at least one) of program modules 8205, such program modules 8205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0167] Bus 830 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0168] The electronic device 800 may also communicate with one or more external devices 900 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 800, and / or communicate with any device that enables the electronic device 800 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 850. Furthermore, the electronic device 800 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 860. As shown, the network adapter 860 communicates with other modules of the electronic device 800 via a bus 830. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 800, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0169] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0170] In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the above method of the present specification is stored. In some possible implementations, various aspects of the present disclosure may also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary implementations of the present disclosure described in the above "Exemplary Method" section of the present specification.

[0171] According to the program product for implementing the above method in the embodiment of the present disclosure, it can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present disclosure is not limited thereto. In this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.

[0172] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0173] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0174] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0175] Program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0176] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0177] Other embodiments of the present disclosure will be readily apparent to those skilled in the art after considering the specification and practicing the inventions invented herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not invented by the present disclosure. The specification and examples are to be considered merely exemplary, and the true scope and spirit of the present disclosure are indicated by the claims.

Claims

1. A method for displaying a screen image, characterized in that: include: Acquire the original screen image currently displayed by the image display terminal, and perform image preprocessing on the original screen image to obtain a standard screen image; Performing afterimage recognition on the standard screen image based on a preset afterimage recognition model to obtain a first afterimage recognition result, and determining a first afterimage compensation parameter of the original screen image according to the first afterimage recognition result; Performing real-time afterimage compensation on the original screen image based on the afterimage compensation parameters to obtain a target screen image; The target screen image is transmitted to the image display terminal, so that the image display terminal replaces the original screen image based on the target screen image.

2. The method for displaying a screen image according to claim 1, characterized in that: Performing image preprocessing on the original screen image to obtain a standard screen image includes: Processing the original image brightness and the original image contrast of the original screen image to obtain a standard image brightness and a standard image contrast, and processing the original color space of the original screen image to obtain a standard color space; Based on the standard image brightness, standard image contrast and standard color space, the original image brightness, original image contrast and original color space are replaced to obtain a standard screen image.

3. The method for displaying a screen image according to claim 2, characterized in that: Processing the original image brightness and the original image contrast of the original screen image to obtain a standard image brightness and a standard image contrast includes: Obtaining preset brightness adjustment parameters and preset contrast adjustment parameters; Adjusting the original image brightness of the original screen image according to the preset brightness adjustment parameter to obtain a standard image brightness; The contrast of the original image is adjusted according to the preset contrast adjustment parameter to obtain a standard image contrast.

4. The method for displaying a screen image according to claim 2, characterized in that: The original color space of the original screen image is processed to obtain a standard color space, including: Obtaining a preset color value conversion matrix; wherein the preset color value conversion matrix includes a first color value conversion matrix, a second color value conversion matrix, and a third color value conversion matrix; Calculating a first standard color value of an X dimension in the standard color space based on the first color value conversion matrix and the original color value in the original color space; Calculating a second standard color value of a Y dimension in the standard color space based on the second color value conversion matrix and the original color value in the original color space; Calculating a third standard color value of a Z dimension in the standard color space based on a third color value conversion matrix and the original color value in the original color space; The standard color space is obtained based on the first standard color value, the second standard color value and the third standard color value.

5. The method for displaying a screen image according to claim 1, characterized in that: Performing afterimage recognition on the standard screen image based on a preset afterimage recognition model to obtain a first afterimage recognition result includes: Performing convolution processing on the standard screen image based on the convolution layer in the preset afterimage recognition model to obtain local image features of the standard screen image; Performing nonlinear processing on the local image features based on the nonlinear layer in the preset residual image recognition model to obtain target image features; Based on the pooling layer in the preset residual image recognition model, the target image processing features are down-sampled to obtain the first residual image recognition result.

6. The method for displaying a screen image according to claim 5, characterized in that: The preset residual image recognition model is obtained in the following manner: The preset reference weight parameters are read from the weight storage component, and the parameters of the convolutional neural network are updated based on the preset reference weight parameters to obtain the preset afterimage recognition model.

7. The method for displaying a screen image according to claim 6, characterized in that: The preset reference weight parameter is obtained in the following manner: Acquire a historical screen image and an image afterimage label corresponding to the historical screen image, and input the historical screen image into a convolutional neural network to obtain an afterimage prediction result; Constructing a loss function according to the residual image prediction result and the image residual image label, and training the convolutional neural network based on the loss function; Extract the parameters in the trained convolutional neural network to obtain the preset reference weight parameters, and store the preset reference weight parameters in the weight storage component.

8. The method for displaying a screen image according to claim 1, characterized in that: Determining a first afterimage compensation parameter of the original screen image according to the first afterimage recognition result includes: A compensation parameter query condition is generated according to the first afterimage recognition result, and based on the compensation parameter query condition, a first afterimage compensation parameter corresponding to the original screen image is matched from a preset color comparison table.

9. The method for displaying a screen image according to claim 1, characterized in that: The first afterimage compensation parameter includes at least one of a brightness compensation coefficient, an R channel compensation coefficient, a G channel compensation coefficient, and a B channel compensation coefficient; The method of performing real-time afterimage compensation on the original screen image based on the afterimage compensation parameter to obtain a target screen image includes: Compensating the original image brightness of the original screen image based on the brightness compensation coefficient to obtain a target image brightness, and compensating the original R pixel channel value of the original screen image based on the R channel compensation coefficient to obtain a target R pixel channel value; Compensating the original G pixel channel value of the original screen image based on the G channel compensation coefficient to obtain a target G pixel channel value, and compensating the original B pixel channel value of the original screen image based on the B channel compensation coefficient to obtain a target B pixel channel value; Based on the target image brightness, target R pixel channel value, target G pixel channel value and target B pixel channel value, the original image brightness, original R pixel channel value, original G pixel channel value and original B pixel channel value are replaced to obtain a target screen image.

10. The method for displaying a screen image according to claim 9, characterized in that: The first residual image compensation parameter also includes an image filtering coefficient; The method further comprises: performing real-time afterimage compensation on the original screen image based on the afterimage compensation parameter to obtain a target screen image; The original screen image is filtered based on the image filtering coefficient to obtain a target screen image.

11. The method for displaying a screen image according to claim 1, characterized in that: The screen image display method further includes: Acquire a target screen image currently displayed by the image display terminal, and perform afterimage recognition on the target screen image based on a preset afterimage recognition model to obtain a second afterimage recognition result; Determining whether the target screen image needs to be compensated again according to the second residual image recognition result; If image compensation is required again, a second afterimage compensation parameter of the target screen image is determined based on the second afterimage recognition result, and the target screen image is compensated based on the second afterimage compensation parameter.

12. A screen image display device, characterized in that: include: An image preprocessing module, used to obtain an original screen image currently displayed by the image display terminal, and perform image preprocessing on the original screen image to obtain a standard screen image; A first compensation parameter determination module, configured to perform afterimage recognition on the standard screen image based on a preset afterimage recognition model to obtain a first afterimage recognition result, and determine a first afterimage compensation parameter of the original screen image according to the first afterimage recognition result; An afterimage real-time compensation module, used for performing afterimage real-time compensation on the original screen image based on the afterimage compensation parameters to obtain a target screen image; The screen image display module is used to transmit the target screen image to the image display terminal, so that the image display terminal replaces the original screen image based on the target screen image.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the screen image display method described in any one of claims 1-11 is implemented.

14. An electronic device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; Wherein, the processor is configured to execute the screen image display method described in any one of claims 1-11 by executing the executable instructions.

Citation Information

Cited By

  • Picture quality compensation method for display apparatuses, and picture quality compensation model, electronic device and storage medium

    WO2026117943A1

  • Compensation method and system, and electronic device and storage medium

    WO2026143692A1